Structural-conformational aspects of tRNA complexation with chloroethyl nitrosourea derivatives: A molecular modeling and spectroscopic investigation

Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

Contents lists available at ScienceDirect

Journal of Photochemistry & Photobiology, B: Biology journal homepage: www.elsevier.com/locate/jphotobiol

Structural-conformational aspects of tRNA complexation with chloroethyl nitrosourea derivatives: A molecular modeling and spectroscopic investigation Shweta Agarwal a,b, Gunjan Tyagi b, Deepti Chadha a,b, Ranjana Mehrotra a,b,⁎ a b

Academy of Scientific & Innovative Research (AcSIR), CSIR-National Physical Laboratory Campus, New Delhi 110012, India Quantum Phenomena and Applications, CSIR-National Physical Laboratory, Dr. K. S. Krishnan Marg, New Delhi 110012, India

a r t i c l e

i n f o

Article history: Received 4 October 2015 Received in revised form 18 September 2016 Accepted 20 September 2016 Available online 06 November 2016 Keywords: CENUs Molecular modeling FTIR spectroscopy CD spectroscopy Drug-RNA interaction

a b s t r a c t Chloroethyl nitrosourea derivatives (CENUs) represent an important family of anticancer chemotherapeutic agents, which are used in the treatment of different types of cancer such as brain tumors, resistant or relapsed Hodgkin's disease, small cell lung cancer and malignant melanoma. This work focuses towards understanding the interaction of chloroethyl nitrosourea derivatives; lomustine, nimustine and semustine with tRNA using spectroscopic approach in order to elucidate their auxiliary anticancer action mechanism inside the cell. Attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR), Fourier transform infrared difference spectroscopy, circular dichroism spectroscopy and UV–visible spectroscopy were employed to investigate the binding parameters of tRNA-CENUs complexation. Results of present study demonstrate that all CENUs, studied here, interact with tRNA through guanine nitrogenous base residues and possibly further crosslink cytosine residues in paired region of tRNA. Moreover, spectral data collected for nimustine-tRNA and semustine-tRNA complex formation indicates towards the groove-directed-alkylation as their anti-malignant action, which involves the participation of uracil moiety located in major groove of tRNA. Besides this, tRNA-CENUs adduct formation did not alter the native conformation of biopolymer and tRNA remains in A-form after its interaction with all three nitrosourea derivatives studied. The binding constants (Ka) estimated for tRNA complexation with lomustine, nimustine and semustine are 2.55 × 102 M−1, 4.923 × 102 M−1 and 4.223 × 102 M−1 respectively, which specify weak type of CENU's binding with tRNA. Moreover, molecular modeling simulations were also performed to predict preferential binding orientation of CENUs with tRNA that corroborates well with spectral outcomes. The findings, presented here, recognize tRNA binding properties of CENUs that can further help in rational designing of more specific and efficient RNA targeted chemotherapeutic agents. © 2016 Published by Elsevier B.V.

1. Introduction The technical advancements in characterization and determination of RNA structure has permitted the overcome of several inherent difficulties regarding RNA structure analysis, preparation and handling of RNA solution. In addition, this has led to the development of knowledge about the relationship between RNA structure and its physiological function [1–7]. As a result, it is now fully considered that RNA intercedes at all the stages of cellular life. Moreover, the significance of RNA in cellular events is not only due to its key sequence motifs, but also devoted to its intricate three-dimensional folds. RNA acts as a key player in many cellular processes such as gene regulation, protein synthesis, mRNA splicing, retroviral replication and transcriptional regulation [8–10]. This apprehension has promoted RNA to be a logical target as a ⁎ Corresponding author at: Quantum Phenomena and Applications, CSIR-National Physical Laboratory, Dr. K. S. Krishnan Marg, New Delhi 110012, India. E-mail address: [email protected] (R. Mehrotra).

http://dx.doi.org/10.1016/j.jphotobiol.2016.09.045 1011-1344/© 2016 Published by Elsevier B.V.

therapeutic intervention for chemotherapeutic agents, rationally designed drugs as well as for natural compounds. Indeed, the binding of ligand with RNA may influence its normal cellular and auxiliary processes like inhibiting the RNA catalysis, preventing the formation of its complex with biologically relevant molecule (protein, DNA or RNA itself) or by inducing the distortion in its native conformation [11–13]. Furthermore, the three-dimensional configuration of RNA molecule gives rise to a complex structure, which augments the possibility of its highly specific interaction with any ligand or drug. Besides this, RNA manifests a greater structural-conformational diversity, a fact that also develops the impact of therapeutics, directed towards RNA [2,6,14]. Keeping this in view, in the present study, we aim to study the interaction of three nitrosourea derivatives lomustine, nimustine and semustine with tRNA to explore their possible interaction with RNA as a part of their anti-proliferative action. Nitrosourea derivatives, known as alkylating agents, are extremely active class of antitumor agents. They are effective against solid tumors, as well as leukemia [15]. In particular, 2-chloroethyl nitrosourea (CENUs) derivatives (lomustine,

2

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

nimustine and semustine) and some of their metabolites show great promises as effective anti-tumor agents [16]. Lomustine (CCNU) [1-(2chloroethyl)-3-cyclohexyl-1-nitrosourea] (Fig. 1A) inhibits cell cycle progression at S- and G2-M phase. It is widely used in the treatment of brain tumors, resistant or relapsed Hodgkin's disease, small cell lung cancer, lymphomas, malignant melanoma and various solid tumors. Lomustine is used alone or in combination with radiotherapy and surgery for the treatment of brain tumors and brain metastasis. Solubility of lomustine in alcohol specifies its affinity towards lipids at physiological level; hence, it can cross the blood-brain-barrier and effectively treat brain tumors [17–20]. Another drug of the family is nimustine (ACNU) [(1-(4-amino-2-methyl-5-pyrimidynyl) methyl-3-(2chloroethyl)-3-nitrosourea hydrochloride)] (Fig. 1B), which is a cellcycle phase nonspecific antineoplastic agent and mainly used for the treatment of malignant gliomas (brain/spine tumor). It provides a major therapeutic option for the high-grade gliomas. Nimustine is soluble in both water and methanol, representing its affinity to lipid bilayer membrane and enables it to cross the blood-brain-barrier for the chemotherapy of gliomas [21–23]. The third member of the family semustine (Fig. 1C) is a 4-methyl derivative of lomustine (me-CCNU), used to treat primary and metastatic brain tumors, Lewis lung tumor and L1210 leukemia. It has also been used to treat cancer of the digestive tract, Hodgkin lymphoma, malignant melanoma, and epidermoid carcinoma of the lung. Like the other two members of the nitrosourea family, semustine is also able to cross the blood-brain barrier making it effective for brain tumors [19,24–26]. Antineoplastic action of nitrosourea derivatives is believed to involve the inhibition of DNA replication, RNA transcription and protein translation by means of alkylation [27–29]. Moreover, they also affect a number of cellular events including ribosomal and nucleoplasmic messenger RNA processing, DNA base structure and DNA polymerase activity. It is known that initial alkylation at O-6 position of a guanine moiety in one strand of DNA and subsequent reaction with the N-3 position of cytosine in the complementary strand to produce interstrand cross-linking is the mechanism of action adopted by nitrosourea drugs [27–33]. However, the spectrum of their diverse action in cell raises the possibility of involvement of some other sites in its course of anticancer action [15]. Working on these lines, we have investigated the nature and mode of interaction of above-mentioned three nitrosourea derivatives with tRNA using various biophysical and spectroscopic techniques (Attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR), Fourier transform infrared difference spectroscopy, circular dichroism (CD) spectroscopy and UV–visible spectroscopy). In addition, molecular docking simulations were performed to predict the preferred orientation of lomustine, nimustine and semustine binding to tRNA molecule, which further help in predicting the strength and site of interaction between drug and tRNA moiety. Parameters involved in the binding of CENUs with tRNA are studied in detail. Binding affinities of these CENUs with tRNA are also calculated.

