Interactions of 2-phenyl-benzotriazole xenobiotic compounds with human Cytochrome P450-CYP1A1 by means of docking, molecular dynamics simulations and MM-GBSA calculations

Accepted Manuscript Title: Interactions of 2-phenyl-benzotriazole xenobiotic compounds with human Cytochrome P450-CYP1A1 by means of docking, molecular dynamics simulations and MM-GBSA calculations Authors: Karel Mena-Ulecia, Desmond MacLeod-Carey PII: DOI: Reference:

S1476-9271(17)30702-8 https://doi.org/10.1016/j.compbiolchem.2018.04.004 CBAC 6836

To appear in:

Computational Biology and Chemistry

Received date: Revised date: Accepted date:

4-10-2017 4-4-2018 6-4-2018

Please cite this article as: Mena-Ulecia, Karel, MacLeod-Carey, Desmond, Interactions of 2-phenyl-benzotriazole xenobiotic compounds with human Cytochrome P450-CYP1A1 by means of docking, molecular dynamics simulations and MM-GBSA calculations.Computational Biology and Chemistry https://doi.org/10.1016/j.compbiolchem.2018.04.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Interactions of 2-phenyl-benzotriazole xenobiotic compounds with human

Cytochrome

P450-CYP1A1

by

means

of

docking,

molecular dynamics simulations and MM-GBSA calculations

1Universidad

SC RI PT

Karel Mena-Ulecia1 and Desmond MacLeod-Carey*2 Autónoma de Chile, Computational and Theoretical Chemistry Center, Instituto de

Ciencias Químicas Aplicadas. Facultad de Ingeniería, Avenida Alemania 1090, Temuco, Chile. 2Universidad

Autónoma de Chile, Computational and Theoretical Chemistry Center, Instituto de

Ciencias Químicas Aplicadas. Facultad de Ingeniería, El Llano Subercaseaux 2801, San Miguel, Santiago, Chile.

A

CC

EP

TE

D

M

A

N

U

Graphical abstract

Highlights  Molecular docking analyses reproduce binding modes of 2-phenil-benzotriazoles pro-mutagens with human Cytochrome P450-CYP1A1 within 1,5 Å of RMSD. 
  Molecular dynamics simulations show stabilization of the PBTAs into the pocket of Cytochrome P450-CYP1A1 through π-π stacking interactions between the triazole group and Phe224, as well as hydrogen bonds of the terminal -NH2 of the benzotriazole units with Asn255 and Ser116 aminoacids. 
The binding energy results from MM-GBSA were consistent with the experimental metabolic activation, with the highest binding energy corresponding to the PBTA-4- CYP1A1 complex.

1



Molecular dynamics and MMGBSA calculations show that modification of the amino group over the phenyl group on PBTAs generate subtle changes in the interaction between PBTAs and Cytochrome P450-CYP1A1.

Abstract 2-phenyl-benzotriazole xenobiotic compounds (PBTA-4, PBTA-6, PBTA-7 and PBTA-8) that were previously isolated and identified in waters of the Yodo river, in Japan (Nukaya et al., 2001; Ohe et

SC RI PT

al., 2004; Watanabe et al., 2001) were characterized as powerful pro-mutagens. In order to predict the activation mechanism of these pro-mutagens, we designed a computational biochemistry

protocol, which includes, docking experiments, molecular dynamics simulations and free energy decomposition calculations to obtain information about the interaction of 2-phenyl-benzotriazole

molecules into the active center of cytochrome P450-CYP1A1 (CYP1A1). Molecular docking calculations using AutoDock Vina software shows that PBTAs are proportionally oriented in the pocket of CYP1A1, establishing - stacking attractive interactions between the triazole group and

the Phe224, as well as, the hydrogen bonds of the terminal NH 2 over the benzotriazole units with

U

the Asn255 and Ser116 amino acids. Molecular dynamics simulations using NAMD package

N

showed that these interactions are stable along 100.0 ns of trajectories. Into this context, free binding energy calculations employing the MM-GBSA approach, shows that some differences exists

A

among the interaction of PBTAs with CYP1A1, regarding the solvation, electrostatic and van der

M

Waals interaction energy components. These results suggest that PBTA molecules might be activated by CYP1A1. Thus, enhancing their mutagenicity when compared with the pro-mutagen

D

parent species.

GBSA

EP

1. Introduction

TE

Keywords: 2-phenyl-benzotriazole, Molecular Docking, Molecular Dynamics Simulations, MM-

The most important source of water pollution is generated by human activities (Crites et al., 2014; Mena-Ulecia and Hernandez, 2015; Michalski and Ficek, 2016; Osibanjo et al., 2011). The

CC

industrial processes generate a large amount of reactive compounds that are introduced into superficial waters, where they accumulate, concentrate and/or react with other elements to generate secondary pollutants with the capacity to affect the genetic material of several species,

A

including human beings (Aleem and Malik, 2003; Garrido-Baserba et al., 2014; Molinos-Senante et al., 2014; Popovic et al., 2014). The textile industry generates one of the most aggressive residuals for ecosystems. Wastewater from this sector contains several kinds of contaminants, such as, heavy metal (mainly chromium and cadmium), nitrates and surfactants, which are used to set dyes (Cirik et al., 2013). Furthermore, several authors have detected mutagenic organic compounds in polluted waters from the textile industry, e.g. 2-phenyl-benzotriazole (PBTA) derivatives, which have been isolated and

2

identified from the waters of the Yodo river, in Japan (Nukaya et al., 2001; Ohe et al., 2004; Watanabe et al., 2001), and proven to be powerful pro-mutagens in assays of Salmonella typhimurium (Ames test) (Morisawa et al., 2003; Oda et al., 2008; Watanabe et al., 2006). This powerful pro-mutagens correspond to 2-[2-(acethylamino)-4-amino-5-methoxyphenyl]-5-amino-7bromo-4-chloro-2H-benzotriazole

(PBTA-4),

2-[2-(acethylamino)-4-[bis(2-hydroxyethyl)amino]-5-

methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole

(PBTA-6),

2-[2-(acethylamino)-4-

(diethylamino)-5-methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole (PBTA-7) and 2-[2-

SC RI PT

(acethylamino)-4-(diallylamino)-5-methoxyphenyl]-5-amino-7-bromo-4-chloro-2H-benzotriazole

(PBTA-8). The concept of pro-mutagen activity refers to those molecules that need a metabolic activation to exert their mutagenic power (Lepers et al., 2014; Oda et al., 2008). Several authors

have proposed a metabolic activation mechanism of xenobiotic compounds, where the first step is a N-hydroxylation reaction mediated by the cytochrome P450 (Borosky, 2008, 2007; Hilal et al., 2005). This protein is also involved in the metabolism of endogenous substrates and drugs, as well

as, the activation of certain toxins and environmental pollutants (Bansal et al., 2014; Uno et al.,

U

2014). In this context, (Oda et al., 2008) studied the most powerful mutagenic 2-phenylbenzotriazoles (PBTA-4, PBTA-6, PBTA-7 and PBTA-8) (Fig.1), with the purpose to establish the

N

role of cytochrome P450-CYP1A1 (CYP1A1) in the metabolic activation of these pollutants.

