The generation of ROS by exposure to trihalomethanes promotes the IκBα/NF-kB/p65 complex dissociation in human lung fibroblast

1. Introduction

Under certain pathological conditions, the excessive generation of reactive oxygen species (ROS) and the subsequent cellular oxidative stress exerted by these damages are unavoidable unless an antioxidant system [1,2] counteracts this oxidative damage. If the cell is inefficient in inducing or regulating the antioxidant protective mechanisms, survival or cell death pathways may be activated to arrest or eliminate the damaged tissue [3,4]. This raises essential inquiries about the molecular mechanisms involved in shifting from pro-survival signaling to pro-cell death signaling due to alterations in oxidative balance. To date, no consensus has been reached on the role of ROS in pro-survival signaling and or cell proliferation of human lung fibroblasts; however, some human fibroblasts, such as those of the skin and palmar fascia, have been shown to proliferate at low ROS, concentrations. Even more, when ROS concentrations were inhibited in cell cultures, these cells' proliferation was also diminished [5].

On the other hand, a diversity of oxidants that occurs in the environment induces lesions, leading to pulmonary fibrosis. These oxidants can also activate genes related to cell growth, cell death, and fibroblasts proliferation [6]. For example, after lung tissue damage, the fibroblasts accumulate excessive amounts of collagen, generally accompanied by the hyper-proliferation of these cells for lung repair [7]. The increase in lung collagen and its fibroblasts induction are associated with pulmonary fibrosis development [8]. In addition to direct damage to lung cells and extracellular matrix, oxidants also contribute to developing pulmonary fibrosis - by affecting cytokines and growth factors. In this regard, nuclear factor-kappa B (NF-κB) activation is involved in pulmonary inflammation pathogenesis [9]. The NF-κB/ proto-oncogene c-Rel family (NF-κB/Rel) plays a role in immune responses such as inflammation and immunity, differentiation, cell growth, apoptosis, and tumorigenesis. In mammals, the NF-κB/Rel family comprises at least five family members: RelA, RelB, NF-κB (p105/p50), NF-κB (p100/p52), and c-Rel [10,11,12,13]. Previous studies have reported that inhibitor of IkappaB kinase isoform α (IκBα) forms a complex with NF-kβ, in which multiple residues dispersed throughout the protein interface collectively determine its binding affinity. Besides, p65-nuclear localization signaling polypeptide (NLS) is the significant component of the complex that contributes to the binding affinity and specificity of IκBα [14]. NF-κB–activating agents -induce - IκB phosphorylation triggers - the release of NF-κB - into the nucleus and regulate the gene expression on the κB sites [10,15,16,17]. However, whether ROS can alter the signaling pathways mediated by NF-κB, possibly involved with hyper-proliferation and other processes in human lung fibroblasts, is not known. This response can occur because phosphorylation is critical in activating NF-κB downstream of physical, oxidative, and genotoxic stress responses, among other stimuli [17].

The interest in studying the role of ROS and biomarkers of the oxidative stress response on activation of NF-κB from our previous study showed that HMs induced hyper-proliferation in human lung fibroblasts accompanied by increases in collagen synthesis [8]. These results have suggested that activation of genes involved in cell adhesion, differentiation, and proliferation was mediated by NF-κB [18]. Therefore, the objective of the current study was to find evidence that relates to the generation of oxidative stress during the production of ROS with the activation of NF-κB in human lung fibroblasts (MRC-5, ATCC cell line) when exposed to chlorinated and brominated HMs such as CH₂Cl₂, trichloromethane (CHCl₃, TCM) and BrCHCl₂.

