The Oral Delivery of Water-Soluble Phenol TS-13 Ameliorates Granuloma Formation in an In Vivo Model of Tuberculous Granulomatous Inflammation

The redox-sensitive signaling system Keap1/Nrf2/ARE is a premier protective mechanism against oxidative stress that plays a key role in the pathogenesis and development of various diseases, including tuberculous granulomatous inflammation. We have previously reported that novel water-soluble phenolic antioxidant TS-13 (sodium 3-(4′-methoxyphenyl)propyl thiosulfonate) induces Keap1/Nrf2/ARE and attenuates inflammation. The aim of this study is the examination of the effect of TS-13 on tuberculous granulomatous inflammation. BALB/c mice were administered TS-13 (100 mg kg−1 day−1) through their drinking water starting immediately after Bacillus Calmette-Guérin (BCG) intravenous injection. Histological changes, production of reactive oxygen species (ROS) (activity of free-radical oxidation processes), and mRNA expression of Nrf2-driven, NF-κB-, AP-1-, and autophagy-dependent signal pathway genes in the liver and peritoneal exudate were evaluated 30 days later. After the 30th day of infection, the activity of the Keap1/Nrf2/ARE system was decreased and its effector genes entailed increasing ROS production in the liver. Therapeutic intervention with TS-13 is aimed at activating the Keap1/Nrf2/ARE system that leads to an increase in Nrf2 and Nrf2-mediated gene expression and a decrease in NF-κB expression. Changes in these pathways resulted in a decline of ROS production and a decrease in the number and the size of granulomas. In total, the results indicate that the Keap1/Nrf2/ARE system can be an effective pharmacological target in host-adjunctive treatment of tuberculosis.


Introduction
Before the advent of specific antituberculosis drugs (streptomycin, isoniazid, ethambutol, rifampicin, and pyrazinamide) in the middle of the last century, pathogenetic therapy was aimed at reestablishing an ablated immune system and correcting disturbed vital processes; this was the main treatment for patients with tuberculosis. The antimycobacterial therapy demonstrated undoubted success, and resorting to treatment using a specific diet (fish oil or cod liver oil rich in vitamin D) and other methods of pathogenetic therapy have fallen into oblivion or have been put aside. The development of combinatorial therapeutic modalities with the involvement of additional antimycobacterial agents and conventional che-motherapy indicated a breakthrough in therapeutic interventions for tuberculosis, especially in cases of multidrugresistant tuberculosis. Among the recent drug discoveries, new "host-targeted" adjunct therapeutic strategies have emerged [1][2][3][4][5].
The key point in Mycobacterium tuberculosis survival and persistence in the host is its uptake by macrophages and the subsequent formation of granulomas. The factors involved in the formation and sustainment of tuberculous granuloma are not well understood, but they involve adaptations of both the host and pathogen to a changing microenvironment and redox state. Whether granuloma plays a protective or destructive role in tuberculosis pathogenesis and treatment is the subject of great debates [1,4,12]. Besides, the precise mechanisms underlying the redox biology of granulomatous tuberculous inflammation are still obscure. Therefore, the master regulator signaling system Keap1/Nrf2/ARE (Kelch-like ECH-associated protein 1/NFE2-related factor 2/antioxidant respons(iv)e element) can serve as an important target in the pathogenetic treatment of tuberculosis affecting antioxidant protection, xenobiotic detoxification, and activity of reactive oxygen species-(ROS-) generating enzymes [10,[13][14][15].
In this connection, we have been investigating the impact of sulfur-containing phenolic antioxidant TS-13 on the activation Nrf2 transcription factor [16], on Bacillus Calmette-Guérin-(BCG-) induced granuloma formation and ROS production in mice.

Materials and Methods
2.1. Reagents. Luminol was purchased from SERVA Electrophoresis GmbH; 2 ′ ,7 ′ -dichlorodihydrofluorescein diacetate and TRIzol Reagent were obtained from Invitrogen; phorbol 12-myristate 13-acetate (PMA) and zymosan were obtained from Sigma-Aldrich; anti-Nrf2 antibodies (ab31163) were purchased from Abcam; RNAlater was purchased from Qiagen; iScript cDNA Synthesis Kit was obtained from Bio-Rad; BioMaster HS-qPCR mix was obtained from Biolabmix; culture media and fetal bovine serum were purchased from Biolot Medical; mounting media were obtained from BioVitrum; BCG vaccine was obtained from Microgen; and sodium 3-(4′ -methoxyphenyl)propyl thiosulfonate (TS-13, Figure 1) was synthesized based on 2,6-di-tert-butyl-phenоl as described previously [16]. The main substance content in the sample measured by HPLC was 98-99.2%. All other chemicals were analytical grade and were obtained from standard commercial supplies.

