Colonic and Hepatic Modulation by Lipoic Acid and/or N-Acetylcysteine Supplementation in Mild Ulcerative Colitis Induced by Dextran Sodium Sulfate in Rats

Lipoic acid (LA) and N-acetylcysteine (NAC) are antioxidant and anti-inflammatory agents that have not yet been tested on mild ulcerative colitis (UC). This study aims to evaluate the action of LA and/or NAC, on oxidative stress and inflammation markers in colonic and hepatic rat tissues with mild UC, induced by dextran sodium sulfate (DSS) (2% w/v). LA and/or NAC (100 mg·kg·day−1, each) were given, once a day, in the diet, in a pretreatment phase (7 days) and during UC induction (5 days). Colitis induction was confirmed by histological and biochemical analyses (high performance liquid chromatography, spectrophotometry, and Multiplex®). A redox imbalance occurred before an immunological disruption in the colon. NAC led to a decrease in hydrogen peroxide (H2O2), malondialdehyde (MDA) levels, and myeloperoxidase activity. In the liver, DSS did not cause damage but treatments with both antioxidants were potentially harmful, with LA increasing MDA and LA + NAC increasing H2O2, tumor necrosis factor alpha, interferon gamma, and transaminases. In summary, NAC exhibited the highest colonic antioxidant and anti-inflammatory activity, while LA + NAC caused hepatic damage.


Introduction
The use of antioxidants is considered an important complementary therapy in several conditions such as cardiometabolic, neurological, and gastrointestinal (GI) diseases, for cancer prevention and others [1]. Among the GI disorders, inflammatory bowel diseases (IBD), composed of ulcerative colitis (UC) and Crohn's disease (CD), affect 150-250/100,000 people, especially in the USA and Europe, with substantial health care costs of approximately US$6 billion and €4.6-5.6 billion, per year, respectively [2,3].
Despite extensive research, the etiology of IBD has yet to be elucidated; however it is considered closely connected to genetic and immunologic factors, microbiota, and oxidative stress, which, in the case of UC, concerns both etiology and symptoms such as increase of intestinal permeability by destruction of tight junctions and increase of infection and inflammation by neutrophil infiltration. Hence, the presence 2 Oxidative Medicine and Cellular Longevity of colonic ulcers, as a consequence of lipid peroxidation and protein damage and development of cancer, is due to DNA damage [4]. However, the cause of these symptoms and clinical manifestations is heterogeneous, varying according to the UC level: mild, moderate, or severe. In humans, the most common and problematic stage of UC is mild colitis [5]. It is characterized by normal albumin, body temperature, pulse, and hematocrit ratio, associated with an erythrocyte sedimentation rate of <20 mm/h, less than 4 bowel movements per day and no weight loss [6]. Nevertheless, this classification may differ according to authors and adopted criteria [7,8].
The extraintestinal manifestations of IBD, such as the systemic effects of alternative therapy used for IBD treatment, are poorly explored. Among these manifestations, emphasis is given to hepatobiliary disorders. The close relationship between the liver and intestine is justified by their common embryogenesis until later in adult life (intestine to portal vein) [9]. However, the main focus of the cited studies was microbiota, which has received special attention due to its intimate connection with metabolic syndrome, obesity, and nonalcoholic fatty liver diseases [10].
A pioneer study showed that, after colitis induction (4% of DSS for 7 days), important liver injury occurred, confirmed by higher serum levels of haptoglobin and various histological findings, such as signs of necrosis and ballooning of hepatocytes [11].
The use of antioxidants has emerged as an alternative therapy for IBD. Among them, lipoic acid (LA) and Nacetylcysteine (NAC) stand out, which have been tested separately in UC [4], and in combination for other clinical conditions [11,12], with positive results.
Due to its antioxidant property in both forms, oxidized (LA) and reduced (dihydrolipoic acid, DHLA), LA is called a "universal antioxidant." The LA/DHLA couple is a scavenger of superoxide anion radical (O 2 •− ), hypochlorous acid (HOCl), peroxynitrite (ONOO − ), and nitric oxide ( • NO). It is able to restore the exogenous antioxidants tocopherol and ascorbic acid and the endogenous antioxidant system of reduced/oxidized glutathione (GSH/GSSG) [6]. Additionally, in vitro studies suggest that LA acts as an inhibitor of I B kinase-2 (IKK2), with subsequent release of nuclear factor kappa B (NF-kB) [13]. Another anti-inflammatory effect is the elevation of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) intracellular levels that occurs through independent mechanisms, by breaking links between Nrf2 and Keap 1 (Kelch ECH associating protein 1). These actions lead to decreased expression of various proinflammatory cytokines and increased expression of enzymatic antioxidants, such as glutathione peroxidase (GPx) and superoxide dismutase (SOD) [14].
NAC is a strong reducing agent [15]. Like LA, NAC is an important scavenger of reactive molecules and plays a role as a metal chelator. However, the most important antioxidant action assigned to NAC is the increase in antioxidant defense activity by providing cysteine, which is required for GSH synthesis. The anti-inflammatory effects of NAC have also been confirmed, through observation of NF-kB inhibition [16].
Taking into consideration that LA and NAC, alone and in combination, were successfully tested in several systems and in regard the intimate bowel-liver association, the aim of this study was to evaluate the action of LA and/or NAC, on oxidative stress and inflammation in mild colitis induced by DSS.