2. Materials & Methods

2.1. Culture of Human Lung MRC-5 Fibroblasts and Treatments

Human lung MRC-5 fibroblasts were acquired from ATCC. MRC-5 fibroblasts was cultured in Dulbecco's Modified Eagle Medium and harvested according to a previous study [8]. A volume of 100 μL of cell suspension (40,000 cells/mL) was seeded per well in sterile 96-well flat-bottom plates with the culture medium in 5% CO₂ and 37 ºC and allowed to rest for 48 h for cellular adherence and initial growth before treatment was initiated. Before this, growth kinetics using 10,000, 20,000, 30,000, 50,000, and 100,000 cells/mL was performed at 24, 48, and 72 h to optimize growth conditions and reach 50% confluence before exposure (results not shown) [8]. The HMs used in this study were selected based on the frequency of their occurrence and reported concentrations in drinking water [8,19]. BrCHCl₂, CHCl₃, and CH₂Cl₂ high-performance liquid chromatography grade were added to the culture medium using autoclave-sterilized 70% glycerol as a vehicle to obtain final concentrations of 10^-4 to 10^-20 mol (9 treatments) in addition to absolute control (culture medium + 10% fetal bovine serum "FBS") and solvent control (culture medium + FBS + glycerol). Treated cells were incubated for 48 h, and the culture medium and each HM concentration were replaced at 24 h. Three independent experiments were performed each time, and vehicle concentration was 1% in all cases. After the treatment and detachment of the cells with trypsin/EDTA solution, cells from 12 wells of the same concentration were collected, mixed, and washed in Eppendorf tubes with phosphate-buffered saline (PBS1X) preheated to 37ºC solution. Samples were centrifuged at 3,200 rpm for 10 min, and the pellet was placed in 100 µL of mammalian protein extraction reagent (M-PER™) lysis solution (Thermo Scientific), sonicated during 30 seg adding 900 µL of PBS 1X. Sample homogenates were divided into two portions: one was centrifuged at 9,000 x g and 4° C for 15 min to obtain the S9 fraction. The uncentrifuged portion and the S9 fraction were stored at −80 °C until the biomarker assay (less than two weeks). The S9 fraction was used for ROS quantification and the enzymatic assays; the uncentrifuged fraction was used to evaluate lipid peroxidation.

2.2. Evaluation of Biomarkers

Quantification of ROS (O₂^• and H₂O₂) and Level of Lipid Peroxidation as Thiobarbituric Acid Reactive Substances (TBARS)

The ROS measurement was performed in previous reports with 20 μL of S9 fraction of the treated and control cells using a final concentration of 7.0 μmol of dihydroethidium and dihydro-fluorescein diacetate. The concentration of H₂O₂ was calculated with a calibration curve from 0 to 7.0 mol. For O₂•, a molar extinction coefficient of 4,669 mol^-1cm^-1 was used for its quantification, considering a light path of 0.67 cm in 96-well plates with 200 μL of final volume. The results were expressed as mol ROS/40,000 cells. The lipid peroxidation evaluated as TBARS was assessed in the uncentrifuged fractions of the treated and control cells by Buege and Aust's method [21]. The lipid peroxidation results were expressed as TBARS with a molar extinction coefficient of 156,000 mol⁻¹cm⁻¹ as mol of malondialdehyde (MDA)/40,000 cells. Regarding prooxidant forces (ROS) and lipid peroxidation levels, results were shown by cell number because these biomarkers are not related to protein concentration.

2.3. Activity of the Antioxidant Defenses (SOD, CAT and GPx)

The activity of the antioxidant defense enzymes was evaluated in the S9 fraction of the treated and control cells. Superoxide dismutase (SOD; EC 1.15.1.1) activity was evaluated by Misra & Fridovich protocol - using a standard curve (3.73–18.65 UI) of SOD from pure bovine erythrocytes. The results were expressed as mmol/min/mg protein. Catalase (CAT; EC 1.11.1.6) was assessed using the Radi et al. test and was estimated using the molar extinction coefficient of H₂O₂ (0.043 mmol⁻¹cm⁻¹). The obtained results were shown as mmol/min/mg protein. The activity of glutathione peroxidase (GPx; EC 1.11.1.9) was evaluated according to Lei et al. method using a test coupled with 1.0 UI of glutathione reductase, 0.03 mol of nicotinamide adenine dinucleotide (NADH), 3.5 mmol of glutathione and 40 mmol H₂O₂ followed by a reading at 340 nm. Enzyme activity was calculated using the molar extinction coefficient of NADH (6.22 mmol⁻¹ cm⁻¹). The results were expressed as mmol/min/mg protein. All tests were performed in triplicate, and the total protein content was determined at 660 nm using a protein assay kit.

2.4. Assay of Phosphorylated NF-κB/p65 Levels

The sandwich ELISA kit, which detects levels of -NF-κB/p65 phosphorylation at serine 536 - was employed. For the assessment, 100 μL of S9 fraction was placed following the manufacturer's instructions.