A

These authors determined that PBTAs, are activated by CYP1A1, however that did not determine the binding mode of these compounds towards the protein. In order to predict the binding

M

mechanism of these pro-mutagens, we designed a computational biochemistry protocol, which includes, docking experiments, molecular dynamics simulations and free energy decomposition

D

calculations to determine the interaction of these molecules into the active center of CYP1A1.

TE

2. Computational methodologies. Four 2-phenyl-benzotriazoles (PBTA-4, PBTA-6, PBTA-7 and PBTA-8) previously detected from

EP

polluted waters of Yodo river, in Japan (Morisawa et al., 2003; Nukaya et al., 2001; Ohe et al., 2004; Ono et al., 2000) (Fig.1) were labeled as potent pro-mutagens (Borosky, 2008). Here, we examine their molecular structure, and the possible interactions of PBTAs with the active center of

CC

CYP1A1, employing docking and molecular dynamics simulations to determine the differences arising from experimental genotoxicity. The molecular structures of the PBTAs were sketched using Maestro’s Molecular Editor (“Suite

A

2011: Maestro, version 9.2, Schrödinger, LLC, New York, NY, .,” 2011), and their geometries were optimized by using density functional theory (DFT) Perdew–Burke–Ernzerhof’s hybrid functional (PBE0)(Perdew et al., 1996) and the 6-311++g(d,p) basis set, as implemented in Gaussian09 code (Frisch et al., n.d.). The optimized geometry of each PBTA was used for further computational docking and molecular dynamics simulations.

2.1. Molecular docking protocol. The docking algorithms reproduce the bound form of target molecules inside the pocket of the 3

active site of proteins (Azam and Abbasi, 2013; Caballero et al., 2011; Faver and Merz, 2010; Janovec et al., 2011; Yuriev and Ramsland, 2013). To accomplish the location of PBTAs inside of the active site of CYP1A1, we used AutoDock Vina software (Azam and Abbasi, 2013). The X-ray crystal structure of human microsomal cytochrome P450 isoform 1A1 (CYP1A1) resolved to 2.6 Å and complexed with -naphthoflavone (PDB ID: 4I8V) (Walsh et al., 2013) was obtained from protein data bank (Berman et al., 2000). AutoDock Tools (ADT) (Morris et al., 2009a; Trott and Olson, 2010) were used to prepare CYP1A1 protein, PBTA ligand structures and a grid box of

SC RI PT

40.0x40.0x40.0 Å3 constructed with the center located at the mass center of α-naphthoflavone in the crystal structure of CYP1A1. The grid center was assigned at: x=-33.9248, y=82.3195 and z=-

22.9317, the energy range was set up at 5.0 kcal/mol and the ten best poses were selected. For each PBTA, these docking poses were analyzed by examining their relative total energy score and

the positional root-mean-square deviation (RMSD) (Gohlke et al., 2000). The most energetically favorable conformation of each PBTA-CYP1A1 complex was selected for further molecular

U

dynamics simulations.

2.2 Molecular Dynamics Simulation protocols.

N

The molecular structures of the PBTA-CYP1A1 complexes derived from docking experiments were

A

placed into a water box (10x10x10 Å3) centred on the mass center of each PBTA, using the TIP3P flexible water model (Kollman et al., 2000). To obtain the PBTAs topologies and parameters,

M

compatible with the CHARMM force field, we use the SwissParam web service (Zoete et al., 2011). The Protein and PBTAs were described by using CHARMM36 force field (Adasme-Carreno et al.,

D

2014) and the CGenFF force field (Mena-Ulecia et al., 2014), respectively. To reduce any close contacts, the conjugated gradient methodology (employing 20.000 steps) was employed as the

TE

energy minimization procedure. The van der Waals cutoff was fixed to 12Å and the temperature was kept at 298.15 K applying the weak coupling algorithm (Brian N and Charles L. Brooks, 1999). In all cases, a constraint of the backbone was applied using the NPT ensemble, and the long range

EP

electrostatic forces were considered by means of the Particle Mesh Ewald (PME) approach (Onufriev et al., 2004). The equations of motion were solved employing the velocity Verlet algorithm

CC

with a time step of 1.0 fs, and all the systems were subjected to 2.0 ns of equilibration. The production took place during 100.0 ns. All molecular dynamics simulations have been performed

A

using the NAMD software package (Phillips et al., 2005).

2.3. MM-GBSA free energy calculation We used molecular dynamics simulation protocols combined with MM-GBSA (Molecular Mechanics-Generalized Born Surface Area) (Kollman et al., 2000) to study the PBTA-CYP1A1 interaction. The MM-GBSA method use the calculated molecular mechanics energies and implicit solvation models to compute the difference between the energy of the bound complex (PBTACYP1A1) and the energies of the unbound protein (CYP1A1) and the isolated promutagen (PBTA) at a reasonable computational cost (Adasme-Carreno et al., 2014; Homeyer and Gohlke, 2012; 4

Mena-Ulecia et al., 2014) using the NAMD package (Phillips et al., 2005). Therefore the TIP3P water model is not included in MM-GBSA free energy calculations. During the 100.0 ns of molecular dynamics simulations, we took 1000 snapshots, to subsequently calculate the free binding energy of each PBTA-CYP1A1 complex through the MM-GBSA methodology. This method allows decomposing the binding free energy into the contributions originated from different types of physical-chemical interactions. Specifically, the energy is calculated for the ligand-protein complex (PBTA-CYP1A1), the unbound ligand (PBTA), and

SC RI PT

unbound protein (CYP1A1) using the CHARMM force field with the generalized Born implicit solvent model, to obtain the averaged binding free energy (∆Gbinding) according to equation 1: ∆Gbinding = ∆GPBTA−CYP1A1 − ∆GPBTA − ∆GCYP1A1

(1)

This binding free energy can be decomposed into three different energy terms as described in equation 2: ∆Gbinding = ∆Egas − ∆Gsolv − T∆S

(2)

Furthermore, ∆Egas correspond to the sum of electrostatic (∆Eelect), van der Waals (∆EvdW) and

U

internal (∆Einternal) energies, see equation 3.