2. Methodology 2.1. Materials RNA (tRNA) from Baker's yeast was purchased from Sigma-Aldrich chemicals (USA). Lomustine, nimustine and semustine were procured from Sigma-Aldrich chemicals (USA). Purity of RNA was estimated by recording UV absorbance at 260 nm and 280 nm followed by the calculation of ratio, A260/A280. The ratio was found to be ~1.89, indicating that RNA is free from any protein contamination [34]. Deionized ultra pure water (resistance 18.2 MΏ) from Scholar-UV Nex UP 1000 system was used for the preparation of desired aqueous solutions (buffer and drug solutions). Other chemicals and reagents used in the study were of analytical grade and used as supplied. 2.2. Preparation of Stock Solutions The solution of tRNA sodium salt was prepared in 10 mM tris-HCl buffer (pH 7.4) and kept at 8 °C for 24 h. The solution was stirred at frequent intervals to ensure its homogeneity. Using extinction coefficient of 9250 cm−1 M−1, concentration of tRNA stock solution was calculated spectrophotometrically [1]. The final concentration of RNA stock solution was found to be 25 mM (due to phosphate group molarity). 2.3. In-Silico Study The computational molecular docking studies of CENUs-tRNA interactions were performed using AutoDock 4.2 [35]. The three dimensional structures of lomustine, nimustine and semustine with Zinc IDs 3831006, 3979156 and 387495 respectively were obtained from Zinc Database [36] in mol2 format. These files were converted to pdb format using Open Babel 3.2.9 [37]. The 3D structure of tRNA (pdb ID: 6TNA) was obtained from the Protein Data Bank [38]. The receptor and ligand coordinate files were prepared in PDBQT format using AutoDockTools (ADT) version 1.5.6 [39]. AutoGrid program was employed to pre-calculate grid maps of interaction energies for various atom types of the drug with macromolecule (tRNA) using grid box of dimensions 45 × 62 × 62 with grid center placed at 44.5, 17.546, 50.0 and spacing set to 0.375 Å. For molecular modeling simulations, tRNA was used as a rigid receptor, whereas the drug molecules being docked were kept flexible. The docking was carried out using random starting positions and orientations of the ligands. The Lamarckian genetic algorithm (LGA) and the pseudo-Solis and Wets methods were applied for conformational searching using default parameters. The docking protocol consisted of 100 trials using 150 individuals in the population, mutation rate of 0.02, crossover rate of 0.80 with maximum number of energy evaluations and generations set to 2.5 million and 27,000 respectively. The analysis of the docked conformations was performed using ranked cluster analysis. The clusters were formed by grouping similar docked

Fig. 1. Chemical structure of (A) Lomustine (B) Nimustine and (C) Semustine. Red dotted circle contrasts the presence of methyl group in semustine as compared to lomustine. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

conformations in increasing order of their binding energy at 2.0 Å RMSD value using ADT. The lowest energy docked structure of the most populated cluster (best-docked conformer) was used as a model to represent the CENUs-tRNA complex that was utilized for subsequent analysis. The molecular docking simulations were performed independently for lomustine, nimustine and semustine using identical simulation conditions and methods. 2.4. FTIR Spectral Measurements

3

of 1 cm is; A ¼ εR lRt

ð1Þ

where, εR is the molar absorptivity of free tRNA. The absorbance of solution (AS) comprising of total concentration of tRNA (Rt) along with total concentration of drug (Dt) is; AS→εRl½R þ εDl½D þ εRDl½RD

ð2Þ

where: [R] is the concentration of uncomplexed tRNA. [D] is the concentration of uncomplexed drug. [RD] is the concentration of drug-tRNA complex. εD is the molar absorptivity of drug. εRD is the molar absorptivity of drug-tRNA complex. After combining with the mass balance of tRNA and drug, the absorbance equation can be written as;

FTIR spectra were recorded with Varian 660-IR spectrophotometer, equipped with DTGS (deuterated triglycine sulphate) detector and KBr beam splitter. All spectra were recorded in 10 mM tris-HCl buffer (pH of 7.4). To remove water vapors from sample chamber, dry nitrogen gas was purged continuously. Liquid samples were analyzed in attenuated total reflectance mode with Miracle® (PIKE) ZnSe-micro horizontal attenuated total internal reflection (HATR) assembly. Ambient humidity of 45% RH was maintained during the entire experiment. 256 interferograms were recorded for each sample in the spectral range of 2400–600 cm−1 with a resolution of 2 cm−1. Background spectra were collected before each spectral measurement. A spectrum of buffer solution was recorded and subtracted from the spectrum of tRNA and CENUs-tRNA complexes. A satisfactory buffer subtraction was considered to be achieved, when the intensity of water combination band at 2200 cm− 1 became zero in all the IR spectra recorded [40]. FTIR difference spectra were produced by subtracting the spectrum of free tRNA from the spectrum of ligand-tRNA complex [(RNA solution + CENUs solution)–(RNA solution)]. For the spectral measurements of CENUs-RNA complexes, drug solutions of three different concentrations (0.3125 mM, 0.625 mM and 1.25 mM) were prepared. Thereafter, these drug solutions were added in a dropwise manner to tRNA solution (25 mM) to achieve 1/80, 1/40 and 1/20 M ratios (r) of CENUs-tRNA complexes. Continuous vortexing for 15 min followed by incubation at room temperature for 2 h was performed to ensure the complete complexation of CENUs with tRNA.