2.5. Molecular Docking Analysis

Docking studies against the NF-βB/p65 were performed using a docking module implemented in MOE2019 (Chemical Computing Group, Montreal, Canada). In silico docking in MOE, the ligand placement method "Alpha PMI" was employed using London dG scoring function and implicit generalized born solvation model [25]. The binding sites of the NF-βB/p65 subunit (Protein Data Bank ID: 6QHL) were estimated by using the "alpha site finder" implemented in MOE2014. By default, thirty docked conformations were retained as a cutoff. Finally, the affinity scoring function, ΔG [U total in kcal/mol], was employed to rank the candidate poses as the sum of the electrostatic and Van der Waals energies. The "reaction model" dielectric function with a cutoff between 8 and 10Å and the radius of the interaction site of 6Å was used for our docking studies. The protein-ligand complexes were set up for molecular dynamics and equilibration methods using NAMD 2.5 software and CHARMM22 force field for proteins [27], along with the TIP3P model for water [28]. The simulations began with a 10,000-step minimization of the designed side chains and solvent to remove wrong contacts. A cutoff of 12 A˚ (switching function starting at 10 A˚) for van der Waals interactions was assumed. An integration time step of 2 fs was used, permitting a multiple-time stepping algorithm [29,30] to be employed, in which interactions involving covalent bonds were computed every time step. Short-range non-bonded interactions were computed every two-time step, and long-range electrostatic forces were computed every four-time steps. The pair list of the non-bonded interaction was recalculated every 10-time steps with a pair list distance of 13.5 A˚. The short-range non-bonded interactions were defined as van der Waals and electrostatics interactions between particles within 12 A˚. A smoothing function was employed for the van der Waals interactions at 10A˚ distance. The backbone atoms were harmonically constrained with a restraining constant of 10.0 kcal/mol A˚2, and the systems were heated to 300 K throughout 6 ps at constant volume. The simulations were equilibrated for 2 ns with NPT ensemble (1 atm, 300 K) while the harmonic constraints were gradually turned off. With no harmonic constraints, the simulations ran for 2 ns in the NPT ensemble using Langevin dynamics at a temperature of 300 K with a damping coefficient of g= 5ps-1 [31]. The pressure was maintained at 1 atm using the Langevin piston method with a piston period of 100 fs, a damping time constant of 50 fs, and a piston temperature of 300 K. Nonbonded interactions were smoothly switched off from 10 to 12 A˚. The list of nonbonded interactions was truncated at 14 A˚. Covalent bonds involving hydrogen were held rigid using the SHAKE algorithm, allowing a 2-fs time step. No periodic boundary conditions were included in the above studies. Atomic coordinates were saved every 1 ps for the trajectory analysis during the last 2 ns of MD simulation. CHARMM22 force field parameters were used in all simulations in this study. Finally, the graph was drawn by taking Root Mean Square Deviation (RMSD) on the X-axis with time (ns) on the Y-axis. The structure with the least RMSD of Ca trace in the trajectory generated was used for further studies. Finally, self-consistent field (SCF) energies are calculated using wavefunction-based methods that construct an initial guess density matrix by atomic orbital basis functions. At each step, the kinetic energy, nuclear attraction, and electron-electron repulsion terms are calculated, and a correction is applied to the density until self-consistency is achieved.

2.6 .Statistical Analysis

Results regarding solvent controls were compared using a one-way analysis of variance (ANOVA) followed by Dunnett's test. For phospho- NF-κB/p65, one-way ANOVA followed by the Kruskal–Wallis test was used. Relations between biomarkers were assessed by Pearson correlation analysis. Results were considered significant at p≤0.05.

3. Results

3.1. The Generation of Hydrogen Peroxide (H₂O₂), Superoxide Anion (O₂^•), and Lipid Peroxidation by Exposure to Halomethanes (CH₂Cl_2, CHCl₃, BrCHCl₂) in Human Lung Fibroblasts (MRC-5)

The results showed that the HMs understudy could induce ROS in human lung fibroblasts exposed to different concentrations of these toxicants, particularly superoxide anion. CHCl₃ induced higher and more significant ROS production in all test concentrations; in contrast, CH₂Cl₂ and BrCHCl₂ elicited different concentrations of ROS production -. Only the lower (10^-20 to 10^-14 mol; 1.76-1.66-fold-higher) and the three-fold higher concentrations of CH₂Cl₂ significantly augmented the H₂O₂ production in MRC-5 cells (p≤0.05), among 2.82 to 1.72-fold-higher, respectively. In the case of BrCHCl2, the superoxide anion levels showed an irregular response; however, at the two-fold higher concentrations of this halomethane, a significant diminution (p≤0.001) of H₂O₂ below 0.17-fold-lower was noted (Figure 1A,B).