(3)

N

∆Egas = ∆Einternal + ∆Eelect + ∆EvdW

Additionally, the ∆Einternal energy comprises the bond, angle and dihedral angle energies. This term

A

is cancelled, since the structures of PBTA, CYP1A1 and the PBTA-CYP1A1 complex were obtained from the same trajectory.

M

The conformational entropy change (−T∆S) can be estimated through normal-mode analysis on a set of conformational snapshots taken from the previous molecular dynamics simulations.

D

Nevertheless, many authors have reported that the lack of the evaluation of the entropy is not critical for the calculation of MM-GBSA (or MM-PBSA) binding free energies for related protein-

TE

ligand complexes (Adasme-Carreno et al., 2014; Genheden and Ryde, 2010; Massova and Kollman, 2000). Due to this reason, the entropy term −T∆S of each PBTA-CYP1A1 complex was

EP

not calculated.

On the other hand, ∆Gsolv correspond to the solvation free energy, which includes the polar and non-polar contributions, according to equation 4:

CC

∆Gsolv = ∆GGB + ∆GSA

(4)

The polar solvation free energy (∆GGB) is calculated using the generalized Born implicit solvent model. This ∆GGB includes the solute-solvent interactions and the work necessary to generate the

A

reaction field of the solvent induced by the charge distribution of the solute. This parameter it is estimated by using the following expression (Miyamoto et al., 2012; Onufriev et al., 2004): 𝑁

∆𝐺GB

𝑁

𝑞𝑖 𝑞𝑗 1 1 = − [1 − ] ∑ ∑ 2 𝜀 𝑓𝐺𝐵 (𝑟𝑖𝑗 𝑅𝑖 𝑅𝑗 )

(5)

𝑖=1 𝑗=1

where rij is the distance between the charges, Ri and Rj are the Born radius and 𝑓𝐺𝐵 is a function calculated as follows:

5

𝑓𝐺𝐵 = (𝑟𝑖𝑗2 + 𝑅𝑖 𝑅𝑗 𝑒

2 𝑟𝑖𝑗 − 4𝑅𝑖 𝑅𝑗

1 2

(6)

)

The non-polar solvation free energy (∆GSA) from equation 4 is estimated based on the accessibility of the solvent to the surface area (SA) where γ and β depends of the radii parameterization used to calculate surfaces; and the surface area is estimated through a spherical probe of 1.4 Å that rolls on the protein surface, according to equation 7: ∆𝐺𝑆𝐴 = γSA + β

SC RI PT

(7)

3. Results and discussion

Docking calculations allow us to predict the interaction of four different 2-phenyl-benzotriazoles,

PBTA-4, PBTA-6, PBTA-7 and PBTA-8, with CYP1A1. These PBTAs were proportionally oriented in the active center of CYP1A1 and their optimized docked structures are presented in Figure 2, showing the most adequate orientation obtained for this series. The orientation obtained from the

U

docking experiments reveal the presence of - stacking interactions between the triazole group

N

and the Phe224 amino acid, as well as the hydrogen bonds of the terminal −NH2 from PBTAs with Asn255 and Ser116 amino acids, see Figure 2. It must be noted that the interaction distances

A

between PBTAs and Phe224, Asn255 and Ser116 amino acids are different in all the studied xenobiotics.

M

The Root Means Standard Deviation (RMSD) was used as a criterion to determine if our docking methodology was adequate to reproduce the correct orientation of the PBTAs, in comparison with

D

the available crystallographic structure (α-naphthoflavone). As we showed in Table 1, the value of the RMSD of PBTAs regarding the X-ray crystal structure of the inhibitors was <1.5 Å in all cases.

TE

Here, we have considered that 2.0 Å it is the referential value that identifies a correct or incorrect resolution of the docking (Gohlke et al., 2000; Mena-Ulecia et al., 2015; Quesada-Romero et al.,

EP

2014). The results of the binding energy, obtained from the docking experiments, correspond to those from experimental metabolic activation (Morisawa et al., 2003; Oda et al., 2008; Ohe et al., 2008). As determined experimentally, PBTA-4 is the most genotoxic compound and PBTA-6 the

CC

less genotoxic from the series. The PBTA-4-CYP1A1 has the most negative binding energy (-9.7 kcal/mol). Hence, high stability of the PBTA-4-CYP1A1 complex into the active pocket of CYP1A1 could be the cause of their high mutagenicity observed experimentally (Morisawa et al., 2003;

A

Nukaya et al., 2001; Oda et al., 2008). On the contrary, PBTA-6-CYP1A1 complex exhibit the less negative binding energy (-8.8 kcal/mol) and the highest RMSD 1.30 (Å), being the least stable complex of the series. This low relative stability of the PBTA-6-CYP1A1 complex obtained from the docking experiments could explain the lower metabolic activation and genotoxicity observed experimentally for PBTA-6. Another aspect to be considered, is the presence of the heme group in the active pocket of CYP1A1, see Figure 3. According to docking calculations, PBTA-4, PBTA-7 and PBTA-8 are perpendicularly oriented to the heme group. The terminal -NH2 group is the closest one for PBTA-4 6

at a distance of 4.7 Å towards the iron atom of the heme group, Figure 3A. This is a large interatomic distance, thus electrostatic and van der Waals interactions should be negligible. Into this context, diethylamino and diallylamino groups in PBTA-7 and PBTA-8 are proximately located towards the heme group at 4.4 and 4.7 Å respectively, Figure 3C and 3D. On the contrary, PBTA-6 is orientated in a different way at the active pocket of CYP1A1, and their C2H4OH groups are closely oriented to the heme group (~2.7 Å), Figure 3B. This demonstrates the relevance of the different substituents toward the steric effects and orientation of these promutagens in CYP1A1,

SC RI PT

which can be related to the experimental degree of metabolic activation and genotoxicity.

