Tobacco Smoke Carcinogens Induce DNA Repair Machinery Function Loss: Protection by Carbon Nanotubes

Purpose: DNA damage is a continuous process occurring within the cells caused by intrinsic and extrinsic factors, but it gets repaired regularly. If the DNA repair process is faulty, the incidences of damages/mutations can accumulate in cells resulting in cell transformation. It is hypothesized that the negative variations in DNA repair pathways in even at one point viz. genetic, translational or posttranslational stage may fairly be crucial for the beginning and development of carcinogenesis. Therefore, we investigated the potential of tobacco specific nitrosamines (TSNs) related carcinogens to interact with the enzymes involved in DNA repair mechanisms in the current study. Methods: The derivatives of cigarettes’ smoke like NNK and NNAL are very well known and recognized carcinogens. Therefore, almost 120 enzymes playing crucial role in the DNA repair process have been analysed for their reactivity with NNK and NNAL. Results: The molecular docking study helped to screen out, 07 possible DNA repair enzyme targets for NNK, and 12for NNAL. Present study revealed the loss of activity of DNA repair enzymes in the presence of NNK and NNAL, and this accumulation may induce the tendency of DNA damage which can lead the transformation of exposed normal cells in to cancerous cells. This study also demonstrated the protective potential of nanoparticles like SWCNTs/MWCNTs against TSN’s induced toxicity; here SWCNT against NNK (-17.16 Kcal/Mol) and MWCNT against NNK -17.01 Kcal/Mol were showing maximum binding affinities than the known biomolecular target of NNK 1UGH (Uracil-DNA glycosylase,-7.82Kcal/Mol). Conclusion: CNTs can be applied as chemo-preventive agents against environmental and tobacco induced carcinogens owing to their scavenging potential and warrants for in vivo and in vitro experimental validation of the results obtained from the present study.

Introduction 1-ol) make non repairable DNA lesions (Huang et al., 2011), and were therefore analysed for their actions on important enzymes in DNA repair machinery (Saito et al., 2012). NNK, a derivative of tobacco alkaloids ,is one of the established strong carcinogen in humans (Hecht and Hoffmann, 1989).
The present study involves simulation procedures to explore the binding efficiency of NNK and NNAL with the enzymes important in DNA repair processes. There are almost 120 enzymes engaged in DNA repair mechanisms . Hence, in this study we planned to explore if chemicals like NNK and its metabolite NNAL, despite straight damaging the DNA, are affecting the DNA repair process also by interacting with these enzymes?
Lately, many nanoparticles have shown their scavenging effects for the reclamation of soil contaminated with heavy PAHs(polycyclic aromatic hydrocarbons) (Karnchanasest and Santisukkasaem, 2007). The rummaging abilities of nanoparticles towards PAHs and other toxicants can be attributed to surface chemistry, large and complimentary surface area and other inherent assets. Titanate Nano Tubes (TNT) and Titanate Nano Sheets (TNS) were employed as additives to get rid of detrimental chemicals from cigarette smoke (Qixin et al., 2011), and were found to successfully remove tar, hydrogen cyanide, nicotine, ammonia, phenolic compounds and selected carbonyls. Single and multi-walled Carbon nanotubes (CNTs) are gaining lot of attention from the scientific community owing to their extraordinary physical and atomic properties. The void one-dimensional and oneatom-thick structure of SWCNTs and more than oneatom-thick structure of MWCNTs are particularly great for adsorption associated applications. CNTs are widely used nowadays for creating sensors of pollutant smokes. Also, CNTs have been used for eliminating dangerous pollutants from gas streams because of their adsorption potentials (Agnihotri et al., 2005).
In view of the above mentioned characteristics of carbon nanotubes, the current study has been deliberated to investigate if CNTs could be exploited to defend the biological system against the harmful effects of TSNs exposure. The current study was performed using various bioinformatic tools for in silico experimentations to achieve a deep insight of the mechanism of protection.

Materials and Methods
Schematic flowchart of methodology is discussed in the Figure 1.