The absorbance of solution (A) measured against the total concentration of drug as reference is

2.5. Circular Dichroism (CD) Spectral Measurements

From the mass balance equation; Rt = [R] + [RD], we get [R] = Rt/ (1 + KRD[D]), that gives following equation-

CD spectral measurements were recorded on Applied Photophysics (Chirascan) spectrophotometer using quartz cuvette having a pathlength of 1 mm. All the spectra of CENUs-tRNA complexes were collected in the far UV range (200 nm–320 nm) subsequent to 2 h of incubation period at room temperature. Six scans were recorded with a scanning speed of 1 nm/s followed by averaging for each sample. To accomplish buffer subtraction, a spectrum of buffer was subtracted from the spectra of free tRNA and CENUs-tRNA complexes. For CD spectral measurements, CENUs solution of varying concentration (in the range of 0.0625 mM to 0.25 mM) were prepared and added into tRNA solution of constant concentration (5 mM) and thus, drug/tRNA molar ratios (r) become 1/80, 1/40 and 1/20. 2.6. UV–Visible Spectral Measurements The absorption spectra of free tRNA and its complexes with CENUs (lomustine, nimustine and semustine) were recorded on Perkin-Elmer Lambda-35 spectrophotometer. Quartz cuvette with a pathlength of 1 cm was used for the spectral measurements. For UV–visible studies, 0.04 mM tRNA solution was used with varying concentrations of CENUs ranging from 4 × 10−2 to 4 × 10−3 mM. Binding constants (Ka) for the formation of CENUs-tRNA complexes were calculated assuming that only one type of interaction occurs between tRNA (R) and drug (D) in aqueous solution resulting in the formation of one type of complex (RD) [41]. It is also presumed that tRNA and ligands follow Lambert-Beer's law for the absorbance of light. The absorbance of tRNA solution (Ao) at its total concentration (Rt) with a path length (l)

AS→εRl Rt þ εDl Dt þ ΔεRDl ½RD ΔεRD →ε RD −ε R −ε D

A→εR l Rt þ Δ εRDl ½RD

ð3Þ

ð4Þ

The stability constant (KRD) for the formation of complex (RD) can be given as K RD →½RD=½R½D

ð5Þ

Combining Eq. (4) and (5) ΔA→KRD ΔεRDl½R½D ΔA ¼ A−A0

ΔA Rt KRDΔεRD½D → l 1 þ KRD½D

ð6Þ

ð7Þ

There is a hyperbolic relation between the free drug molecule concentration and its interaction with tRNA. Linear transformation of Eq. (6) is done by taking the reciprocal of both side of Eq. (7) that can be presented as; l 1 1 → þ ΔA RtKRDΔεRD ½D RtΔεRD

ð8Þ

The double reciprocal plot of 1/ΔA versus 1/[D] is linear and the binding constant (Ka) can be calculated by estimating the ratio of intercept to slope. 3. Results and Discussion 3.1. Molecular Modeling Analysis The molecular docking results, presented here (Fig. 2), demonstrate the ability of lomustine, nimustine and semustine to form stable complexes with tRNA molecule. Lomustine manifested the binding energy of − 4.76 kcal/mol with nucleobase residues G51, U52, G53, 5MU54, 1MA58 and A62 of tRNA, found in the close vicinity of the drug (Fig. 2A, left panel). In addition, LMT-tRNA complex is stabilized by two hydrogen bonds, formed between lomustine moiety and nitrogenous base residues U52 (uracil) and G53 (guanine) of tRNA (Fig. 2A, right

4

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

5

Table 1 Hydrogen bond information.

Complex

Donor residue

Acceptor residue

LMT-tRNA

H12 of lomustine (hydrogen atom at 12th position of lomustine) H7 of G53 (seventh hydrogen atom of Guanine located at 53rd position of tRNA) NMT-tRNA H9 of nimustine (hydrogen atom at 9th position of nimustine)

SMT-tRNA

H2′ of C48 (second hydrogen atom of ribose sugar attached with Cytosine located at 48th position of tRNA) H14 of semustine (hydrogen atom at 14th position of semustine) H7 of G53 (seventh hydrogen atom of Guanine located at 53rd position of tRNA)

panel) with distances 1.779 Å and 1.923 Å respectively (details are shown in Table 1). The free binding energy for nimustine-tRNA complexation is predicted as −5.2 kcal/mol. The docking outcomes indicate the interaction of nimustine with C48, 5MC49, U50, 1MA58 and U59 residues of tRNA (Fig. 2B, left panel). In addition, nimustine and tRNA complexation is strengthen by the generation of two hydrogen bonds. The first hydrogen bond is formed between nimustine and phosphate group attached with the uracil nitrogenous base (U50) of tRNA having 1.932 Å distance (Fig. 2B, right panel). The second hydrogen bond of length 2.162 Å is predicted between the ribose sugar residue (attached with cytosine base located at 48th position) of tRNA and the oxygen atom at first position (O1) of nimustine (Table 1). Besides this, the molecular docking outcomes suggest the involvement of U50, G51, U52, G53, 5MU54, 1MA58, C60 and A62 ribonucleotides in semustine-tRNA interactions with free binding energy estimated as − 5.07 kcal/mol (Fig. 2C, left panel). Two hydrogen bonds are also observed with nucleobases uracil (U52) and guanine (G53) (Table 1) during the formation of semustine-tRNA complex. H14 of semustine and O4 of U52 participate as donor and acceptor residues respectively to form a hydrogen bond of length 1.817 Å. The second hydrogen bond is observed between 7th hydrogen atom of G53 (donor residue) and second oxygen atom of semustine (acceptor residue) with 1.864 Å distance (Fig. 2C, right panel). 3.2. FTIR Spectral Analysis The infrared spectrum of tRNA is comparable to double stranded DNA in its A-geometrical form. Major infrared bands assigned to different composition units of RNA are shown in Fig. 3. Infrared vibrations emerging between 1700 and 1500 cm−1 are assigned to in-plane stretching vibrations of RNA nitrogenous bases. The strong infrared band at 1705 cm−1 is assigned to the C_O and C_N stretching vibrations of guanine residues. The band at 1650 cm−1 is primarily attributed to the C_O stretching vibrations of uracil. C_N and C_O stretching vibrations of adenine residues lead to the emergence of band at 1602 cm−1. Infrared region between 1500 and 1250 cm−1 comprises of bands that are arising due to in-plane ring vibrations of bases and A- conformation of RNA. Out-of-phase and in-plane stretching