Molecular dynamics (MD) simulations allow us to obtain the trajectories that contain all the

structural information regarding the stability and relevance of molecular interactions on the PBTACYP1A1 complexes and their evolution through time.

The stability of complexes along the MD trajectories was evaluated by using the root mean standard deviation (RMSD) of all the backbone atoms positions as a function of simulation time. This

parameter is critical to analyze the equilibration of MD trajectories. As it is displayed in Fig. 4A, the

U

RMSD had small fluctuations after 2 ns of simulation for all systems. None of these PBTA-CYP1A1

complexes exhibit values higher than 1.3 Å, this RMSD values are below of the 2.0 Å limit

N

established previously (Mena-Ulecia et al., 2015; Quesada-Romero et al., 2014), which indicates

A

that these complexes remain stable during the simulation time (Figure 4B). PBTA-4-CYP1A1 exhibits an average value of 1.13 Å during the trajectory, being the most stable complex along the

M

simulation time. That result is in agreement with experimental ones, where PBTA-4 presents the highest genotoxicity in vitro and also the highest experimental metabolic activation.

D

Root mean square fluctuation (RMSF) is a parameter used in molecular dynamics simulations to describe the flexibility of several residues. The RMSF of the backbone atoms of each residue in

TE

PBTA-CYP1A1 complexes were calculated to analyze the flexibility of the backbone structure. The higher RMSF value is, the larger the flexibility, whereas lower RMSF values indicate limited movements during simulation in relation to its average position. Calculated RMSF values of the

EP

residues are shown in Figure 5.

CC

It must be pointed out that those amino acid residues showing the higher fluctuations are mainly located at the active center of CYP1A1 interacting with the different PBTAs as follows. The residue Ser116 can form HBs with PBTA-6, PBTA-7 and PBTA-8, as well as Asn255 can form HBs with

A

PBTA-4 and PBTA-6. On the other hand, Leu496 is present at the active center, but does not form HBs with any PBTA. However, their flexible alkyl lateral chain can increase the fluctuations along the MD simulation. Also, Leu496 can establish hydrophobic interactions with the PBTAs, as well as Phe224 that exhibit π-π staking with all PBTAs. Figure 5 show that residues Gly36, Asn185, Ser260 and Lys499 present large RMSF values along the MD simulations. However, they do not form part of the active pocket of CYP1A1, so that these residues can’t establish direct interactions with the PBTAs. Thus, we assume that Gly36, Asn185, Ser260 and Lys499 only have a structural auxiliary supporting role. 7

The Radius of gyration (Rg) is determined to understand the level of compaction of the structure in PBTAs-CYP1A1 complexes. Rg value is defined as the mass weighted root mean square distance of an atoms collection from their common center of mass (Kumar and Purohit, 2014; Kumar et al., 2014; Lavanya et al., 2016). Hence this analysis gives the overall dimensions of the protein. The calculated Rg values over the simulation time scale for all PBTA-CYP1A1 complexes are shown in Figure 6. This parameter is stable during the whole simulation time for all complexes. The highest level of compaction was found for PBTA-4-CYP1A1 complex, and the lower one for PBTA-8-

SC RI PT

CYP1A1 complex. This result matches with the ones obtained employing the RMSF parameter.

The numbers of hydrogen bonds (HBs) formed between the PBTAs and CYP1A1 during the MD simulations are shown in Figure 7. The HBs were identified by measuring their donor-acceptor distances and considering a cutoff of 3.0 Å. As can be extracted from these graphs, the net number

of hydrogen bonds (HB) between PBTAs and CYP1A1 remains quite low. PBTA-4 form the lowest

and PBTA-6 the highest number of HBs during the 100 ns of simulation. Quantitative analyses of

the occupancies obtained from the trajectories of Figure 7 are presented in Table 2. It is noteworthy

U

that water molecules present a stable interaction with all the studied PBTAs. From these trajectories, we determine that the HBs formed at the active center of CYP1A1 were not

N

permanently stable during all the time of simulation. The most stable interactions of PBTA-4 and

A

PBTA-6 are with the Ser122 residue of CYP1A1 and two molecules of water that offer additional stability into the pocket of CYP1A1, see Figure 8. Also, their occupation percentages and HB

M

distances along the simulation are consistent with this model as showed in Table 2. On the contrary, PBTA-7 and PBTA-8 does not form stable hydrogen bonds with the active center of

D

CYP1A1, despite of exhibiting the formation of several HBs during the simulation time, as depicted from the quantitative analysis of their occupancies, see Table 2. This is the main cause of the

TE

difference in the stability of these PBTA-CYP1A1 complexes, which can be associated with the observed experimental genotoxicity (Kumar and Purohit, 2014; Kumar et al., 2014; Lavanya et al., 2016).

EP

The calculated binding free energy of PBTAs using the Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) methodology is presented in Table 1. Each calculated binding free

CC

energy (∆Gbinding) obtained from the snapshots of the MD trajectories were subsequently averaged and decomposed into van der Waals (∆Evdw), electrostatic (∆Eelect.) and solvation (∆Gsolv) contributions for all the PBTA-CYP1A1 complexes. It is important to note that after starting

A

molecular dynamics simulations, a 2.0 ns of equilibration are necessary, since the input structure is not within the equilibrium under the present simulation conditions. Once the system reaches a thermodynamic stationary point, we start to collect data, see Table 1. Thus, the 2.0 ns of equilibration were not considered for the data collection. The major favorable contribution to the total binding free energy corresponds to ∆EvdW . It must be highlighted that the difference of ∆Evdw values among PBTA-4, PBTA-6 and PBTA-8-CYP1A1 complexes is similar, being the ∆Evdw value for PBTA-7-CYP1A1 different to the rest of the series. On the contrary, the electrostatic term destabilizes the PBTA-CYP1A1 complexes. However, the gas phase interaction energy (∆Ggas), 8

that correspond to the sum of ∆Eelect and ∆EvdW contribution terms shows that PBTA-4-CYP1A1 is the less stable complex of the series, since PBTA-4 form the lowest number of hydrogen bonds at the active center of CYP1A1. With respect to ∆Gsolv, Table 1 shows that the polar term (∆GGB) is almost negligible, while the nonpolar interaction (∆GSA) is the major stabilizing contribution of PBTAs on the protein. The resulting binding free energy considering the above results, determines that PBTA-4-CYP1A1 interaction is the most stable of the series and PBTA-6-CYP1A1 the less stable. It can be

SC RI PT

concluded that the substitution of the -NH2 terminal group of PBTA-4, derives into the variation of

the components that determine the binding free energy, and ultimately the observed pro-mutagen activation and mutagenicity trend. These results can be associated with those of metabolic activation and genotoxicity assays (Morisawa et al., 2003; Morris et al., 2009b; Oda et al., 2008; Ohe et al., 2009, 2004; Ono et al., 2000).