Preparation of protein structures
The structures of all enzymes involved in DNA repair process, used in the current study, were acquired from Protein Data Bank RCSB. Those structures of enzymes, which were not found in there, were obtained from I-Tasser and Homology Modelling-Modeler 9.10 online server (Eswar et al., 2006;Roy et al., 2010). The modelled protein structures had already been mentioned in the study of Dhasmana et al., (2015). The published structures were amended to eliminate HETATM with help of Discovery Studio Visualizer. The energy was minimized, steric collision was removed with the sharpest descent steps 1000, sharpest descent size 0.02 Å, conjugated gradient steps 1000 and the conjugate gradient step size 0.02 Å for the conjugate gradient minimization with help of Chimera (Wang et al., 2004;.

Procurement of carbon based nanotube and fullerene
The designing and procurement of .pdb files of carbon-based nanoparticle ( Figure 4) was done using Nanotube Modeler software program (Firdaus et al., 2017). The parameters of the designed single walled carbon nanotubes (SWCNTs) were bond length 1.421nm, tube length 2.5nm, whereas the parameters of multi walled carbon nanotubes (MWCNTs) were with bond length 1.421 nm, two wall layers and tube length 2.5 nm, while the parameter of fullerene was with diameter of 1nm.

Molecular docking studies
The docking evaluations were executed by using AutoDock 4.0 suite based on genetic algorithm (GA) (Morris et al., 2009;Rarey et al., 1996). The Cygwin interface was used with 10 times run of each docking, in Intel (R) i7-5500U, CPU 2.40 GHz and 16.0 GB of RAM of DELL machine. The molecular docking method was implemented followed by searching the best conformation of DNA repair enzymes, both CNTs, fullerene, NNK and NNAL complexes on the basis of molecular binding energy. All AutoDock 4.0 parameters were used for this study have been discussed in our previously published article . For the characterization of the docked binding sites of Protein-Ligands complex, the LigSite online server was used (Hendlich et al., 1997) ( Table 1). The most favourable functional partners of enzymes involved in DNA repair process were obtained with help of STRING 9.0. The known database (http:// string-db.org/) for predicting protein-protein interactions was used for predicting the interaction of the query molecules. The ZDOCK protein-protein docking program (Fast Fourier Transform algorithm) was used for protein-protein interaction (Chen et al., 2003). The best ZDOCK score of protein-protein interaction and protein complex (protein+ NNK or NNAL) interaction were considered for the most suitable protein conformational pose , here protein complex (protein+ NNK or NNAL) interaction analysis will be used to explore the inhibition tendency of NNK/NNAL between two DNA repair enzymes and how these carcinogens interfering between normal protein-protein interaction as shown. For NNAL may induce carcinogenesis was elucidated by implementing protein-protein docking technique. The function loss of the enzymes was evaluated by using ZDOCK scores. The ZDOCK scores of protein-protein protective potential analysis AutoDock 4.0 tool was used for the docking analysis of nanoparticles (SWCNT, MWCNT and Fullerene) against NNK and NNAL.

Molecular docking studies
The adverse effects of NNK and NNAL on the DNA repair process enzymes were analysed after procuring all 120 PDB files of DNA repair enzymes as reported by Dhasmana et al., (2015). These enzymes of DNA repair mechanism were docked with tobacco smoke carcinogens (NNK, and its metabolites NNAL). The interaction of these enzymes with NNK and NNAL is comprehensively scrutinized in current article. NNK and NNAL have shown the binding capability with all 120 enzymes involved in DNA repair ranged from +28.15 to -7.82 Kcal/Mol and 7.81 Kcal/Mol, respectively. Docking simulations were achieved and hydrogen bonds were constructed in all the docked assemblies. The 1IRD(RCSB ID of human carbon monoxy haemoglobin) was used as a positive control , to select and authenticate the most plausible biomolecular targets among the 120 DNA repair enzymes. After setting the threshold energy of positive control, the complete docking involved was screened for the selection of most probable biomolecular targets. The NNK and NNAL bind with positive control 1IRD with -6.67 and -6.30 Kcal/Mol of energy, respectively).After finishing the screening of whole docking process, the most plausible biomolecular targets of NNK and NNAL were selected. The selection was based on the interactions among 120 DNA repair enzymes, which have binding energy either equal or higher than the positive control. The NNK and NNAL binding efficiency with their probable bimolecular targets ranges from -6.49 to -7.82 Kcal/Mol and -6.79 to -7.81 Kcal/Mol, respectively. Hydrogen bond sizes and inhibition contest (Ki) were measured by using Discovery Studio Visualizer and docked .dlg file of AutoDock 4.0 respectively as mentioned in Table 1.