Distance (Å)

O4 of U52 (fourth oxygen atom of Uracil located at 52nd position of tRNA) O2 of lomustine (oxygen atom at 2nd position of lomustine) OP2 of U50 (second oxygen atom of phosphate group attached with Uracil located at 50th position of tRNA) O1 of nimustine (oxygen atom at 1st position of nimustine)

1.779

O4 of U52 (fourth oxygen atom of Uracil located at 52nd position of tRNA) O2 of semustine (oxygen atom at 2nd position of semustine)

1.817

1.923 1.932

2.162

1.864

vibrations of cytosine residues correspond to the infrared bands at 1532 cm−1 and 1482 cm−1 respectively. The band at 1456 cm− 1 (assigned to C6_N6 vibrations of adenine base) is regarded as A-conformational marker band of RNA. The band at 1396 cm−1 arises due to C4C5H and C6C5H stretching vibrations of ribouridine nucleosides. Moreover, C3′endo/anti sugar conformation in rA_rU base pairs correspond to the band at 1338 cm− 1. The sugar-phosphate backbone stretching vibrational frequencies appeared as strong bands in 1250– 1080 cm−1 region of free tRNA spectrum. The two vibrational bands at 1237 and 1083 cm−1 are attributed mainly to the asymmetric and symmetric stretching vibrations of the phosphate groups respectively. Ring vibrations of ribose sugar lead to rise of infrared bands at 1223 cm−1 and 1119 cm−1. The spectral region 1080–800 cm−1 contains various infrared marker bands of RNA in A-conformation, corresponding to ribose sugar conformation coupled with phosphodiester chain stretching vibrational modes. The infrared band at 996 cm−1 is assigned to sugar-phosphate backbone stretching vibrations involving 2′OH group. Other bands at 968 cm− 1, 914 cm− 1, 866 cm−1, and 810 cm−1 are attributed to ribose sugar conformation in C3′endo/anti type [1,42,43]. Alteration in band position as well as in intensity of these IR spectral bands is due to the corresponding structural-conformational transitions in tRNA allied to its interaction with drug. 3.2.1. Lomustine (LMT)-tRNA Complexes Lomustine-tRNA complex formation leads to major shift of 3 cm−1 in the IR bands accredited to cytosine (1532 cm− 1 and 1482 cm− 1) and guanine (1705 cm−1) nitrogenous bases (Fig. 3A). No appreciable shift was observed for adenine (1602 cm−1) and uracil (1396 cm− 1 and 1650 cm−1) bands in the complex spectrum. No change in rA = rU base pair vibrations was observed as evident from the unchanged position of the IR band at 1338 cm−1. Furthermore, the A-form marker band at 1456 cm−1 also did not undergo any noticeable change. However, a shift of around 2–3 cm−1 and 1–2 cm−1 was observed in the vibrational frequencies of ribose sugar vibrations (1223 cm−1, 866 cm−1 and 1119 cm−1) and phosphate stretching vibrations (1237 cm− 1 and 1083 cm−1) respectively. Moreover, a small shift has also been noticed in sugar-phosphate backbone vibrations (996 cm−1) of tRNA in the complex spectrum. IR spectral shifts in cytosine of tRNA are also accompanied with an enhancement in intensity as evident by the positive

Fig. 2. A Left panel: The docked model of lomustine with tRNA illustrating the best-docked conformer of lomustine-tRNA complex, Right panel: Close view of lomustine-tRNA docked conformer with lowest energy depicting residues of tRNA, which are in close contact with the drug. The hydrogen bonds between the drug and tRNA are shown as green dotted lines. The van der Waals interactions are presented as thin wireframe spheres on those pairs of atoms, which are closer than the sum of their van der Waals radii. B Left panel: Docked model of tRNA with nitrosourea derivative nimustine, Right panel: Nimustine-tRNA complex illustrating residues of tRNA that are near to drug. The hydrogen bonds formed between drug and tRNA are shown as green dotted lines. The van der Waals interactions are presented as thin wireframe spheres on those pairs of atoms, which are closer than the sum of their van der Waals radii. C Left panel: The docked model of semustine with tRNA, Right panel: Semustine-tRNA complex showing residues of tRNA, which are in close contact with the drug. Moreover, the hydrogen bonds formed between drug and tRNA are depicted as green dotted lines. The van der Waals interactions are presented as thin wireframe spheres on those pairs of atoms, which are closer than the sum of their van der Waals radii. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

6

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

Fig. 3. (A) Stacked view of FTIR spectra of free tRNA and its complexes with nitrosourea derivative lomustine at different molar ratios. The spectra were collected in the region of 1800 cm−1 to 600 cm−1. (B) Difference spectra of lomustine-tRNA complexes in the region of 1800 cm−1 to 600 cm−1. [Difference spectra = (tRNA solution + lomustine solution) − (tRNA solution)].

bands at 1527 cm−1 and 1490 cm− 1 in the difference spectra of lomustine-RNA complexes [(tRNA solution + lomustine solution) – tRNA solution] (Fig. 3B). Besides this, guanine base stretching vibrations exhibit reduction in intensity (as shown by negative band at 1710 cm−1) upon tRNA complexation with lomustine. Moreover, deviations in intensity due to stretching vibrations were also observed in ribose sugar residues, phosphate groups and phosphate-sugar backbone (as apparent by positive bands at 1238, 1118, 1081, 989 and

861 cm−1) of nucleic acid. Spectral variations observed in terms of shifts and intensity deviations for base vibrations of tRNA suggest the direct binding of lomustine with heterocyclic bases of tRNA with major emphasis on cytosine and guanine [1,44–47]. Previous investigations conducted on LMT-DNA complexes suggest that LMT binds to DNA by transferring its chloroethyl moiety to guanine residue [20]. Some reports available in literature specify that lomustine undergoes oxidation and form geometric mono-hydroxylated isomers (trans-4′-CCNU and

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

7

Fig. 4. (A) Upper panel exhibits stacked view of FTIR spectra of yeast free tRNA and its complexes with chloroethyl nitrosourea derivative nimustine at different molar ratios. (B) Lower panel illustrates corresponding difference spectra of nimustine-tRNA complexes. All the spectral were collected in the scan range of 1800 cm−1 to 600 cm−1.