The increase of lipid peroxidation was significantly augmented by exposure to high concentrations of CHCl3 (10-10 to10-6 mol) with increments between 10 to 15.6-fold-higher, and with BrCHCl₂ at 10^-10 mol (1.87- fold-higher). However, higher levels (10^-8 and 10^-6 mol) of BrCHCl₂ elicited a significant diminution near to 0.12-0.78-fold lower in TBARS. In contrast, an insignificant change of lipid peroxidation was observed in MRC-5 cells exposed to CH₂Cl₂ except for diminishing the oxidative damage close to 0.15-fold-lower in lung fibroblasts exposed to 10^-6 mol (Figure 2).

3.2. The activity of the Antioxidant Enzymes

There were no significant changes in the activity of SOD in the MRC-5 cells exposed to CHCl₃. However, the treatments with CH₂Cl₂ and BrCHCl₂ irregularly increased this enzyme's catalysis (Figure 3A). CAT activity was augmented in all treatments of CH₂Cl₂, particularly at the higher concentrations (3.48-3.51-fold-higher; p≤0.01). In contrast, MRC-5 exposed to BrCHCl₂ suffered a significant abolishment in CAT activity at 10^-12 to 10^-8 mol (1.83-2.16, fold-higher, p≤0.01); nevertheless, the response of this enzyme could be considered inconsistent in this treatment (Figure 3B). The catalysis of GPx was increased in all treatments with CHCl₃ and with CH₂Cl₂. However, a concentration-dependent response was not observed. Quite the opposite, BrCHCl₂ treatment irregularly improved this enzyme activity on MRC-5 cells (Figure 3C).

3.3. Evaluation of Phosphorylated NF-κB/p65 Levels

Interestingly, the degree of halogenation of halomethanes seems to be a critical factor in the phosphorylation of NF-κB. Levels of phosphorylated NF-κB at Ser536 on MRC-5 cells exposed to BrCHCl₂ increased in a concentration-dependent manner from 10^-14 to 10^-6 mol, ranging from 7.15 to 13.5-fold-higher (p≤0.05). A similar response was observed in MRC-5 cells exposed to CHCl₃ from 10^-12 to 10^-6 mol, with increments from 1.73 to 3.07-fold higher. However, the phosphorylation of NF-κB induction was lower with BrCHCl₂. In contrast, the treatment of CH₂Cl₂ does not induce significant differences regarding control cells (Figure 4).

3.4. Relation between Biomarkers

Levels of O₂• were statistically (p<0.001) related to H₂O₂ in MRC-5 cells treated with the three HMs under study. The lipid peroxidation was linked with ROS content in this cell line exposed to CH₂Cl₂ and BrCHCl₂. In the case of CHCl3, only statistical relationships between TBARS and H₂O₂ were found. The activity of SOD did not show relations with ROS except for BrCHCl₂ treatment. In contrast, CAT activity was related to H2O2 on MRC-5 cells exposed to the three HMs, and with O₂• and SOD in the treatments of CH₂Cl₂ and BrCHCl₂.

Nevertheless, CAT activity had statistical relationships with TBARS in cells exposed to chlorinated halomethanes. By contrast, GPx showed relations with H₂O₂ in human lung fibroblasts exposed to BrCHCl₂ and TBARS in the treatments with CHCl₃. Interestingly, both ROS levels were inversely related to phospho- NF-κB/p65 protein at Ser536 on MRC-5 cells exposed to chlorinated halomethanes; besides, O₂• concentration was also linked with this nuclear factor. On the other hand, levels of phospho- NF-κB/p65 protein at Ser536 showed an inverse association with TBARS in the treatment with CH2Cl2 and CAT activity in cells exposed to BrCHCl2 (Table 1).