UC Induction and Experimental
Design. Animals were divided into 5 groups. Healthy rats received commercial feed (control group). Mild UC was induced in four groups with 2% w/v DSS, administered in drinking water for 5 days. Antioxidants-treated rats were supplemented 7 days before mild colitis induction with lipoic acid (LA, 100 mg/kg/d), N-acetylcysteine (NAC, 100 mg/kg/d), or LA + NAC (100 mg/kg/d, each). Doses of the two antioxidants used are considered safe to rats/mice (LA up to 1800 mg/d [17] and NAC up to 6000 mg/d) and have been tested successfully in several physiological/pathophysiological conditions in mice/rats [16,18].

Euthanasia, Blood Sample Collection, General Biochemical
Profile, and Preparation of Tissue Homogenates. At the 13th day, after 12 h of fasting, the animals were anesthetized (ketamine, 100 mg/kg, and xylazine, 15 mg/kg, via i.p.), and blood was collected by cardiac puncture. After perfusion with heparine solution (1 : 50 v/v), aorta was sectioned. The organs were, then, dissected, placed on liquid nitrogen, and immediately stored at −80 ∘ C. Tissue homogenates were prepared, within ice cold, with Ripa buffer with protease inhibitor cocktail (one tablet for 50 mL of Ripa buffer), and were centrifuged at 19.600 ×g/20 min, at 4 ∘ C. The supernatants were stored at −80 ∘ C.
Biochemical analyses were made by certified laboratory, using well-established methods. Colon and liver tissues were washed in saline and weighed. The left lobe of the liver and distal colon were cut, for use in histological analysis.
2.6. Histological Analysis. After fixation with 10% buffered formalin, the organs were cleaved and the sections were processed by embedding in paraffin and stained with hematoxylin and eosin (HE) to histological evaluation.
Total collagen was evaluated by Masson trichrome staining kit. Briefly, one section of each, liver and bowel, was obtained and stained by using Masson's original trichrome stain, with collagen stained in blue, nuclei in purple-brown, and cytoplasm in pink. Collagen area was defined as the distinct blue color region and was distinguished from muscle, blood, and inflammatory cells. Afterwards, ImageJ5 software was used to quantify the blue area (pixel/camp).