Animal
Model. The experimental protocol was approved by the Animal Care Committee of the Federal Research Center for Fundamental and Translational Medicine (Novosibirsk, Russia). Male BALB/c mice (weight: 18-22 g; age: 2 months) were purchased from the Federal Research Center Institute of Cytology and Genetics, Siberian Branch of the Russian Academy of Sciences (Novosibirsk, Russia). The animals were housed in an environment where temperature and light were controlled, and food and water were provided ad libitum. Mice were injected with 0.5 mg BCG in 0.2 mL saline solution into the tail vein (two groups, 8 animals per group) or 0.2 mL saline solution as a vehicle control (control group, n = 6). One group of BCG-infected animals was supplemented with a daily dose of 100 mg kg -1 day -1 TS-13 dissolved in drinking water for 30 days, while the other groups received water. Mice were sacrificed by cervical dislocation. Peritoneal macrophages were isolated to evaluate the oxidative metabolism of these cells. The livers were quickly removed, weighed, and processed for histological and immunohistochemical examination and preparation of liver homogenates. Samples were collected and stored in RNAlater with subsequent nucleic acid extraction.

Histological Examination.
Liver fragments were fixed in 10% neutral formalin, dehydrated in ascending alcohol solutions, and embedded in paraffin. Sections (4-5 μm thick) were stained using the hematoxylin/eosin techniques and studied using light microscopy (AxioImager A1, Carl Zeiss). Specific histochemical staining by the Ziehl-Neelsen stain was used to visualize Mycobacterium bovis in the tissues. The numerical density of the granulomas and their diameters were determined by using the morphometry method (Axio-Vision software, rel. 4.8). These parameters were used as the morphological criteria for tuberculosis activity.
For immunohistochemical study, 3 μm sections were dewaxed and dehydrated, and epitopes were retrieved in citrate buffer solution in a 700 W microwave oven for 20-35 minutes. Endogenous peroxidase was blocked. The samples were incubated with primary anti-Nrf2 antibodies according to instructions. Then, the sections were incubated with horseradish peroxidase (HRP) conjugate (using a Spring Bioscience Corporation detection system) and 3,3′-diaminobenzidine (DAB) substrate and then stained with Mayaer's hematoxylin. Sections were dehydrated using increasing concentrations of ethanol and xylol mounted with synthetic mounting media "Bio Mount," placed under cover glasses and studied using light microscopy (AxioImager A1, Carl Zeiss).

Activity of Free-Radical Oxidation Processes
2.4.1. Chemiluminescence (CL). The livers were rinsed with saline, minced with scissors, and homogenized on ice in a Potter-Elvehjem tissue grinder with 5 vol (w/v) of Hanks' balanced salt solution without phenol red (HBSS) (200 mg/mL). After recording the background CL of the measuring cuvette at 37°C in a chemiluminometer (Photon, Russia) for 2 minutes, 2 mL of liver homogenate was placed in the cuvette and then incubated for 2 minutes; then, the spontaneous CL was measured for 2 minutes. Afterwards, 0.1 mL of 100 nM luminol solution was injected, and the luminol-amplified CL (LACL) was measured for 2 minutes. 2 Oxidative Medicine and Cellular Longevity

Flow Cytometry.
Peritoneal exudate cells (PECs) were obtained by peritoneal lavage with cold RPMI 1640 medium supplemented with 1% (v/v) fetal bovine serum and kept on ice until measured. To measure the total ROS production, isolated PECs were incubated for 15 min in 1 mL of HBSS containing 10 μM 2′,7′-dichlorodihydrofluorescein (DCF) diacetate at +37°C. The former is deacylated intracellularly and rapidly oxidized by ROS (predominantly H 2 O 2 ) to yield the highly fluorescent product DCF. We measured the intensity of the DCF-dependent fluorescence of PECs stimulated with 100 nM PMA (λ Em = 488 nm, λ Ex = 520 nm) by using the FACSCalibur (Becton Dickinson, USA) flow cytometer. The gating of the viable macrophages and neutrophils was based on light scattering (forward and side scatter) properties. The results of the cell fluorescence intensity were normalized to the fluorescence in control mice.