4. Conclusions

U

The understanding of the molecular basis of the binding modes of novel 2-phenil-benzotriazoles

N

pro-mutagens with human Cytochrome P450-1A1 is essential to know the mechanism of metabolic activation of these pollutants, which is crucial to exerting its genotoxic power. In the present work,

A

we report a computational protocol to predict the binding mechanism of these pro-mutagens towards Cytochrome P450-1A1. In summary, the protocol consists of three steps: i) molecular

M

docking, to reproduce the bound form of ligands inside the pocket of proteins, ii) molecular dynamics simulations of the molecular structures representing the ligand-protein (PBTA-CYP1A1)

D

complexes using water as solvent, to determine their interaction through time and iii) determine the binding affinity between ligands (PBTAs) and protein (CYP1A1) by means of Molecular Mechanics-

TE

Generalized Born Surface Area (MM-GBSA) free energy calculations, to establish the type of interactions that contribute to the ligand-protein (PBTA-CYP1A1) stabilization. By using this simple

EP

protocol, it can be concluded that the calculations can reproduce the trends experimentally observed for genotoxicity values. The binding energy results of our calculations were consistent with previous experimental metabolic

CC

activation trends, with the highest binding energy corresponding to the PBTA-4-CYP1A1 complex. The studied PBTAs present an adequate orientation in the pocket of cytochrome P450, with - stacking interactions between the triazole group and the Phe224, as well as, hydrogen bonds of the

A

terminal -NH2 of the benzotriazole units with Asn255 and Ser116 amino acids. The behavior of PBTA-CYP1A1 complexes through time was determined employing molecular dynamics simulations. MM-GBSA binding free energy calculations allow us to analyze the role of physical-chemical parameters such as van der Waals interactions, electrostatic interactions, and solvation interactions. It was found that the substitution of the terminal -NH2 groups of PBTA-4, derives into the variation of the components that determine the binding energy of PBTAs to CYP1A1. Since our results agree with those of the metabolic activation and the genotoxicity measured experimentally, 9

we can conclude that our theoretical model can be used to analyze the influence of the molecular interactions with the binding modes of PBTA’s over the active site of CytP450, extensible to other pollutants and enzymes, in order to propose new strategies for their treatment and disposal.

5. Acknowledgements

A

CC

EP

TE

D

M

A

N

U

SC RI PT

The authors acknowledge to FONDECYT-Chile grant 1171118 for their financial support.

10

References Adasme-Carreno, F., Munoz-Gutierrez, C., Caballero, J., Alzate-Morales, J.H., 2014. Performance of the MM/GBSA scoring using a binding site hydrogen bond network-based frame selection: the

protein

kinase

case.

Phys.

Chem.

Chem.

Phys.

16,

14047–14058.

doi:10.1039/C4CP01378F Aleem, A., Malik, A., 2003. Genotoxicity of water extracts from the River Yamuna at Mathura, India.

SC RI PT

Environ. Toxicol. 18, 69–77. doi:10.1002/tox.10101 Azam, S.S., Abbasi, S.W., 2013. Molecular docking studies for the identification of novel

melatoninergic inhibitors for acetylserotonin-O-methyltransferase using different docking routines. Theor. Biol. {&} Med. Model. 10, 63. doi:10.1186/1742-4682-10-63

Bansal, S., Leu, A.N., Gonzalez, F.J., Guengerich, F.P., Chowdhury, A.R., Anandatheerthavarada,

H.K., Avadhani, N.G., 2014. Mitochondrial targeting of cytochrome P450 (CYP) 1B1 and its

role in polycyclic aromatic hydrocarbon-induced mitochondrial dysfunction. J. Biol. Chem. 289, 9936–51. doi:10.1074/jbc.M113.525659

U

Berman, H.., Westbrook, Z., Feng, G., Gilliland, T.., Bhat, H., Weissig, I.., Shindyalov, P, E.,

N

Bourne, P.., 2000. The Protein Data Bank. Nucleic Acids Res. 28, 235–242. Borosky, G.L., 2008. Carcinogenic carbocyclic and heterocyclic aromatic amines: A DFT study their

mutagenic

potency.

J.

Mol.

Graph.

Model.

27,

459–465.

A

concerning

doi:10.1016/j.jmgm.2008.08.002

M

Borosky, G.L., 2007. Ultimate carcinogenic metabolites from aromatic and heterocyclic aromatic amines: A computational study in relation to their mutagenic potency. Chem. Res. Toxicol. 20,

D

171–180. doi:10.1021/tx600278q

Brian N, D., Charles L. Brooks, *, 1999. Development of a Generalized Born Model Parametrization

TE

for Proteins and Nucleic Acids. J. Chem. Phys. B 103, 3765–3773. doi:10.1021/JP984440C Caballero, J., Alzate-Morales, J.H., Vergara-Jaque, A., 2011. Investigation of the differences in

EP

activity between hydroxycycloalkyl N1 substituted pyrazole derivatives as inhibitors of B-Raf kinase by using docking, molecular dynamics, QM/MM, and fragment-based de novo design: study of binding mode of diastereomer . J. Chem. Inf. Model. 51, 2920–2931.

CC

doi:10.1021/ci200306w Cirik, K., Dursun, N., Sahinkaya, E., Çinar, Ö., 2013. Effect of electron donor source on the

A

treatment of Cr(VI)-containing textile wastewater using sulfate-reducing fluidized bed reactors (FBRs). Bioresour. Technol. 133, 414–420. doi:10.1016/j.biortech.2013.01.064

Crites, R.W., Middlebrooks, E.J., Bastian, R.K., 2014. Natural Wastewater Treatment Systems, Second Edition, Civil and Environmental Engineering Series. Taylor & Francis.