Malondialdehyde (MDA), Hydrogen Peroxide (H 2 O 2 )
, and Nitrite Levels. MDA contents were measured by reverse phase ion pair HPLC, with UV detector at 270 nm, according to Tatum et al. [19]. HPLC system conditions were C18, ultrasphere column 150 mm × 4.6 mm and 45 mm × 4.6 mm guard column, the mobile phase comprised acetonitrile HPLC/UV grade, and Trizma buffer (pH 7.4) (1 : 9). The intestinal or hepatic tissue was homogenized with Trizma buffer, BHT, and acetonitrile. Then, the homogenate was centrifuged at 872 ×g/10 min, at 4 ∘ C, and the supernatant was filtrated with HPLC filter (0.22 mm). The flow rate was 1.0 mL⋅min −1 and MDA was calculated from the standard curve, generated using 1,1,3,3-tetramethoxypropane and expressed as nmol MDA⋅mg tissue −1 . The retention time was around 2 min 40 s. H 2 O 2 was measured according to Pick and Keisari [20]. Hepatic or colonic tissues were homogenized in 1.0 mL of assay buffer containing 140 mM NaCl, 10 mM PBS, pH 7.0, and 5.5 mM dextrose. After centrifugation (2000 ×g/5 min at 4 ∘ C), the supernatant was transferred to microtubes, containing 0.28 mM phenol red, 8.5 U/mL of horseradish peroxidase, and assay buffer. After incubation (37 ∘ C for 30 min), NaOH 1 M was added. The samples were read at 610 nm. The concentration was expressed as nmol⋅mg protein −1 .
Nitrite assay was performed based on Griess method adapted for microplates. Supernatant was mixed with 2,3diaminonaphthalene (50 g⋅mL −1 in 0.2 M HCl) and incubated at room temperature/10 min. The reaction was interrupted by the addition of NaOH (2.8 M) and monitored at 540 nm. The result was expressed as mol⋅mg protein −1 .

Measurement of Enzyme
Activity. SOD was measured following S. Marklund and G. Marklund [21]. Liver or colon supernatant was added to 0.2 mM pyrogallol (dissolved in 50 mM potassium phosphate buffer (PBS), pH 6.5) to initiate the reaction, and the decrease of the absorbance related to pyrogallol was monitored at 420 nm. One unit of SOD was defined as the amount required for inhibiting pyrogallol autoxidation by 50% per min. The result was expressed as U⋅mg protein −1 [22].
CAT activity was measured as the rate of decomposition of H 2 O 2 , as described elsewhere [23], and was monitored at 540 nm. Relative activity was expressed as IU⋅min⋅mL⋅mg protein −1 .
Total GPx activity was measured according to Flohe and Gunzler [24], adapted to microplate. The hepatic or colonic tissue was homogenized with assay buffer (PBS 0.1 M, EDTA 5 mM, pH 7.4) and centrifuged (14.000 ×g/20 min, at 4 ∘ C). Supernatant was added to wells (in duplicate), followed by the addition of glutathione reductase (GR) (0.048 U) and GSH (10 mM), incubated at 37 ∘ C/10 min and afterwards, nicotinamide adenine dinucleotide phosphate (NADPH) (1.5 mmol) and tert-butyl hydroperoxide (0.5 mM) were added. The decrease in the absorbance of the system was measured, for 5 min, at 340 nm. One unit of tGPx was defined as the amount of enzyme able to catalyze the oxidation of 1 mol of NADPH to NADP + in 1 min. The result was expressed as U⋅mg protein −1 .
MPO activity was measured according to Bradley et al. [25]. Hepatic or colonic tissue was homogenized using assay buffer pH 6.0 (PBS, 50 mM, 0.5% hexadecyltrimethylammonium bromide and EDTA, 5 mM) and centrifuged at 1550 ×g/15 min (4 ∘ C). Supernatant was collected and centrifuged at 14.000 ×g/15 min (4 ∘ C). The sample was transferred (duplicate) to microplate and ortho-dianisidine solution (0.8 mg/mL) was added. After incubation (37 ∘ C/15 min), a solution of H 2 O 2 (0.3%) was added. A new incubation was performed (37 ∘ C/10 min) and the reading was made at 460 nm. One unit of MPO was defined as the quantity that decomposes 1 mol of H 2 O 2 . The result was expressed as U⋅mg protein −1 .
To measure GSSG, the supernatant was diluted (1 : 50) in assay buffer, containing N-ethylmaleimide (NEM), and centrifuged (10,000 ×g/20 min at 4 ∘ C). This solution was incubated for 50 minutes to complete GSH complexation and removal. To exclude NEM, this supernatant with assay buffer were eluted through Sep-pak5 Classic C18 cartridges. Afterwards, the eluent was transferred to microplate and Reaction Mixture 1 and NADPH,1% w/v, were added.
In both analyses, the absorbance was measured over 3 min at 412 nm with 30 s intervals. GSH was determined according to the following equation: GSH = tGSG − (GSSG/ 2). Results were expressed in nM⋅mg protein −1 .