cis-4′-CCNU), which could either bound to O6 position of guanine (produces O6-chloroethyl guanine) or N7 atom of guanine (produces N7chloroehtly guanine). Thereafter, O6-chloroethyl guanine can stimulate the formation of intra-molecular N1-O6-ethano guanine adduct, which subsequently may react with N3 of adjacent cytosine (in complementary strand) to produce G-C crosslink (1-(3-cytosinyl)-2-(1-guanosinyl)-

ethane) as a secondary product [20]. Since DNA and RNA both are nucleic acid, similarity in the binding pattern of lomustine with both of these vital molecules can be expected. Based on IR spectral data analysis of LMT-tRNA complexes, we propose that lomustine interacts with tRNA through guanine nitrogenous base and can possibly crosslink cytosine residues in the paired region of tRNA. Lomustine binding and

8

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

Fig. 5. (A) Stacked view of FTIR spectra in the region of 1800–600 cm−1 for free yeast tRNA and semustine-tRNA complexes at different molar ratios. (B) Difference spectra of semustine– tRNA complexes in the region of 1800–600 cm−1.

cross-linking to RNA probably renders RNA non-functional and halts its normal cellular processes such as protein translation, which may lead to cell death. 3.2.2. Nimustine (NMT)-tRNA Complexes Detailed analysis of NMT-tRNA complex using IR spectrum reveals major shift of around 4 cm−1 in guanine at 1705 cm−1 (from 1705 to 1701 cm− 1) and 2–3 cm−1 in uracil (1650 cm− 1 and 1396 cm− 1) base stretching vibrations upon NMT-RNA complex formation at highest molar ratio (r-1/20) (Fig. 4A). Cytosine stretching vibrational frequencies at 1532 cm−1 and 1482 cm−1 exhibit a shift of 2 cm−1 at highest drug concentration taken in the study. No appreciable shift was observed for frequencies corresponding to adenine (1602 cm−1) and adenine-uracil base (1338 cm−1) stretching vibrations. In addition, the A-conformation marker band (1456 cm−1) also remains same in the

position of NMT-RNA complexation. Besides this, 2 to 4 cm−1 spectral shift was noticed for the ribose sugar stretching vibrations (1223 cm−1, 866 cm−1 and 1119 cm−1), while bands assigned to phosphate stretching vibrations and sugar-phosphate backbone (1237, 1083, 996 and 968 cm− 1) underwent minor shift (1 to 2 cm− 1). In the difference spectra of nimustine-RNA complexes [(RNA solution + nimustine solution) – RNA solution] (Fig. 4B), IR spectral shifts in the heterocyclic nitrogenous bases of tRNA were also accompanied with deviation in intensity. Negative band at 1712 cm−1 is ascribed to decrease in stretching vibrational intensity of guanine residues in tRNA after its complexation with nimustine. Moreover, positive bands at 1659 cm−1 and 1380 cm−1 signify the enhancement in vibrational intensity of uracil residues stretching. In our previous study, we have observed that NMT binds DNA by interacting primarily with guanine and thymine nitrogenous bases located in the major groove of DNA [48]. Our Spectral

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

9

Fig. 6. Circular dichroism spectra of free tRNA and CENUs -tRNA complexes at different molar ratios (r) 1/20 (blue line), 1/40 (green line) and 1/80 (red line) after 2 h of incubation period. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

data collected for NMT-RNA complex formation also indicates towards the possibility of NMT initial interaction with guanine and uracil moiety located in the major groove of RNA. As the geometry of RNA molecule determines the recognizability of the major and minor grooves. In Aconformation of RNA helix, major groove is deep and narrow displaying a richer assembly of arrayed hydrogen bond donors and acceptors [49, 50]. Thus, interaction of nimustine with guanine and uracil residues in RNA groove, helps in widen the major groove of RNA helices thereby increasing the accessibility of other proteins [50] (such as alkylating enzymes). Preliminary positioning of nimustine within RNA major groove is strengthened by hydrophobic interaction that facilitates generation of stable contact of nimustine with RNA helix [49,50]. This can be followed by the alkylation of nitrogenous bases, which leads to the generation of rG-rC cross-links in nucleic acid. Spectral outcomes, presented here, corroborates well with the fact that “nitrosoureas are known for alkylation and majority of alkylating agent are major groove binders” [51]. 3.2.3. Semustine (SMT)-tRNA Complexes SMT behaved in a much similar manner of NMT, while interacting with RNA. The prominent vibrations affected by the SMT-tRNA complex formation are guanine and uracil bases. Guanine band position underwent a change of 4 cm−1 from 1705 to 1709 cm−1, while uracil band is shifted to 1653 cm−1 from 1650 cm−1 by experiencing a shift of 3 units (Fig. 5A). Bands assigned to cytosine stretching vibrations (1532 cm− 1 and 1482 cm− 1) exhibit a shift of 2–3 cm−1 at highest drug-RNA molar ratio studied. No considerable shift could be seen for adenine (1602 cm− 1), adenine-uracil base pair (1338 cm− 1) and Aform marker band (1456 cm−1). Ribose sugar vibrations (1223 cm−1,

866 cm−1 and 1119 cm−1) were noticed for a shift of around 2 to 4 cm−1. Neither phosphate stretching vibrations in sugar-phosphate backbone (996 cm− 1 and 968 cm− 1) nor phosphate symmetric and asymmetric stretching vibrations (1223 cm− 1 and 1083 cm− 1) underwent any alteration in terms of their vibrational bands position. In the difference spectra of semustine-tRNA complexes [(RNA solution + semustine solution) – RNA solution] (Fig. 5B), deviation in intensity was observed due to stretching vibrations of guanine and uracil residues as indicated by positive bands (1661 cm−1, 1530 cm− 1, and 1491 cm−1) and negative bands (1710 cm−1). Based on these spectral alterations, it could be suggested that semustine interacts directly with uracil and G = C base pair along with a minor binding to sugar residues in RNA backbone [1,46,47]. 3.3. CD Spectral Analysis To further probe the conformational transitions in tRNA due to its binding with nitrosourea derivatives (lomustine, nimustine and semustine) CD spectra of free yeast tRNA and CENUs-RNA complexes at different molar ratios were collected (Fig. 6). Native conformation (A-form) of RNA is characterized by a typical CD spectrum exhibiting a strong positive band at 268 nm, a negative band at 210 nm and a weaker negative band at 242 nm. Positive band at 268 nm is attributed to nitrogenous base-stacking interaction in nucleic acid helices. Band at 242 nm arises due to rigid helical domains of single stranded region of RNA formed by parallel orientation of adjacent aromatic rings of the nitrogenous bases. Altogether, shifts and intensity variation in characteristic CD bands (268 nm and 242 nm) are used to identify helix-coil transition (denaturation) in ribonucleic acid due to its interaction with

Fig. 7. UV–visible spectra of free yeast tRNA (0.04 mM) in the absence and presence of CENUs. Intersects show double reciprocal plot of CENUs binding to tRNA. A0 and A is the absorption of tRNA at 260 nm in free and complexed state respectively. C is the analytical concentration of drug in solution.