Faver, J., Merz, K.M., 2010. Utility of the Hard/Soft Acid−Base Principle via the Fukui Function in Biological Systems. J. Chem. Theory Comput. 6, 548–559. doi:10.1021/ct9005085 Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A., Cheeseman, J.R., Scalmani, G., Barone, V., Mennucci, B., Petersson, G.A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H.P., Izmaylov, A.F., Bloino, J., Zheng, G., Sonnenberg, J.L., Hada, M., Ehara, M., 11

Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Montgomery Jr., J.A., Peralta, J.E., Ogliaro, F., Bearpark, M., Heyd, J.J., Brothers, E., Kudin, K.N., Staroverov, V.N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J.C., Iyengar, S.S., Tomasi, J., Cossi, M., Rega, N., Millam, J.M., Klene, M., Knox, J.E., Cross, J.B., Bakken, V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J., Cammi, R., Pomelli, C., Ochterski, J.W., Martin, R.L., Morokuma, K., Zakrzewski, V.G., Voth, G.A., Salvador, P., Dannenberg, J.J., Dapprich, S.,

SC RI PT

Daniels, A.D., Farkas, Ö., Foresman, J.B., Ortiz, J. V, Cioslowski, J., Fox, D.J., n.d. Gaussian 09 Revision E.01.

Garrido-Baserba, M., Hospido, A., Reif, R., Molinos-Senante, M., Comas, J., Poch, M., 2014. Including the environmental criteria when selecting a wastewater treatment plant. Environ. Model. Softw. 56, 74–82. doi:http://dx.doi.org/10.1016/j.envsoft.2013.11.008

Genheden, S., Ryde, U., 2010. How to obtain statistically converged MM/GBSA results. J. Comput. Chem. 31, 837–846. doi:10.1002/jcc

U

Gohlke, H., Hendlich, M., Klebe, G., 2000. Knowledge-based scoring function to predict proteinligand interactions. J. Mol. Biol. 295, 337–56. doi:10.1006/jmbi.1999.3371

N

Hilal, R., Khalek, A.A.A., Elroby, S.A.K., 2005. DFT investigation of nitrenium ions derived from

A

metabolism of antitumor 2-(4-aminophenyl)benzothiazoles. J. Mol. Struct. THEOCHEM 731, 115–121. doi:10.1016/j.theochem.2005.04.017 Poisson−Boltzmann

Surface

Area

Method.

Mol.

Inform.

31,

114–122.

D

doi:10.1002/minf.201100135

M

Homeyer, N., Gohlke, H., 2012. Free Energy Calculations by the Molecular Mechanics

Janovec, L., Kožurková, M., Sabolová, D., Ungvarský, J., Paul’\iková, H., Plš’\iková, J., Vantová, Z.,

TE

Imrich, J., 2011. Cytotoxic 3,6-bis((imidazolidinone)imino)acridines: synthesis, DNA binding and molecular modeling. Bioorg. Med. Chem. 19, 1790–1801. doi:10.1016/j.bmc.2011.01.012 Kollman, P. a, Massova, I., Reyes, C., Kuhn, B., Huo, S., Chong, L., Lee, M., Lee, T., Duan, Y.,

EP

Wang, W., Donini, O., Cieplak, P., Srinivasan, J., Case, D. a, Cheatham, T.E., 2000. Calculating structures and free energies of complex molecules: combining molecular

CC

mechanics and continuum models. Acc. Chem. Res. 33, 889–97. Kumar, A., Purohit, R., 2014. Use of Long Term Molecular Dynamics Simulation in Predicting Cancer Associated SNPs. PLoS Comput. Biol. 10. doi:10.1371/journal.pcbi.1003318

A

Kumar, K.M., Anbarasu, A., Ramaiah, S., 2014. Molecular docking and molecular dynamics studies on

β-lactamases

and

penicillin

binding

proteins.

Mol.

Biosyst.

10,

891–900.

doi:10.1039/c3mb70537d

Lavanya, P., Ramaiah, S., Anbarasu, A., 2016. A Molecular Docking and Dynamics Study to Screen Potent Anti-Staphylococcal Compounds Against Ceftaroline Resistant MRSA. J. Cell. Biochem. 117, 542–548. doi:10.1002/jcb.25307 Lepers, C., Dergham, M., Armand, L., Billet, S., Verdin, A., Andre, V., Pottier, D., Courcot, D., Shirali, P., Sichel, F., 2014. Mutagenicity and clastogenicity of native airborne particulate 12

matter samples collected under industrial, urban or rural influence. Toxicol. In Vitro 28, 866– 74. doi:10.1016/j.tiv.2014.03.011 Massova, I., Kollman, P., 2000. Combined molecular mechanical and continuum solvent approach (MM-PBSA/GBSA) to predict ligand binding. Perspect. Drug Discov. Des. 18, 113–135. doi:10.1023/A:1008763014207 Mena-Ulecia, K., Hernandez, H., 2015. Decentralized peri-urban wastewater treatment technologies assessment integrating sustainability indicators. Water Sci. Technol. 72, 214–222.

SC RI PT

doi:10.2166/wst.2015.209

Mena-Ulecia, K., Tiznado, W., Caballero, J., 2015. Study of the Differential Activity of Thrombin

Inhibitors Using Docking, QSAR, Molecular Dynamics, and MM-GBSA. PLoS One 10, e0142774. doi:10.1371/journal.pone.0142774

Mena-Ulecia, K., Vergara-Jaque, A., Poblete, H., Tiznado, W., Caballero, J., 2014. Study of the Affinity between the Protein Kinase PKA and Peptide Substrates Derived from Kemptide Using

Molecular

Dynamics

Simulations

and

MM/GBSA.

U

doi:10.1371/journal.pone.0109639

PLoS

One

9,

e109639.

Michalski, R., Ficek, A., 2016. Environmental pollution by chemical substances used in the shale extraction—a

review.

Desalin.

Treat.

57,

1336–1343.

A

doi:10.1080/19443994.2015.1017331

Water

N

gas

Miyamoto, N., Oguro, Y., Takagi, T., Iwata, H., Miki, H., Hori, A., Imamura, S., 2012. Design,

novel

VEGFR2

kinase

inhibitors.

Bioorg.

Med.

Chem.