Results
For the sake of clarity, results are divided into three topics: general, colonic, and hepatic results.

Mild Colitis and Supplementation Did Not Alter Body
Weight, Food Intake, or Liver and Colon Weights. In both phases (PT and T) of the study, DSS or supplementation by antioxidants did not induce effects on body weight or food intake patterns compared to the control group (Figures 1(a) and 1(b)). Similarly, body weight was unchanged over the evaluation period (Figure 1(c)). Absolute and relative liver and colon weights were unchanged (Table 1). Water ingestion modification was also not observed (data not shown).

LA and NAC, but Not LA + NAC, Decreased Anemia and Leukocytosis Caused by Mild Colitis.
Anemia in the mild colitis group was confirmed by a decrease in red blood cells (RBC) and hemoglobin (HB) (Figures 2(a) and 2(b)), since clinical changes such as rectal blood, diarrhea, and weight loss were not observed or did not show differences between the groups. No macroscopic change was observed after euthanasia (data not shown). All treatments increased the parameters (RBC, HB). However, LA and NAC alone ameliorated the typical UC inflammation, represented by a decrease in leukocytes, while the combination of LA + NAC did not show beneficial effects (Figure 2(c)).

LA + NAC Increased Levels of Aminotransferases.
The use of DSS and NAC did not cause a change in the available serological biomarkers, unlike LA supplementation, which promoted a decrease of globulin levels versus the mild colitis group ( < 0.01). However, this biochemical alteration did not exhibit physiological relevance, since albumin, the most important biomarker of hepatic function, remained statistically unaltered among the groups (Table 2). Moreover, it is important to observe that the combined action of LA and NAC on biomarkers of hepatic injury differed from the control ( < 0.05) and NAC ( < 0.05) groups. Compared to the NAC group, ALT and AST in the LA + NAC group were 2.2x and 2x higher, respectively. Despite the fact that these enzymes are not exclusive markers of liver damage, their increase in clinical situations such as heart disease and myopathies, when analyzed together with oxidative ( Figure 7) and inflammatory ( Figure 8) parameters, may be considered as a disruption of the liver metabolism balance. The other systemic biomarkers analyzed were not seen to be statistically significant.

LA or NAC Reduced Histological Damage on the Colon
Induced by DSS. Samples from DSS-treated animals (mild colitis) showed higher histological damage (Figure 3(a)) than other groups. Although all the treated groups showed these changes, damage to the mucosal architecture was reduced when antioxidants were used, showing a protective effect of the antioxidants, when compared to the mild colitis group. Collagen deposition (Figure 3(b)), marked with a blue color, confirms the presence of fibrotic tissue in mild colitis (Figure 3(b)). LA and/or NAC treatments were able to decrease this deposition but were not sufficient to prevent mild UC lesions and their collagen counts were equal to the mild colitis and control groups ( > 0.05) (Figure 3(c)). colonic oxidative damage. In mild colitis, NAC activity may be maintained due to an increase in GSH (Figure 4(g)) and consequently GSSG (Figure 4(h)), as a response attempt of the body to oxidative damage. There were no alterations in the GSH/GSSG ratio (Figure 4(i)) and GPx levels for all groups (Figure 4(j)).

Changes in Intestinal Cytokines Were Not Observed in
Mild UC and LA + NAC Provoked Inflammation. Colonic inflammation represented by proinflammatory cytokines TNF-and INF-, involved in innate immunity, and the antiinflammatory cytokine IL-10, was not altered in the mild colitis group, compared to the control (Figures 5(a), 5(b), and 5(c)). However, LA + NAC promoted an increase in TNF-(versus control, LA, and NAC groups) ( Figure 5(a)) and IL-10 (versus all groups) ( Figure 5(c)). Probably IL-10 increased to minimize the proinflammatory effects caused by TNF-.