3.5. Molecular Docking Analysis

In this study, the MOE was executed in as much time as required to give several docked conformations. The predicted energy analysis and the consistency of results were considered criteria to identify the best solution for the ligand/receptor model. The in-silico results showed that the estimated free energy of binding with the p65 subunit of NF-κB was -7.0 kcal/mol for both BrCHCl₂ and H₂O₂: -6.5 kcal/mol for CHCl₃, and -7.6 kcal/mol for O₂•. These interaction energy values implied that these ligands could bind with the nuclear factor with sufficient bond strength. In the case of CHCl₃, Cl1 bound with Thr191 with a distance of 3.94 Å, Cl2 with Thr191, Ala192, Leu194, and with Ser281 showing distances of 3.37, 2.96, 3.79, and 3.60 Å, respectively. Additionally, Cl₃ was in contact with Leu¹⁹⁴ at 3.76 Å in region RHD of IPT domain (Figure 5A). The Cl₁ anion of BrCHCl₂ is - in close contact with – Glu193 and Leu194 and interacts with backbone nitrogen atoms, respectively with the distances of 2.9 and 2.83 Å and electrostatic energy was 0.9 kcal/mol (Figure 5B). The Cl₂ and the Br ion contacts with backbone carbon and side chain oxygen of Thr191 with 3.0 Å distances. This finding suggests that backbone nitrogen plays a crucial factor in the interactions of Cl1 and Cl2 with the region RHD of the IPT domain of p65 of NF-κB. The biotransformation of CHCl₃ and BrCHCl₂ generated ROS (O₂• and H₂O₂), indication both chemical species interact with the NF-κB/p65 complex. Superoxide anion (O₂•) shows only four hydrogen bonds with back bone nitrogens and carbon atoms of Thr191, Ala192, Glu¹⁹³, Leu¹⁹⁴ predicting that H₂O₂ shows higher stability than (O₂•) with NF-κB/p65 complex. (Figure 6A). However, for H₂O₂, close contacts were detected showing five hydrogen bonds with backbone nitrogens and carbon atoms of Thr¹⁹¹, Ala¹⁹², Glu¹⁹³, Leu¹⁹⁴, and side chain oxygen of Leu²⁸⁰ (Figure 6B). The RMSD of Ca trace on X-axis and time (ns) on Y-axis generated using MD trajectory clearly show that H₂O₂ is highly stable than other ionic species CHCl₃, O₂•, and BrCHCl₂ showing in order CHCl₃<O₂• < BrCHCl₂ < H₂O₂ used in this study (Figure 7A). In addition, NF-κB/p65- H₂O₂ complex also forms higher number of hydrogen bonds that other three complexes predicting to be high stable (Figure 7B). Furthermore, difference in HUMO-LUMO energies calculated for the bound NF-κB/p65-ligand complex also clearly indicated that H₂O₂ and O₂• are chemically stable with -0.3 and -0.5 kcal/mol, respectively, showing a good correlation with ΔG calculated using thermodynamic integration and RMSD stability using MD simulation studies. However, SCF calculations have shown that CHCl₃ shows higher affinity with -46.4 a.u with ΔH of -23.6 kcal/mol, as shown in the Table 2.

4. Discussion

The lungs are commonly affected by ROS, and almost all cell types of this organ can produce these highly reactive molecules. Under basal conditions, ROS are synthesized by NADPH oxidase and act as second messengers to activate numerous intracellular proteins and enzymes involved in physiological and pathological stages [19,32]. In the lungs, pulmonary fibroblasts produce ROS, especially after stimulation by inflammatory cytokines with a diverse range of stimuli. However, uncontrolled ROS production could directly cause DNA damage, morphologic transformations in cells, and lung injury if the antioxidant system cannot control the generation of ROS [34,35]. Besides, in the environment, many toxicants can induce these prooxidant forces [36,37]. Nevertheless, there is no previous information about ROS production in human lung fibroblasts induced by HMs, a by-product of disinfection procedures to produce drinking water. The current study found that the toxicants understudy could induce ROS in vitro, particularly O₂•. Similar findings were also observed in peripheral blood mononuclear cells (PBMCs) of Cyprinus carpio carpio i.p dosed with CH₂Cl₂, CHCl₃ and BrCHCl₃ [19]. Despite this lack of information, we can propose that the generation of ROS in human fibroblasts treated with halomethanes is due to a series of events, not necessarily consecutive, as well as its feedback with prooxidant forces and oxidative damage summarized as follows: 1) Biotransformation processes are fundamental in the generation of ROS. 2) ROS generation is not directly regulated by mitochondrial activity. 3) The higher the number of halogens involved, the greater the probability of ROS generation. 4) Some physicochemical properties, such as the electronegativity of the metabolites generated during biotransformation, influence the ROS generation rate. 5) ROS induction depends mainly on specific and non-specific antioxidant defenses. The arguments presented are discussed below. ROS could be generated more by an independent pathway than mitochondrial complex I activity [19], such as the case of decoupling the flow of electrons during redox processes of cytochrome P450 isoenzymes (i.g. cytochrome P450 isoform 2E1 “CYP 2E1”). Other enzymes are involved in the HM's bioactivation process, as in the isoform tetha of the glutathione S-transferase (GSTT) [8,19,38,39,40]. Regarding the topic, the obtained results showed that the degree of chlorination of HMs is a critical factor in the induction of ROS in MRC-5 cells. In the first step of the biotransformation of the HMs, the three compounds under study can induce similar amounts of O₂• mediated by the activity of CYP 2E1 [19]. Next to this step, the HMs possessing three halogens suffer reductions, dehalogenations, and reductive dehalogenation mediated by some enzymes such as GSTT and other enzymes [8,41], allowing ROS generation [19]. However, the higher levels of O₂• elicited by CHCl₃ compared with BrCHCl₂ could be explained by HCl release, which possesses an electronegativity value (δ=1.06) higher than HBr (δ=0.86) released from BrCHCl₂ after dehalogenation processes [19]. Contrary to these halomethanes, the amounts of O₂• were the lowest in the cells treated with CH₂Cl₂. This response could be explained by a lesser number of reductive dehalogenations involved in the biotransformation processes that this compound undergoes [19]. Also, on the final step of biotransformation of CHCl₃ mediated by GSTT [42], the release of H⁺ after reductive dehalogenation probably enhances the respiratory rate and proton motive force [19], allowing ROS generation by mitochondrial complex I activity as well as by mitochondrial electron transport chain. In contrast, the significant diminution of O₂•levels in MRC-5 fibroblasts exposed to high concentrations of CH₂Cl₂ is feasible to explain by enzymatic dismutation of H₂O₂ through SOD activity as demonstrated in the current study. The activity of SOD plays a crucial role in the antioxidant system in human lungs [6,43,44]. This enzyme plays a vital role in the dismutation of O₂• into H₂O₂ and prevents chain reactions that induce •OH and peroxynitrite radicals (ONOO-). However, the O₂• may further convert into OH• in the presence of metal ions via Fenton and Haber-Weiss reactions [45]. The excessive production of ROS in the lungs can disrupt the oxidant/antioxidant balance, vital in developing pathologies in the airways [46,47,48,49]. Besides, excessive ROS levels can alter critical cellular components such as proteins, unsaturated lipids, and DNA by exposure to free radicals, thus compromising cell homeostasis [45,50].