20,

7051–7058.

D

doi:10.1016/j.bmc.2012.10.004

M

synthesis, and evaluation of imidazo[1,2-b]pyridazine derivatives having a benzamide unit as

Mizuno, T., Takamura-Enya, T., Watanabe, T., Hasei, T., Wakabayashi, K., Ohe, T., 2007.

TE

Quantification of a potent mutagenic 4-amino-3,3’-dichloro-5,4’-dinitrobiphenyl (ADDB) and the related chemicals in water from the Waka River in Wakayama, Japan. Mutat. Res. 630, 112– 121. doi:10.1016/j.mrgentox.2007.03.004

EP

Molinos-Senante, M., Gómez, T., Garrido-Baserba, M., Caballero, R., Sala-Garrido, R., 2014. Assessing the sustainability of small wastewater treatment systems: A composite indicator Sci.

Total

Environ.

497?498,

607–617.

CC

approach.

doi:http://dx.doi.org/10.1016/j.scitotenv.2014.08.026

Morisawa, T., Mizuno, T., Ohe, T., Watanabe, T., Hirayama, T., Nukaya, H., Shiozawa, T., Terao,

A

Y., Sawanishi, H., Wakabayashi, K., 2003. Levels and behavior of 2-phenylbenzotoriazoletype mutagens in the effluent of a sewage treatment plant. Mutat. Res. 534, 123–132.

Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009a. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 30, 2785–2791. doi:10.1002/jcc.21256 Morris, G.M., Huey, R., Lindstrom, W., Sanner, M.F., Belew, R.K., Goodsell, D.S., Olson, A.J., 2009b. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 30, 2785–91. doi:10.1002/jcc.21256 13

Nukaya, H., Shiozawa, T., Tada, A., Terao, Y., Ohe, T., 2001. Identification of 2- [ 2- ( acetylamino ) -4-amino- ( PBTA-4 ) as a potent mutagen in river water in Kyoto and Aichi prefectures , Japan. Mutat. Res. 492, 73–80. Oda, Y., Watanabe, T., Terao, Y., Nukaya, H., Wakabayashi, K., 2008. Genotoxic activation of 2phenylbenzotriazole-type

compounds

by

human

cytochrome

P4501A1

and

N-

acetyltransferase expressed in Salmonella typhimurium umu strains. Mutat. Res. 654, 52–7. doi:10.1016/j.mrgentox.2008.04.013

SC RI PT

Ohe, T., Suzuki, A., Watanabe, T., Hasei, T., Nukaya, H., Totsuka, Y., Wakabayashi, K., 2009.

Induction of SCEs in CHL cells by dichlorobiphenyl derivative water pollutants, 2phenylbenzotriazole (PBTA) congeners and river water concentrates. Mutat. Res. 678, 38–42. doi:10.1016/j.mrgentox.2009.06.003

Ohe, T., Watanabe, T., Nonouchi, Y., Hasei, T., Agou, Y., Tani, M., Wakabayashi, K., 2008.

Identification of a new mutagen, 4,4’-diamino-3,3’-dichloro-5-nitrobiphenyl, in river water

flowing through an industrial area in Wakayama, Japan. Mutat. Res. 655, 28–35.

U

doi:10.1016/j.mrgentox.2008.06.008

567, 109–49. doi:10.1016/j.mrrev.2004.08.003

N

Ohe, T., Watanabe, T., Wakabayashi, K., 2004. Mutagens in surface waters: a review. Mutat. Res.

A

Ono, Y., Somiya, I., Oda, Y., 2000. Identification of a carcinogenic heterocyclic amine in river water. Water Res. 34, 890–894.

M

Onufriev, A., Bashford, D., Case, D.A., 2004. Exploring protein native states and large-scale conformational changes with a modified generalized born model. Proteins Struct. Funct.

D

Bioinforma. 55, 383–394. doi:10.1002/prot.20033 Osibanjo, O., Daso, A.P., Gbadebo, A.M., 2011. The impact of industries on surface water quality of

TE

River Ona and River Alaro in Oluyole Industrial Estate, Ibadan, Nigeria. African J. Biotechnol. 10, 696–702. doi:10.5897/AJB10.1065 Perdew, J.P., Burke, K., Ernzerhof, M., 1996. Generalized Gradient Approximation Made Simple.

EP

Phys. Rev. Lett. 77, 3865–3868. doi:10.1103/PhysRevLett.77.3865 Phillips, J.C., Braun, R., Wang, W., Gumbart, J., Tajkhorshid, E., Villa, E., Chipot, C., Skeel, R.D.,

CC

Kalé, L., Schulten, K., 2005. Scalable molecular dynamics with NAMD. J. Comput. Chem. 26, 1781–802. doi:10.1002/jcc.20289

Popovic, T., Kraslawski, A., Heiduschke, R., Repke, J.-U., 2014. Proceedings of the 8th

A

International Conference on Foundations of Computer-Aided Process Design, Computer Aided Chemical Engineering. Elsevier. doi:10.1016/B978-0-444-63433-7.50105-X

Quesada-Romero, L., Mena-Ulecia, K., Tiznado, W., Caballero, J., 2014. Insights into the interactions between maleimide derivates and GSK3β combining molecular docking and QSAR. PLoS One 9. doi:10.1371/journal.pone.0102212 Suite 2011: Maestro, version 9.2, Schrödinger, LLC, New York, NY, ., 2011. Trott, O., Olson, A.J., 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 31, 455–61. 14

doi:10.1002/jcc.21334 Uno, S., Sakurai, K., Nebert, D.W., Makishima, M., 2014. Protective role of cytochrome P450 1A1 (CYP1A1) against benzo[a]pyrene-induced toxicity in mouse aorta. Toxicology 316, 34–42. doi:10.1016/j.tox.2013.12.005 Walsh, A.A., Szklarz, G.D., Scott, E.E., 2013. Human cytochrome P450 1A1 structure and utility in understanding drug and xenobiotic metabolism. J. Biol. Chem. 288, 12932–12943. doi:10.1074/jbc.M113.452953

SC RI PT

Watanabe, T., Nukaya, H., Terao, Y., Takahashi, Y., Tada, a, Takamura, T., Sawanishi, H., Ohe,

T., Hirayama, T., Sugimura, T., Wakabayashi, K., 2001. Synthesis of 2-phenylbenzotriazoletype mutagens, PBTA-5 and PBTA-6, and their detection in river water from Japan. Mutat. Res. 498, 107–15.