Mild Colitis, LA, and/or NAC Did Not Cause Histological
Alterations in the Liver. Biochemical and histological analyses (Table 2) suggest an absence of hepatic injury caused by DSS. However, despite the fact that the association between LA and NAC indicated concerning effects on this tissue, which can be observed by an increase in ALT and AST levels, only one morphological alteration could be identified: a periportal zone with disorganized hepatocyte cords (Figure 6(a)) without a necrosis area or collagen deposition (Figures 6(b) and 6(c)).  (Figure 7(b)).

LA + NAC Caused Inflammation in the Liver.
In the liver, the levels of cytokines were not modified in the mild colitis group compared to the control group (Figures 8(a), 8(b), and 8(c)). However, it is important to notice, in both tissues (colon and liver), the proinflammatory effects of LA associated with NAC (LA + NAC). In hepatic tissue, this combination provoked the increase of TNF-and INFlevels, when compared to all groups (Figures 8(a) and 8(b)).

Discussion
Analyses of histology, oxidative stress, and inflammatory biomarkers were performed on colonic and liver tissues, in order to investigate the role of added antioxidants (NAC and/or LA) in controlling damage caused by DSS. In this context, we observed that redox imbalance was the first alteration in mild colitis; NAC was able to reduce oxidative stress and cell damage, DSS (2% w/w) did not cause hepatic modification, and LA + NAC increased inflammation in the colon and liver.