On the other hand, the H₂O₂ generation by dismutation of O₂• could occur spontaneously (non-catalyzed) and too mediated by the activity of SOD [45]. The highest levels of H₂O₂ observed in MRC-5 cells treated with CH₂Cl₂ at 10^-20 mol were not accompanied by the highest concentration of its precursor (O₂•), which reinforces the hypothesis about a non-catalyzed dismutation, which probably occurred early in the present study, nor the enzymatic activity of SOD had a role, as has been previously reported [45]. In the current study, the H₂O₂ was related to O₂• levels in MRC-5 fibroblasts exposed to CHCl₃ but not with SOD activity. This finding suggests a non-catalyzed dismutation of O₂• to H₂O₂ in the presence of H⁺ generated during biotransformation at the late stage as observed - for MRC-5 cells treated with CH₂Cl₂. If H₂O₂ levels remain stable in the intracellular medium and escape to the enzymatic action of either CAT or GPx or some non-enzymatic element of the antioxidant system. This compound can activate signaling pathways that stimulate diverse cellular processes such as cell proliferation, differentiation, migration, or apoptosis [19,51,52,53,54,55]. In contrast, levels of H₂O₂ in MRC-5 fibroblasts treated with CH₂Cl₂ and with BrCHCl₂ were statistically related to O₂• concentration and showed a similar response in the low treatment. However, in the fibroblasts exposed to BrCHCl₂, the diminution of H₂O₂ was linked with the activity of CAT and GPx. Despite the crucial role of CAT in the dismutation of H₂O₂, the role of GPx in maintaining the redox status of lung cells exposed to BrCHCl₂ stands out. This enzyme also removes organic hydroperoxides and participates in glutathione's redox cycle, which is abundant in the alveolar epithelial lining fluid [56]. Regardless of the enzymes involved in the antioxidant defense activity, ROS induction elicited lipid peroxidation in vitro in human lung fibroblasts.