Protein functional loss analysis by using protein-protein docking
The probable mechanism how NNK and its metabolite           (Table 4).

Discussion
Smoking and consuming tobacco causes oral/lung cancer, is almost an established fact (Keyler et al., 2003). The consumers of tobacco are generally in contact with Tobacco Specific Nitrosamines (TSNs) like NNK and its derivative NNAL present in cigarette smoke. The NNK interacts with DNA and forms DNA adduct, accountable for the tumor development (Gharavi and Kadi, 2004). The current study attempts to explicate the process of carcinogenesis induced by cigarette smoke carcinogens NNK and NNAL by using molecular simulation techniques. The study included120 enzymes that participate in various DNA repair processes. The pdb files for all 120 enzymes were collected from the previous study conducted by Dhasmana et al.,(2015). To determine the potential biomolecular targets of TSNs, NNK and NNAL were docked with each of the 120 DNA repair enzymes. The enzymes, binding to TSNs with a score higher than the docking score of TSNs with positive control (1IRD), were considered as the potential biomolecular targets of TSNs. The assessment projected 7 potential targets of NNK, and 12 of NNAL as mentioned in the   Figure 5. Protein-Protein Interaction of HUS1with Their Functional Partner RAD9,ZDOCK Score was Reduced from 14.06 to 12.7 and NNK Binds between the Interface ofHUS1 and RAD9. Orange and yellow coloured ribbon like structures represents the RAD9 andHUS1 proteins respectively and red and green coloured ball like structures shows binding residues of RAD9 and HUS1, which involve an intermolecular hydrogen bond. NNK is interfering between the interfaces of HUS1 andRAD9. were considered to be most vulnerable targets, namely, HUS1, SMUG1, Uracil-DNA glycosylase (1UGH), AlkB homolog 2 (3BTY), DNA mismatch repair protein Msh3 (3THW), POLG1 and DPOLA. Further, these targets were examined for the loss of molecular propertiesas a result of TSNs binding, by using an in silico ZDOCK server. The score of PPI (Protein-protein interaction) with each of the 12 probable target DNA repair proteins and their utmost standard functional partner (carefully chosen from the STRING Database) was estimated, and compared with protein complex's (target DNA repair enzymes+TSNs) interaction with their respective potential functional partner. The results as shown by the current study clearly indicates the loss of function of 7 important DNA repair proteins ( Table 2).
The greatest decrease in ZDOCK score was found in between HUS1 and RAD9 from 14.06 to 12.7 (NNK bound HUS1 with RAD9 as shown in Figure 5). RAD9, HUS1, and RAD1 heterotrimeric complex performs dual role in checkpoint activation of cell cycle and repair of damaged DNA in human cells (Min et al., 2009), but NNK interference may acts as an obstacle for this hetromeric complex formation and can obstructs the cell cycle check point activation and hence reduce DNA repair tendency.
The reduction in ZDOCK score was also found between Uracil-DNA glycosylase & PCNA. It was reduced from 14.16 to 13.64 (NNK bound Uracil-DNA glycosylase with PCNA). This uracil-DNA glycosylase acts as an essential DNA repair protein to eliminate uracil bases from DNA. Earlier, Kiyonari et al., (2008), reported that human nuclear Uracil-DNA glycosylase forms a complex with other proteins and starts the base excision repair by acting with proliferating cell nuclear antigen (PCNA). But, our study shows that NNK may induce interference between Uracil-DNA glycosylase and PCNA function leading to interrupt the base excision pathways.
The normal binding of MUS81 with EME1 plays an important role in upholding genomic integrity while DNA replication occurs in recombination repair process (Taylor and McGowan, 2008). The human MUS81-EME1 complex is a flap/fork endonuclease which processes stalled replication fork intermediates (Alberto et al., 2003). This study revealed that NNAL causes the interference between MUS81 and EME1 and reduces their ZDOCK score from 16.68 to 14.08.