10

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11

ligand. Appearance of band at 210 nm depends upon the β-N glycosidic linkage occurring between the ribose sugar moiety and nitrogenous base residue. Besides this, there is an emergence of a band at 223 nm with minimum positive ellipticity. This band (223 nm) is accredited to hydrogen bonding between the nitrogenous base pairs in the double stranded region of RNA [52–55]. Therefore, alterations in position and in intensity of these CD spectral bands can provide information about structural and conformational variations in RNA upon its complexation with nitrosourea derivatives. Upon complexation of RNA with nitrosourea derivatives, no major shift was observed in CD bands corresponding to ribonucleic acid component. However, the band at 242 nm, which is attributed to rigid helical domains of single stranded portions of RNA, shows a blue shift of 5 nm (from 242 nm to 237 nm) in the CD spectra of nimustine-RNA and semustine-RNA complexes. This blue shift is accompanied with increase in positive molar ellipticity (or decrease in magnitude) at 242 nm. This shift and intensity variations are more likely to be associated with the formation of long stem-loop structure in RNA molecule [56,57]. The development of such stem-loop structure in biomolecule (tRNA) can be the consequence of guanine residues alkylation. The alkylated nitrogenous bases may generate intrastrand crosslink, which in turn promote the alignment of RNA strands adjacent to each other and produces stem-loop structure. Furthermore, it has been reported that nimustine and semustine induces groove-directed alkylation in DNA [19,23]. In addition, double stranded region of RNA has a wideshallow minor groove and narrow-deep major groove [49,50]. Therefore, formation of long stem-loop structure in RNA molecule can be considered as a part of anticancer action of nimustine and semustine, which provides accessibility to CENUs to bind with RNA in a groove-oriented manner. This observation also falls in line with our FTIR spectral outcome of nimustine and semustine interaction with RNA. However, such shift and intensity deviation related to 242 nm CD band has not been observed in case of lomustine-tRNA complexes, indicating simple alkylation of nitrogenous bases by lomustine as its anti-malignant action mechanism, in comparison to groove-directed-binding of its other two counter molecules. Besides this, at other dichroic components (268 nm, 223 nm and 210 nm) of RNA, slight augmentation in molar ellipticity was noticed after complexation of tRNA with all three nitrosourea derivatives (lomustine, nimustine and semustine). The band at 268 nm (assigned to base stacking interaction) shows an enhancement in positive molar ellipticity with the increase in CENUs concentration. Nimustine-RNA and semustine-RNA complexes exhibit ~ 44% of amplification in molar ellipticity at 268 nm band, while lomustine-RNA complexes illustrate ~ 25% of enhancement. This enhancement in molar ellipticity (at 268 nm) suggests that interaction of drug with RNA induces an augmentation in secondary structure of nucleic acid favored by additional stacking of single stranded regions of RNA [58]. Moreover, slight variation in molar ellipticity at 210 nm and 223 nm indicates that nitrosourea derivatives interaction with tRNA causes reduction of water molecules surrounding nucleic acid due to the presence of drug's additional hydrophobic alkyl groups [54]. Since, there is no major spectral shifts in the bands at 268 nm and 210 nm, hence, RNA remains in A-conformation after its interaction with all three nitrosourea derivatives studied. 3.4. Absorption Spectral Analysis 3.4.1. Binding Mode The absorption spectra of free yeast tRNA and its complexes with varying concentrations of CENUs (lomustine, nimustine and semustine) are shown in Fig. 7. The absorption spectra of tRNA exhibit similar behavior on addition of CENUs. When CENUs interact with tRNA, hyperchromic effect was observed in the absorption maxima of tRNA with increasing concentration of drugs in solution. This hyperchromic effect can be ascribed to the interaction of CENUs with tRNA (by the means of nitrogenous base alkylation). Subsequently, alkylated bases

may generate interstrand cross-links resulting in localized distortion in stem-loop structure of tRNA (as indicated by FTIR and CD spectral results). Further, this localized deformation may induce more exposure of bases to UV radiation that leads to an enhancement in UV radiation absorption [1]. 3.4.2. Binding Strength of CENUs-tRNA Complexes The binding constant (Ka) is calculated for the quantitative measurement of binding of CENUs with tRNA. It is calculated by observing the changes in optical density at 260 nm for free tRNA (represented by A0) and its complexes with CENUs (represented by A). It includes the preparation of a series of drug-tRNA complex solutions in which the concentration of tRNA is held constant, while the drug concentration [C] is varied from 0.4 × 10−1 to 4 × 10−2 mM. The double reciprocal plot of 1/(A − A0) vs. 1/ [C] is linear and the binding constant (Ka) can be estimated from the ratio of intercept to slope (intersects of Fig. 7). The binding constants (Ka) for the interaction of lomustine, nimustine and semustine with tRNA are 2.55 × 102 M− 1 , 4.92 × 102 M−1 and 4.22 × 102 M−1 respectively, which indicate weak type of CENUs binding with tRNA [40]. Furthermore, experimentally observed binding free energies for lomustine-tRNA (−3.2 kcal/mol), nimustine-tRNA (−3.67 kcal/mol) and semustine-tRNA (−3.58 kcal/mol) complexes, as calculated from the binding constant (Ka) (obtained from UV–visible spectroscopic results) are compared with the binding free energy values, predicted by docking simulations and correlated well [35,59,60]. 4. Conclusion In the present work, we carried out molecular docking and spectroscopic investigations on the binding properties of chloroethyl nitrosourea derivative lomustine, nimustine and semustine with tRNA. FTIR spectral outcomes suggest that CENUs interact with guanine and cytosine residues of tRNA in addition to slight binding with its sugarphosphate backbone, authenticating molecular modeling prediction. However, in case of nimustine-tRNA and semustine-tRNA complexes, preliminary binding with uracil residue is observed as indicated by IR spectral analysis, which augments the possibility of groove-directed alkylation as their anticancer action mechanism. CD spectroscopic data signifies no conformational change in native A-form of biomolecule after its complexation with CENUs. Furthermore, UV–visible studies affirm weak type of binding of CENUs with tRNA. These findings may further contribute to understand the action mechanism of chloroethyl nitrosourea derivative at molecular level. Finally, we can conclude that ligand-RNA interaction studies are fundamental to unravel the mystery of molecular recognition in general and RNA targeting in particular. Acknowledgment The authors thank Director, CSIR-National Physical Laboratory, New Delhi, India, for granting the permission for publication of the work. S. A is thankful to Council of Scientific & Industrial Research, New Delhi, India, for providing financial support. References [1] G. Tyagi, S. Agarwal, R. Mehrotra, tRNA binding with anti-cancer alkaloids–nature of interaction and comparison with DNA–alkaloids adducts, J. Photochem. Photobiol. B Biol. 142 (2015) 250–256. [2] T. Hermann, E. Westhof, Rational drug design and high-throughput techniques for RNA targets, Comb. Chem. High Throughput Screen. 3 (2000) 219–234. [3] W.D. Wilson, K. Li, Targeting RNA with small molecules, Curr. Med. Chem. 7 (2000) 73–98. [4] K.A. Xavier, P.S. Eder, T. Giordano, RNA as a drug target: methods for biophysical characterization and screening, Trends Biotechnol. 18 (2000) 349–356. [5] J. Gallego, G. Varani, Targeting RNA with small-molecule drugs: therapeutic promise and chemical challenges, Acc. Chem. Res. 34 (2001) 836–843. [6] T. Hermann, Rational ligand design for RNA: the role of static structure and conformational flexibility in target recognition, Biochimie 84 (2002) 869–875. [7] Y. Tor, Targeting RNA with small molecules, Chembiochem 4 (2003) 998–1007.