Quantitative
Real-Time PCR. Total RNA was extracted from tissue using the TRIzol Reagent. Total RNA was converted to cDNA using the iScript cDNA Synthesis Kit. The real-time PCR reactions were carried out using a BioMaster HS-qPCR mix. All reactions were run in triplicate in 96well plates on a CFX96 system (Bio-Rad), and threshold cycles (Cts) were obtained after 40 cycles. A quantitative assessment was carried out by the delta delta Ct method,  The relationships between the variables were assessed by Spearman's rank correlation coefficient (r S ). Differences between experimental groups and correlations were considered as statistically significant at a p value of less than 0.05. BCG injection induced a marked granulomatous response in the liver of mice with the formation of diffuse, principally perivascular granulomas consisting primarily of macrophages and containing a considerable content of lymphocytes. Morphometric analysis showed that 30 days of TS-13 treatment was followed by a significant 2.5-and 1.4time decrease in granuloma numerical density and diameter, respectively (Figure 2(a)). Besides, granulomas in the liver of BCG-injected TS-13-treated mice contained significantly less epithelioid cells (by 29.8%) and neutrophils (complete absence) in comparison with BCG-injected TS-13untreated animals (Figure 2(b)).

Results and Discussion
Immunohistochemical analysis of liver samples has demonstrated weak Nrf2 expression with cytoplasmic localization   Oxidative Medicine and Cellular Longevity in the control group. In the BCG-infected group, Nrf2 expression was more pronounced in hepatocytes with minimal expression in granulomas. In BCG-infected TS-13-treated mice, Nrf2 was abundant with cytoplasmic and nuclear localization and strong expression in granulomas (Figure 3).

Bidirectional Effect of TS-13 on Free-Radical Oxidation
Processes. Spontaneous CL is a measure of the generation of lipid alkoxyl and peroxyl radicals, i.e., the intensity of lipid peroxidation [17,18]. Spontaneous CL levels of liver homogenates did not differ significantly among groups, but when luminol was introduced into a registration system, we observed a significant increase of CL intensity in the BCGinfected group; the same was observed after adding H 2 O 2 (Figure 4(a)). The levels of liver LACL and H 2 O 2 -LACL intensity of the BCG-infected mice receiving TS-13 were sig-nificantly lower in comparison with those of the positive control (BCG-injected mice that received water) (Figure 4(a)). The PMA-stimulated ROS production by PECs in total and by peritoneal neutrophils from BCG-infected mice was similar to the negative control, whereas DCF-dependent fluorescence of peritoneal macrophages was significantly lower (Figure 4(b)). In BCG-infected TS-13-treated mice, PMA-stimulated fluorescence of total PECs, peritoneal macrophages, and peritoneal neutrophils was statistically higher compared to the corresponding positive control group (Figure 4(b)). 7 Oxidative Medicine and Cellular Longevity involved in redox-dependent signaling pathways, namely, Nrf2-, NF-κB-, and AP-1-mediated ( Figure 5). 30 days after BCG infection, mRNA expression of Nrf2 significantly decreased and mRNA content of ARE-driven enzymes (NQO1, glutathione S-transferase P1, and heme oxygenase 1) tended to decrease in the liver of BCG-infected mice in comparison with negative control. An evaluation of mRNA expression of those genes in BCG-infected animals receiving TS-13 with drinking water revealed higher levels of Nfe2l2, Nqo1, Gstp1, and Hmox1 mRNAs relative to positive control (BCG-infected mice). The expression pattern of p65 (Rela), an important subunit of the classical NF-κB pathway, was similar in all 3 groups with a tendency of increasing (p = 0:088) in the third group compared to negative control, whereas p100 (Nfkb2) mRNA content increased in the BCG-infected group compared to negative control, indicating the alternative activation of NF-κB, which tended to decrease in BCG-infected TS-13-treated mice. Expression of the c-Jun protooncogene (AP-1, Jun) was upregulated in the BCG-infected group with a significant decrease in the TS-13-treated group.