Watanabe, T., Ohba, H., Asanoma, M., Hasei, T., Takamura, T., Terao, Y., Shiozawa, T., Hirayama,

T., Wakabayashi, K., Nukaya, H., 2006. Isolation and identification of non-chlorinated phenylbenzotriazole (non-ClPBTA)-type mutagens in the Ho River in Shizuoka Prefecture,

U

Japan. Mutat. Res. 609, 137–45. doi:10.1016/j.mrgentox.2006.06.033

Yuriev, E., Ramsland, P. a, 2013. Latest developments in molecular docking: 2010-2011 in review.

N

J. Mol. Recognit. 26, 215–39. doi:10.1002/jmr.2266

A

Zoete, V., Cuendet, M.A., Grosdidier, A., Michielin, O., 2011. SwissParam: A fast force field generation tool for small organic molecules. J. Comput. Chem. 32, 2359–2368.

A

CC

EP

TE

D

M

doi:10.1002/jcc.21816

15

Figure Captions

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 1: Chemical structures of PBTA-type mutagens studied.

16

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 2: Alignment of all docked PBTAs in complex with CYP1A1, denoting - stacking interactions between the triazole group and Phe224 aminoacid and amino group eith Ser116 and Asn255 aminoacids of CYP1A1.

17

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 3: Interactions of group heme of the (A) PBTA-4, (B) PBTA-6, (C) PBTA-7 and (D) PBTA-8.

18

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 4: Plots of Root Means Squared Deviation (RMSD) values in the simulation times corresponding to molecular dynamics of all the systems under study. (A) RMSD during 100 ns of molecular dynamics simulation. (B) Average of RMSD for all systems.

19

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 5: Root mean square fluctuation of the backbone atoms of the PBTAsCYP1A1 complexes during 100 ns of molecular dynamics simulations at 298.15 K.

20

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 6: Radius of gyration of Cα atoms of the PBTAs-CYP1A1 complexes during 100 ns of molecular dynamics simulation at 298.15 K.

21

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 7: Number of hydrogen bonds between the CYP1A1 and PBTAs during 100 ns of molecular dynamics simulation.

22

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure 8: Principal interactions of PBTAs observed during molecular dynamics simulations. (A) PBTA-4, (B) PBTA-6, (C) PBTA-7 and (D) PBTA-8.

23

Table 1. Experimental Genotoxicity data for PBTAs; calculated binding energies of the first ranked

Autodock Vina pose and calculated binding free energies and their individual components from MD simulations through the MM-GBSA protocol for PBTA-CYP1A1 complexes. MM-GBSA from MD simulations

Energy (Rank) (kcal/mol)

RMSD (Å)

∆Eele

∆EvdW

-9.7 (1)

0.845

21.55 ± 0.22

-52.81 ± 0.18

PBTA-6

32

485000

-8.8 (4)

1.301

18.03 ± 0.09

-50.11 ± 0.27

PBTA-7

247

1430000

-9.4 (2)

PBTA-8

152

2213000

-9.3 (3)

Ligand

a

SC RI PT

Docking Experiment

PBTA-4

Experimental Genotoxicity Promutagen Activation Mutagenicity (umu units·min-1·nmol (revertant/μg)b P450-1)a 279 7800000

1.224

25.29 ± 0.23

-57.78 ± 0.15

0.986

16.20 ± 0.08

-50.43 ± 0.31

PBTA by human cytochrome P450 preparations (Oda et al., 2008). b Experimental mutagenicity in Salmonella typhimurium

A

CC

EP

TE

D

M

A

N

U

(Ames test) (Nukaya et al., 2001; Ohe et al., 2004; Watanabe et al., 2001)

24

Table 2. Hydrogen bonds occupancy analysis from MD simulations for PBTA-CYP1A1 complexes. PBTA-4 Donnor-Aceptor Ser122-H···PBTA Thr321-H···PBTA PBTA···Ser116 PBTA···Ala317 H-O···PBTA

a

Dist (Å)b 1.70 5.31 4.08

54.53

2.38

51.70

2.46

Occ. (%)a 90.91 52.64 0.10 54.30 62.11 55.55

PBTA-7 Dist (Å)b 1.80 2.96 3.35 3.48 2.39 2,43

PBTA-8

Occ. (%)a 0.01 8.42 41.47

Dist (Å)b 4.36 4.58 3.37

Occ. (%)a 18.14 0.01 29.09

Dist (Å)b 2.94 4.60 1.96

95.52

1.80

87.28

1.75

49.72

2.10

17.77

2.47

SC RI PT

PBTA···O-H

PBTA-6

Occ. (%)a 92.46 0.01 1.54

The occupancies reflect the % of the time that the hydrogen bond exists with respect to the whole time.

b

The averaged

between hydrogen-acceptor and hydrogen-donor heavy atoms during the time that the hydrogen bond is formed.

References

Nukaya, H., Shiozawa, T., Tada, A., Terao, Y., Ohe, T., 2001. Identification of 2- [ 2- ( acetylamino ) -4-amino- ( PBTA-4 ) as a potent mutagen in river water in Kyoto and Aichi prefectures , Japan. Mutat. Res. 492, 73–80.

N

U

Oda, Y., Watanabe, T., Terao, Y., Nukaya, H., Wakabayashi, K., 2008. Genotoxic activation of 2-phenylbenzotriazole-type compounds by human cytochrome P4501A1 and N-acetyltransferase expressed in Salmonella typhimurium umu strains. Mutat. Res. 654, 52–57. doi:10.1016/j.mrgentox.2008.04.013

A

Ohe, T., Watanabe, T., Wakabayashi, K., 2004. Mutagens in surface waters: a review. Mutat. Res. 567, 109–49. doi:10.1016/j.mrrev.2004.08.003

A

CC

EP

TE

D

M

Watanabe, T., Nukaya, H., Terao, Y., Takahashi, Y., Tada, a, Takamura, T., Sawanishi, H., Ohe, T., Hirayama, T., Sugimura, T., Wakabayashi, K., 2001. Synthesis of 2-phenylbenzotriazole-type mutagens, PBTA-5 and PBTA-6, and their detection in river water from Japan. Mutat. Res. 498, 107–15.

25