S. Agarwal et al. / Journal of Photochemistry & Photobiology, B: Biology 166 (2017) 1–11 [8] R.P. Perry, The cellular sites of synthesis of ribosomal and 4S RNA, Proc. Natl. Acad. Sci. U. S. A. 48 (1962) 2179–2186. [9] J.D. Watson, Involvement of RNA in the synthesis of proteins, Science 140 (1963) 785–808. [10] U.E. Loening, Molecular weights of ribosomal RNA in relation to evolution, J. Mol. Biol. 38 (1968) 355–365. [11] P. Cohen, Protein kinases—the major drug targets of the twenty-first century? Nat. Rev. Drug Discov. 1 (2002) 309–315. [12] P. Kandagal, S. Ashoka, J. Seetharamappa, S. Shaikh, Y. Jadegoud, O. Ijare, Study of the interaction of an anticancer drug with human and bovine serum albumin: spectroscopic approach, J. Pharm. Biomed. Anal. 41 (2006) 393–399. [13] A.M. Gallina, P. Bork, D. Bordo, Structural analysis of protein-ligand interactions: the binding of endogenous compounds and of synthetic drugs, J. Mol. Recognit. 27 (2014) 65–72. [14] A. Speck-Planche, V.V. Kleandrova, F. Luan, M. Cordeiro, Rational drug design for anti-cancer chemotherapy: multi-target QSAR models for the in silico discovery of anti-colorectal cancer agents, Bioorg. Med. Chem. 20 (2012) 4848–4855. [15] J.E. McCormick, R. Stanley McElhinney, Nitrosoureas from chemist to physician: classification and recent approaches to drug design, Eur. J. Cancer Clin. Oncol. 26 (1990) 207–221. [16] L. Zhao, R. Zhong, Y. Zhen, An ONIOM study on the crosslinked base pairs in DNA reacted with chloroethylnitrosoureas, J. Theor. Comput. Chem. 6 (2007) 631–639. [17] J. Larkin, S. Hughes, D. Beirne, P. Patel, I. Gibbens, S. Bate, K. Thomas, T. Eisen, M. Gore, A phase I/II study of lomustine and temozolomide in patients with cerebral metastases from malignant melanoma, Br. J. Cancer 96 (2006) 44–48. [18] R. Mehrotra, D.K. Jangir, S. Agarwal, B. Ray, P. Singh, A. Srivastava, Interaction studies of anticancer drug lomustine with calf thymus DNA using surface enhanced Raman spectroscopy, Mapan 28 (2013) 273–277. [19] S. Agarwal, D. Chadha, R. Mehrotra, Molecular modeling and spectroscopic studies of semustine binding with DNA and its comparison with lomustine–DNA adduct formation, J. Biomol. Struct. Dyn. 33 (2014) 1653–1668. [20] S. Agarwal, D.K. Jangir, P. Singh, R. Mehrotra, Spectroscopic analysis of the interaction of lomustine with calf thymus DNA, J. Photochem. Photobiol. B Biol. 130 (2014) 281–286. [21] K. Mineura, S. Fushimi, Y. Itoh, M. Kowada, DNA lability induced by nimustine and ramustine in rat glioma cells, J. Neurol. Neurosurg. Psychiatry 51 (1988) 1391–1394. [22] M. Tanaka, S. Shibui, K. Nomura, Y. Nakanishi, Radiotherapy combined with nimustine hydrochloride and etoposide for malignant gliomas: results of a pilot study, Jpn. J. Clin. Oncol. 31 (2001) 246–250. [23] S. Agarwal, D.K. Jangir, R. Mehrotra, N. Lohani, M. Rajeswari, A structural insight into major groove directed binding of nitrosourea derivative nimustine with DNA: a spectroscopic study, PLoS One 9 (2014), e104115. [24] J.D. Boice Jr., M.H. Greene, J.Y. Killen Jr., S.S. Ellenberg, R.J. Keehn, E. McFadden, T.T. Chen, J.F. Fraumeni Jr., Leukemia and preleukemia after adjuvant treatment of gastrointestinal cancer with semustine (methyl-CCNU), N. Engl. J. Med. 309 (1983) 1079–1084. [25] W. Tang, L.-J. Zhao, R.-g. Zhong, Agarose Gel Electrophoresis, Fluorometric Assays for the Determination of DNA Cross-Linking Induced by Semustine, Bioinformatics and Biomedical Engineering, 2008. ICBBE 2008, The 2nd International Conference on, IEEE, 2008, pp. 322–325. [26] L.-J. Zhao, T. Ren, B.-Q. Bai, R. Zhang, R.-G. Zhong, Comparative investigations of deoxyribonucleic acid interstrand crosslinks induced by semustine using fluorescence and high performance liquid chromatography-mass spectrometry, Chin. J. Anal. Chem. 4 (2011) 476–480. [27] V. Bono Jr., Review of mechanism of action studies of the nitrosoureas, Cancer Treat. Rep. 60 (1976) 699–702. [28] M. Johnston, A Review of Studies on the Mechanism of Action of Nitrosoureas, 1976 87–119. [29] F. Schabel Jr., Nitrosoureas: a review of experimental antitumor activity, Cancer Treat. Rep. 60 (1976) 665–698. [30] C.T. Gombar, W.P. Tong, D.B. Ludlum, Mechanism of action of the nitrosoureas—IV: reactions of bis-chloroethyl nitrosourea and chloroethyl cyclohexyl nitrosourea with deoxyribonucleic acid, Biochem. Pharmacol. 29 (1980) 2639–2643. [31] W.P. Tong, M.C. Kirk, D.B. Ludlum, Mechanism of action of the nitrosoureas—V: formation of O6-(2-fluoroethyl) guanine and its probable role in the crosslinking of deoxyribonucleic acid, Biochem. Pharmacol. 32 (1983) 2011–2015. [32] A. Naghipur, M.G. Ikonomou, P. Kebarle, J.W. Lown, Mechanism of action of (2haloethyl) nitrosoureas on DNA: discrimination between alternative pathways of DNA base modification by 1, 3-bis (2-fluoroethyl)-1-nitrosourea by using specific