In addition to the cytotoxicity elicited by HMs previously reported in MRC-5 cells [8], the current study shows that CHCl₃-induced lipid peroxidation evaluated as TBARS on this cell line exposed to higher concentrations of this toxicant. The induction of lipid peroxidation was dependent on halomethane concentrations in MRC-5 cells. Indeed, the damage is induced by the remaining amount of ROS generated and not detoxified through the antioxidant defense, as previously discussed. However, this imbalance between the prooxidant forces (ROS) and antioxidants (SOD, CAT, and GPx) produces a series of radical reactions, including, 1) the initiation caused by the predominance of pro-oxidant forces, 2) the propagation of free radicals to reach deep levels of damage and destruction of the lipid bilayer and3) formation of polymers between free radicals might occur in the best of cases. The background on the subject is set out below. Lipid peroxidation is an oxidative process that involves the deterioration of the lipid bilayer mediated by hydroxyl radical (^•OH) to form carbon-carbon double bonds at the initial stage. - The ^•OH is considered the most critical oxidizing chemical species in the initiation of this process [57]. However, through chain reactions, other chemical species such as alkyl hydroperoxides (LOOH), alkyl peroxyl radicals (LOO•), and alkoxyl radicals (LO) may also be formed [45]. The level of lipid peroxidation induced by CHCl₃ was related to H₂O₂ concentration. This oxidation probably implies damages in lung fibroblast membranes (both in the cell and organelles) that can reach several levels of gravity, such as altering the fluidity of the lipid bilayer on its functionality even on its breaking. The consequences can be diverse, but it is possible to emphasize modifications in hormone receptors and proteins involved in signal transduction pathways [45]. A previous report found that a high concentration of CHCl₃ elicited a decrease in cell growth of MRC-5 cells in all concentrations assessed; however, higher concentrations of this toxicant induced cytotoxicity [8]. It is possible to speculate that lipid peroxidation elicited by CHCl₃ was directly related to various damages, including diminution of lung fibroblast growth and cytotoxicity; however, the lungs could counteract these damages. In this regard, Kornbrust and Mavis reported that lipid peroxidation is about 25 to 50 times higher in the lungs and heart, which are highly oxygenated tissues, than in the liver, kidneys, testes, and brain. Indeed, in lung cells, lipid peroxidation is a process that occurs regularly, but pulmonary fibroblasts are highly effective in remodeling the cells and repairing the damaged tissues. In this regard, the lung can regenerate new cells efficiently after injury. Lung regeneration comprises progenitor cell activation and cell replacement through the proliferation of intact cells [59,60,61]. Interestingly, the higher concentration of CHCl₃ induced a diminution in TBARS levels in MRC-5 cells about the treatment of 10^-6 mol, which could be due to GPx activity. However, a significant inverse relationship between TBARS and CAT activity also stands out. This response suggests that oxidative damage was provoked by H₂O₂, whose observed decrease to 10^-4 for 10^-6 was consistent with increased CAT activity and a decrease in lipid peroxidation. In contrast, in human lung fibroblasts exposed to BrCHCl₂, only the median concentrations elicited an increase in lipid peroxidation, which was related to levels of ROS. However, higher concentrations of this toxicant cause a significant reduction in this oxidative damage, which could be related to increase GPx activity even though a significant relationship was not found. Considering the lung's capacity to repair damaged cells or regenerate new cells after an oxidative injury elicited by ROS, observing the relationships among other biomarkers involved in regulating the cell cycle case of NF-κB is probably helpful.

Remarkably, only the HMs possessing three halogens (BrCl₂CH and Cl₃CH) were able to increase the levels of phospho-NF-κB/p65 protein at Ser536; however, the magnitude of this induction and relationships with the biomarkers of oxidative stress response was different. Although the quantitative structure-activity relationships (QSARs) would offer some explanations about the differentiated response of NF-κB translocation to the nucleus, the molecular docking tools provide detailed information about the disruption of the IκBα- NF-κB/p65 complex by the effects of HMs treatment.

The in-silico results using the software molecular operating environment (MOE) showed that the estimated free energy of binding with the p65 subunit of NF-κB was -7.0 kcal/mol for BrCHCl₂ and Cl₃CH binds with -6.5 kcal/mol. The interaction energy values implied that these ligands could bind with the nuclear factor with sufficient bond strength, which coincided with the higher phosphorylation levels of NF-κB/p65 in a concentration-dependent manner, particularly for BrCHCl₂. The bonds of chloride and bromide from HM have occurred in the C-terminal domain of p65 of NF-κB/p65; this is a relevant finding because the C-terminal transcription activation domain (TAD) regulates the gene expression, and is highly active [62]. Notably, we found that Cl₁, Cl₂, and Br are in contact with four contiguous amino acid residues (Thr¹⁹¹, Ala¹⁹², Glu¹⁹³, Leu¹⁹⁴) into region RHD (Rel homology domain), which is involved in DNA-binding. In this sense, it has been reported that the IPT domain is the dimerization segment of the NF-κB/p65 and related transcription factors [63,64], covering 101 a.a. from Thr¹⁹¹ to Asp²⁹¹. Also, the IPT domain acts as a dominant regulator of diverse stressful conditions, including physical stress, oxidative stress, and exposure to certain chemicals [65,66,67]. Besides, there are critical conserved residues in this domain: Arg198, Glu211, Leu215, and Cys216 in the dimer interface, which is dominated by hydrophobic interactions close to residues of the bond of Cl₁, Cl₂ and bromide anions [68].