Thus, the results discussed above show that TSNs may reduce the normal binding and interfere in the normal functioning of DNA repair mechanisms. This impediment in DNA repair pathway can cause damage accumulation and finally cell transformation. Therefore, the current study suggests that TSNs (NNK and NNAL) may alter the mechanisms of DNA repair pathways and enzymes execution, along with these TSNs directly damaging the DNA. These carcinogens from tobacco (like NNK) also acts as a site-selective high affinity agonist for the α7 nicotinic acetylcholine receptor or nAch (-5.47 Kcal/Mol), which controls the development of a significant small cell lung carcinoma by stimulating the secretion of serotonin. Signalling pathways/events are initiated after binding of NNK to α7 nicotinic acetylcholine receptor. The binding of NNK to its receptor, activates Raf -1 /MAP Kinase pathway, which causes phosphorylation of c-myc (Jull et al., 2001).
Following the 'dose makes poison' theory of Paracelsus (the father of toxicology), the low dose or concentration of substances toxic at high dose, can be used as drugs or at least as delivery vehicles. The same can hold true for CNTs (SWCNTs/MWCNTs), which at low doses, can be used for the reduction of hazardous effect of TSNs on DNA repair enzymes. The adsorption of TSNs on both the CNTs is a contribution from different adsorption sites. The strong van der waals interactions of CNTs may play an important role in scavenging NNK and NNAL from the biological system and can help to reduce the impairment of the function of DNA repair pathway enzymes ( Figure 6 and Figure 7). On the other hand, fullerenes were unable to bind NNK and NNAL probably because of low surface area and atomic arrangement (Figure 8). Fullerene was also found to be poor adsorbent for analytical application in contrast to convenience adsorbents (Vergari, 2008). Contrary to fullerene, CNTs were good adsorbents. The current study also showed using in silico experiments that fullerene does not have comparatively good energy than CNTs. The shape and size of carbon nanotubes can vary in diameter, length, number of atoms and their molar weights. They are known for the adsorption of amines like tobacco specific nitrosamine (TSNs ex, NNK and NNAL). The adsorption properties of CNTs at equilibrium deprotonated by the Lewis-basic oxygen of TSNs reveals that free amine base can adsorb to SWCNTs. Hence, even at high concentrations of TSNs, there are high chances that free CNT surface would be available for the nonspecific amine-CNT adsorption (Yoosefian, 2018). TSNs adsorption on nanoparticle surface could be an effect of CNTs to minimize its high free surface energy, to accomplish the atomic coordination at the surface and to establish electronic neutrality.
The possible cause for the reduced toxic effect of co-exposure may be the straight adsorption of carcinogens onto nanoparticles which renders the carcinogens unavailable to its target bio-molecules. Though, the mentioned reason of the adsorption of carcinogens to nanoparticles needs further analysis to prove the point more strongly. The scavenging potential of both the CNTs (SWCNTs and MWCNTs) can thwart the genotoxic effects of TSNs and can be applied as a chemo-preventive agent against tobacco induced carcinogenesis. And also, computer based structural analysis of biomacromolecules and CNTs can serve as a strong basis for cancer management.
In conclusion, the current analysis demonstrated the impairment of the functions of DNA repair enzymes following TSNs exposure, which can be considered as a possible cause for hampering of DNA repair mechanisms. This may result in damage build-up and finally leading to the development of cancer. The current study also proposes that carcinogens (NNK and NNAL) surely affect the process of DNA repair pathways by meddling with enzymes involved. This hazardous effect of carcinogens on DNA repair enzymes can be reduced using both CNTs. The overall adsorption on both CNTs is influenced by their adsorption sites. Due to strong van der waals interactions CNTs can scavenge the NNK & NNAL from the biological system which may further reduce the impairment of the function of DNA repair enzymes. In contrast, fullerenes were unable to bind NNK and NNAL. Their inabilities to bind NNK and NNAL may be because of low surface area. The scavenging potential of both CNTs can avert the genotoxicity posed by TSNs and therefore are proposed as chemo-preventive agent against environmental and tobacco induced carcinogenesis. This study warrants to conduct a profound analysis to describe the failure of DNA repair processes and further look into the scavenging properties of both CNTs with best suitable tools and techniques. It also suggests in vivo and in vitro experimental confirmation to validate the in-silico results acquired from the present study.