Enhancement of ARE-Driven
Expression of autophagy effector gene Map1lc3b (Figure 6(a)) after 30 days of infection did not change in both groups.Becn1 mRNA expression tended to be more pronounced in the BCG+TS-13 group as compared to positive control (p = 0:056) (Figure 6(b)). Expression of lysosomal biogenesis genes Tfeb (Figure 6(c)) and Tfe3 (Figure 6(d)) tended to increase in the BCG group (Tfeb mRNA content enhanced significantly). However, Tfeb expression decreased to control value in the liver of TS-13-treated mice.

Discussion
The main purpose of our study is to establish ROS production during experimental tuberculosis in mice and to deter-mine whether induction of the Keap1/Nrf2/ARE signaling pathway could affect granuloma formation.
The lack of Nrf2 nuclear translocation in the BCGinfected group would suggest a defect in Nrf2-mediated antioxidant defense, as previously described [10,19]. This fact was confirmed by evaluating the expression of the following Nrf2-mediated genes: Nqo1, Gstp1, and Hmox1. Decreased activity of the Keap1/Nrf2/ARE system and its effector genes entails increasing ROS production in the liver. Therapeutic intervention with TS-13 is aimed at activating the Keap1/Nr-f2/ARE system that leads to an increase in Nrf2 and Nrf2mediated gene expression and a decrease in NF-κB expression. The changes in these pathways resulted in a reduction of ROS generation and a decrease in the number and size of granulomas.
We examined the association between the severity of granuloma formation, the intensity of free-radical oxidation processes, and the expression of the Keap1/Nrf2/ARE signaling pathway genes ( Table 2). A multivariate regression analysis showed that the free-radical processes involving ROS were important for BCG-induced formation of granulomas in the liver. This is confirmed by our previous assumption of the steady-state hydrogen peroxide concentration increasing in granulomas [20].
Contribution of classical pathway activation of the proinflammatory transcription factor NF-κB was limited. Furthermore, there was an inverse relationship between Rela expression and ROS generation and a direct dependence between Rela and Nfe2l2 expression. In contrast, the infection process was accompanied by alternative NF-κB pathway activation that correlated with the H 2 O 2 production and the c-Jun subunit of transcription factor AP-1 expression.
After 30 days of infection, there was an inverse relationship between the severity of granulomas or ROS formation and Becn1 expression. This phenomenon is  Oxidative Medicine and Cellular Longevity consistent with observations that tuberculous granulomatous inflammation is accompanied by autophagy inhibition, including a decline of Becn1 expression [5,10,11,21]. At the same time, the elevated levels of expression of lysosomal biogenesis genes Tfe3 and Tfeb positively correlated with expression of Nfkb2 and Jun genes, characterizing the alternative macrophage activation status. The TFEB transcription factor ortholog is a key regulator of immune responses in Candida elegans, coordinating expression of genes that are controlled by NF-κB (absent in C. elegans) in mammals. Probably, TFEB is one of the oldest high-level regulators of nonspecific immune resistance, which lost the leading role in immune reactions in mammals in connection with the appearance of the more universal regulator NF-κB. Increased PMA-stimulated ROS production by peritoneal macrophages in BCG-infected TS-13-treated animals might be due to the enhancement of Nox4 expression as a result of Nrf2 activation under the influence of TS-13 [22,23]. Mice with partial Keap1 knockout (constitutive activation of Nrf2) is known to have an increased expression of NADPH oxidase homolog NOX4 [15]. PMA stimulation leads to the activation of the NOX4 enzyme and ROS production [23,24]. At the same time, Nrf2 contributed to the regulation of purine biosynthesis by controlling the expression of genes encoding enzymes in the pentose phosphate pathway, and induction of the Keap1/Nrf2/ARE system might be followed by an increase in the generation of NADPH that is essential for NADPH oxidase-dependent ROS production during a phagocyte respiratory burst [25][26][27].

Conclusions
It was shown that the induction of the redox-sensitive signaling system Keap1/Nrf2/ARE in a tuberculous inflammation model reduces the number and size of granulomas and changes the cellular composition of granulomas by decreasing the content of neutrophils (the main destructive factor of inflammation) and epithelioid cells (a factor of the matu-rity of granulomas). Our results suggest that the Keap1/Nr-f2/ARE system could be a plausible and promising pharmacological target for tuberculosis treatment.

Data Availability
The data obtained in this study are available from the corresponding author upon request.

Conflicts of Interest
The authors have no conflicts of interest to declare.