On the other hand, post-translational modifications such as phosphorylation, acetylation, and methylation of transcription factors may affect NF-κB transcriptional activity [69]. Remarkably, the phosphorylation of IκB proteins followed by ubiquitination and degradation by proteasomes releases the NF-κB/p65 homodimer, allowing the bind of p65 with DNA as a homodimer or as heterodimers [70,71,72]. In this context, there are several phosphorylation sites in the IPT domain close to the binding residues of Cl₁, Cl₂ and bromide anions of the HMs, i.e. residues 254, 276, 281, particularly stands out the contact with Ser²⁸¹. A previous study on a site-directed mutagenesis screen for potential phosphorylation sites within the p65 RHD identified S205, S276, and S281 essential for p65 transcriptional activity [17,73]. These studies and current results suggest that phosphorylation of the NF-κB subunits profoundly affects the function of NF-κB. Certain phosphorylation events contribute to the selective regulation of NF-κB transcriptional activity in a gene-specific manner [74]. In agreement with our result, the binding of the backbone carbon of CHCl₃ with Ser281 may affect the complex IκBα/p65 conformation, allowing the hiper-proliferation of MRC-5 as documented in a previous study [8]. In this regard, it has been reported that the phosphorylation of p65 induces a conformational change, which impacts p65 ubiquitination and stability and protein-protein interactions. However, NF-κB/p65 is a highly dynamic and flexible protein not fixed in a globular and well-ordered shape but can change dynamically after NF-κB activation [75]. We consider that the bond of Cl₁, Cl₂ and bromide of BrCHCl₂ and CHCl₃ would affect the structure of the p65 homodimer as well as its interaction with IκBα, and the dissociation of the Iκβα-NF-κB complex involved in the activation of target genes transcription. However, to evaluate the effects of these HMs in human fibroblasts, the influence of some of the more reactive metabolites derived from biotransformation processes must be considered, such as in the case of ROS.

Despite the preceding results showing relationships among oxidative stress response, and NF-κB/p65 activation [76,77,78,79,80,81,82,83,84,85,86] there is no previous information about ROS production in human lung fibroblasts induced by HMs. However, in a previous study, ROS was inducted in peripheral blood mononuclear cells (PBMCs) of Cyprinus carpio carpio i.p treated with CH₂Cl₂, CHCl₃ and BrCHCl₂ was documented [19]. Results of the current study showed that the toxicants under study induced ROS in vitro with an estimated free energy of binding with NF-κB/p65 of -7.0 kcal/mol for H₂0₂ and -7.6 kcal/mol for O₂•. The interactions of these ROS occurred in the IPT domain of NF-κB/p65, suggesting the activation of target genes is involved in the antioxidant response, among others, as shown by Pearson correlation analysis. Certain NF-κB-regulated genes play a major role in regulating the amount of ROS in the cell because ROS have various inhibitory or stimulatory roles in NF-κB signaling. The interactions of ROS with cysteine residues are essential to disturb the NF-кB pathways. For example, hydrogen peroxide inhibits IKK activation by acting on its cysteine residues, eliciting, in this way, inhibitory effects in the catalytic domains of tyrosine phosphatases [87,88].

Biochemical analysis in MRC-5 cells treated with HMs and computational study using MOE2019 showed that BrCHCl₂ and CHCl₃ induce the phosphorylation of NF-κB/p65 by activating the IPT domain. The consequence of these interactions is probably linked with the hyper-proliferation of these cells, as previously documented [8]. However, the ROS generated during the metabolism of these toxicants disturb the activation of the IPT domain involved in the antioxidant response mediated by NF-κB/p65 as observed by negative relationships among ROS with SOD, CAT, and GPx found in cells treated with CHCl₃.

5. Conclusions

The degree of chlorination of CHCl₃ and BrCHCl₂ is a critical factor in the induction of ROS on MRC-5 lung human fibroblasts, which could be a counterpart to this damage mainly by CAT and GPx activities. Nevertheless, during the biotransformation of these toxicants, the generation of toxic metabolites such as H₂O₂ and O₂• interacts with the phosphorylation site located on IPT domain of NF-κB/p65 complex. However, the higher binding energy of BrCHCl₂ with NF-κB/p65 complex accompanied by the significant levels of phospho- NF-κB/p65 allows us to suggest that this halomethane contributes to the dissociation of the IκBα-NFκB complex as well as the transcription of target genes. In contrast, ROS probably disrupts the transcription of other genes, such as the antioxidant enzymes.