[33]

[34] [35]

[36] [37] [38]

[39] [40]

[41] [42] [43] [44]

[45] [46]

[47] [48]

[49]

[50] [51] [52]

[53]

[54]

[55] [56] [57] [58] [59] [60]

11

deuterium labeling and identification of reaction products by HPLC/tandem mass spectrometry, J. Am. Chem. Soc. 112 (1990) 3178–3187. K.U. Schallreuter, F.K. Gleason, J.M. Wood, The mechanism of action of the nitrosourea anti-tumor drugs on thioredoxin reductase, glutathione reductase and ribonucleotide reductase, Biochim. Biophys. Acta Mol. Cell Res. 1054 (1990) 14–20. J. Glasel, Validity of nucleic acid purities monitored by 260 nm/280 nm absorbance ratios, BioTechniques 18 (1995) 62–63. G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639–1662. J.J. Irwin, T. Sterling, M.M. Mysinger, E.S. Bolstad, R.G. Coleman, ZINC: a free tool to discover chemistry for biology, J. Chem. Inf. Model. 52 (2012) 1757–1768. N.M. O'Boyle, M. Banck, C.A. James, C. Morley, T. Vandermeersch, G.R. Hutchison, Open babel: an open chemical toolbox, J. Cheminform. 3 (2011) 1–14. J.L. Sussman, S.R. Holbrook, R.W. Warrant, G.M. Church, S.-H. Kim, Crystal structure of yeast phenylalanine transfer RNA: I. Crystallographic refinement, J. Mol. Biol. 123 (1978) 607–630. M.F. Sanner, Python: a programming language for software integration and development, J. Mol. Graph. Model. 17 (1999) 57–61. S. Agarwal, D.K. Jangir, R. Mehrotra, Spectroscopic studies of the effects of anticancer drug mitoxantrone interaction with calf-thymus DNA, J. Photochem. Photobiol. B Biol. 120 (2013) 177–182. K.A. Connors, Binding Constants: The Measurement of Molecular Complex Stability, Wiley, New York, 1987 141–188. M. Banyay, M. Sarkar, A. Gräslund, A library of IR bands of nucleic acids in solution, Biophys. Chem. 104 (2003) 477–488. B. Stuart, Infrared Spectroscopy, Wiley Online Library, 2005. M.M. Islam, P. Pandya, S.R. Chowdhury, S. Kumar, G.S. Kumar, Binding of DNA-binding alkaloids berberine and palmatine to tRNA and comparison to ethidium: spectroscopic and molecular modeling studies, J. Mol. Struct. 891 (2008) 498–507. C. N'Soukpoé-Kossi, H. Tajmir-Riahi, DNase I-tRNA interaction, Oligonucleotides 18 (2008) 65–72. R. Marty, C.N. N'soukpoe-Kossi, D.M. Charbonneau, L. Kreplak, H.-A. Tajmir-Riahi, Structural characterization of cationic lipid–tRNA complexes, Nucleic Acids Res. 37 (2009) 5197–5207. A.A. Ouameur, P. Bourassa, H.-A. Tajmir-Riahi, Probing tRNA interaction with biogenic polyamines, RNA 16 (2010) 1968–1979. S. Agarwal, B. Ray, R. Mehrotra, SERS as an advanced tool for investigating chloroethyl nitrosourea derivatives complexation with DNA, Int. J. Biol. Macromol. 81 (2015) 891–897. L. Chen, A.D. Frankel, A peptide interaction in the major groove of RNA resembles protein interactions in the minor groove of DNA, Proc. Natl. Acad. Sci. 92 (1995) 5077–5081. K.M. Weeks, D.M. Crothers, Major Groove Accessibility of RNA, Science-New York then Washington, 261, 1993 1574. S. Pelengaris, M, Khan, The Molecular Biology of Cancer, A Bridge from Bench to Bedside, Wiley. com, 2013. G.T. Walker, Deoxycytidine methylation does not affect DNA-RNA hybrid formation or BA transitions of (dG) n.(dC) n sequences, Nucleic Acids Res. 16 (1988) 3091–3099. L. Dassonneville, F. Hamy, P. Colson, C. Houssier, C. Bailly, Binding of Hoechst 33258 to the TAR RNA of HIV-1. Recognition of a pyrimidine bulge-dependent structure, Nucleic Acids Res. 25 (1997) 4487–4492. M.P. McPike, J. Goodisman, J.C. Dabrowiak, Footprinting and circular dichroism studies on paromomycin binding to the packaging region of human immunodeficiency virus type-1, Bioorg. Med. Chem. 10 (2002) 3663–3672. S.R. Martin, M.J. Schilstra, Circular dichroism and its application to the study of biomolecules, Methods Cell Biol. 84 (2008) 263–293. X. Sun, Characterization of E Coli Hfq Structure and its RNA Binding Properties, Georgia Institute of Technology, 2006. S. Nafisi, M. Adelzadeh, Z. Norouzi, M.N. Sarbolouki, Curcumin binding to DNA and RNA, DNA Cell Biol. 28 (2009) 201–208. T. Antony, M. Atreyi, M. Rao, Interaction of methylene blue with transfer RNA—a spectroscopic study, Chem. Biol. Interact. 97 (1995) 199–214. R. Huey, G.M. Morris, A.J. Olson, D.S. Goodsell, A semiempirical free energy force field with charge-based desolvation, J. Comput. Chem. 28 (2007) 1145–1152. G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility, J. Comput. Chem. 30 (2009) 2785–2791.