Colonic Injury and the Effects of Supplementation.
Previous studies have provided compelling evidence for the association between DSS and different degrees of UC [27,28], from mild colitis up to carcinogenesis, according to its continuous administration, at 2-5% w/v, for a short period of time (4-9 days). Moreover, histological, biochemical, and immunological alterations caused by DSS are similar to UC in humans [29]. The exact mechanism of colitis induction by DSS is not known, but it seems to involve dysfunctional macrophages, luminal bacterial alterations, and direct toxicity to epithelium [30]. The redox profile was the first biochemical parameter to be altered, before immunological changes, particularly in    [31,32]. However, LA and/or NAC supplementation could not prevent this damage, including collagen deposition (Figures 3(b) and 3(c)).
Even alone, LA has shown to exhibit negative effects in human studies. Wray et al. analyzing cardiovascular risk in elderly people observed that 600 mg/d of LA + Vit C (1000 mg) and Vit E (600 IU), 3x per week for 6 weeks, nullified positive effects on blood pressure, caused by exercise [33]. Additionally, McNeilly et al. [34], studying cardiovascular risk in obese patients with glucose intolerance, tested 1 g/d of LA for 12 weeks, with or without exercise, and detected no improvement on serum lipid profile and increased oxLDL. Showkat et al. [35] tested LA (600 mg), 30 minutes prior to iron administration in patients with chronic renal failure on hemodialysis, and observed an increase in F2 isoprostanes and lipid hydroperoxide, * * * * * * biomarkers of lipid peroxidation, thereby confirming cell disruption.
LA was able to scavenge H 2 O 2 and decrease LP (MDA levels), but NAC was more effective, completely preventing the increase of these markers. NAC was also shown to be effective in chronic UC, induced by DSS (5% w/v for 5 days), as observed by Amrouche-Mekkioui and Djerdjouri [28]. According to these authors, NAC (150 mg⋅kg⋅d −1 for 45 days) decreased colitis symptoms, inflammation, cell apoptosis, and MPO and • NO levels. Collectively, these results indicate the antioxidant and anti-inflammatory effects of NAC at different stages of UC.
• NO is a reactive molecule associated with UC progression, especially regarding toxic megacolon. In addition, • NO reacts with O 2 •− , forming peroxynitrite, which causes LP and consequent ulcers in the colon mucosa. Both lesions are common in IBD [4]. • NO production is catalyzed by the enzyme nitric oxide synthase (NOS). In inflamed tissue, such as the colon in UC, the inducible isoform of the enzyme (iNOS) is highly expressed in DSS-inflamed colons and the colon of UC patients [36]. This enzyme, present in its own colonocyte, may be responsible for the increase of nitrite levels in the mild colitis group.
F2-isoprostane is the best general indicator of nonenzymatic lipid peroxidation in complex biological systems [37]. However, in a recent systematic review on antioxidant therapy published by our group, we observed that the majority of the studies used MDA (identified directly by HPLC, or indirectly by thiobarbituric acid reactive substances, TBARS) to measure LP. In our study, a MDA assay was chosen.
In the present study, alterations in colonic SOD and GPx (Figures 4(f) and 4(j)) in the mild colitis group were not observed, which is similar to data reported by Akman et al. [38], who studied patients with active intestinal inflammation. On the other hand, the lower SOD activity observed in the NAC versus LA group, without evidence of increased H 2 O 2 levels, confirms the major antioxidant power of NAC in our model. Recently, antioxidant therapy for IBD has gained increased recognition [4]. However, SOD activity has been poorly investigated regarding LA and NAC administration, probably because SOD is an enzyme whose activity is modestly reduced, during tissue inflammation, unlike GPx2, a gastrointestinal-specific form of GPx [36], which is closely associated with H 2 O 2 metabolism, and whose gene expression is downregulated in several experimental models of UC [39].
Our findings on elevated GSH and GSSG levels in the mild colitis group are different from the findings of Amrouche-Mekkioui and Djerdjouri [28]. This increase may be explained by the role played by GSH in inhibiting apoptosis signaling not only by scavenging intracellular ROS but also by inhibiting cytochrome c release from mitochondria and regulating the activity of redox-sensitive caspases [40]. In our results, the influence of oxidative stress on GSH cycling was confirmed by an increase of GSSG. However, the concomitant elevation of GSH levels did not cause a change in the GSH/GSSG ratio, the most important biomarker.
The GI tract is a major site for generation of prooxidants, whose production is primarily due to the presence of a plethora of microbes, food ingredients, and interactions between immune cells [15]. The enhanced production of reactive species is associated with chronic intestinal inflammation in the early stages of IBD. Their destructive effects on DNA, proteins, and lipids may contribute to initiation and progression of UC, causing several symptoms, such as loss of blood and anemia, carcinogenesis, hepatotoxicity, nephrotoxicity, and hypersensitivity [41]. Besides that, oxidative stress increases inflammation and stimulates activation of NF-kB, with consequent production of proinflammatory cytokines, chemokines, growth factors, and adhesion molecules, which cause inflammation and fibrosis, identified in our study by the increase of leukocyte levels and collagen deposition.
TNF-has been shown to play a critical role in the pathogenesis of IBD and biological therapy with TNF--blockers has been used as a mainstream treatment for downregulating aberrant immune responses and inflammatory cascades [42]. Additionally, INF-is involved with overexpression of several chemokines, such as IFN--inducible protein 10 and IFNinducible T-cell chemoattractant, in the intestinal mucosa for colitis induced in mice, and in UC patients [43]. On the other hand, IL-10, an anti-inflammatory cytokine, is required for regulating immune functions by promoting the widespread suppression of immune responses through its pleiotropic effects [44]. Imbalance in the production of these cytokines, such as TNF-, plays a pivotal role in the signaling cascade of inflammatory pathways.
Guijarro et al. [45] in studying the effect of NAC plus mesalamine in UC patients also observed no changes in TNF-plasma levels. It is possible that alterations were not identified because the model used in this study is for mild UC and upregulation of proinflammatory cytokines, such as IFN-and TNF-, was observed more consistently in severe inflammation, such as colitis associated with carcinogenesis.
Unexpectedly, LA + NAC did not promote a beneficial action, even in increasing colonic IL-10 levels, which may be explained by the increase of leukocytes and dendritic cells, the latter responsible for its secretion and that are increased in colonic infiltrates [46]. IL-10 elevation has an anti-inflammatory response, especially in Th2 (T helper 2 lymphocytes) [47], related to autoimmune disorders such as UC [48]. In contrast to other studies [49,50], a colonic increase of proinflammatory cytokines ( Figures 5(a), 5(b), and 5(c)), despite the increase of leukocytes (Figure 2(c)), was not observed in the present study.

Hepatic Injury and the Effects of Supplementation.
The extraintestinal manifestations of IBD are poorly explored by the scientific community. However, recent results have associated these manifestations to IBD activity [51] and the use of TNF-inhibitors [52], exemplified by hepatobiliary manifestations, in terms of frequency and severity [53][54][55]. At the same time, these IBD complications remain underdiagnosed [56].
Despite the intimate connection between liver and colon, the oxidative and inflammatory alterations found in our model of mild colitis (2% of DSS, v/v, for 5 days) could have been insufficient to cause liver damage, unlike the results of Trivedi and Jena [57] and Farombi et al. [58]. However, when histological ( Figure 6(a)), serological (Table 2) evaluated, hepatic damage in the LA + NAC group was identified. Unlike hepatic • NO (Figure 7(d)), H 2 O 2 production was seen to be stimulated by the combination of LA + NAC (Figure 7(f)), while no modification in histology and collagen depositions was observed in this group.
Mitochondria and redox-active enzymes can generate O 2 •− and H 2 O 2 as byproducts in liver cells and these reactive molecules are increased under different conditions of chronic liver injury, caused by alcohol, xenobiotics, viral infections, nonalcoholic fatty liver disease, and others. Additionally, high concentrations of oxidative species, such as H 2 O 2 and • OH, induce hepatic stellate cell death and cause reduction of collagen deposition [59], which would explain the absence of changes in collagen deposition despite the higher levels of ALT and AST found in the serum of the LA + NAC group. Hepatic injury, whether acute or chronic, eventually results in an increase of serum concentrations of aminotransferases [60], suggesting stronger harmful action on the liver tissue, caused by the combination of these two antioxidants.
NAC supplementation led to an increase in hepatic SOD but without increasing H 2 O 2 levels (versus mild colitis and LA), that is, improved redox status by activation of the antioxidant defense system. On the contrary, LA significantly increased CAT, GSH, and GPx, possibly due to increased oxidative stress observed in this group and confirmed by an elevation of MDA. Oxidative stress may be observed by GSSG levels, and although without statistical significance, these levels were approximately 61% higher than in the NAC group (Figure 7(i)).
Relative to lower CAT and GSH levels observed in the NAC group, these levels could be justified by the shorter period of supplementation. NAC protection against oxidative stress occurs by directing cysteine into the GSH synthesis pathway, with a consequent increase on the intracellular GSH content [16].
In a recent review, the authors showed that LA is generally administered associated with other substances and that this multiple therapy impairs the identification of the specific role of each component, raising difficulties in attributing  beneficial, synergistic, or antagonistic effects [18]. In this context, El-Gowelli et al., also using an animal model, observed that LA plus cyclosporine, an immunosuppressant used routinely in UC treatment, aggravates colon damage. Pop-Busui et al. studying patients with type 1 diabetes observed similar noxious effects in diabetic patients, using LA plus allopurinol (xanthine oxidase inhibitor) and nicotinamide whose combination did not prevent the progression of cardiovascular autonomic neuropathy. It is important to emphasize that the LA + NAC group received 200 mg⋅kg⋅d −1 (100 mg⋅kg⋅d −1 of each antioxidant), an amount much lower than the maximum limit established for safety in the oral delivery of LA (10x less) [61] and NAC (30x less) [62]. Taken together, these results cast doubt on the concept of the "universal antioxidant" given to LA.

Conclusions
In our study, oxidative stress was the first biochemical manifestation of mild UC and happened before an increase in TNF-and INF-. Additionally, NAC exhibits better antioxidant effects, especially regarding MDA and H 2 O 2 levels. LA, administered daily, as a single dose increased hepatic MDA. LA + NAC increased oxidative and inflammatory profiles in the colon and liver (Figure 9).
In summary, our work provides evidence that the antioxidant and anti-inflammatory power of NAC involves not only the colon but also the liver. This fact confirms, for UC, the necessity to broaden the investigation to the liver, which is intimately connected to the colon.
The management of UC with alternative therapies is a large field of investigation and experimental colitis must mimic the human disease spectra. As presently shown, investigation on several tissues and organs is necessary, before a definite choice of a treatment can be made.