**3. Results and Discussion**

#### *3.1. BG-4 Exert Antioxidant and Antiradical Activity*

Antioxidant capacity of BG-4 was estimated as 4.72 ± 0.41 μM of Trolox Equivalent per gram, being comparably higher to that reported for protocatechuic acid and protocatechuates (<3 μM of Trolox Equivalent/g) [20]. The ability of BG-4 to convert DPPH radical into it picrylhydrazine form increase about 1.6 times when doubling the concentration, as it can be seen in Figure 1, exerting up to 26% inhibition at a concentration of 0.5 mg/mL of BG-4.

**Figure 1.** Radical scavenging activity of BG-4 calculated as % of inhibition of DPPH free radical for different concentrations. Results are presented as mean ± SD.

#### *3.2. BG-4 Decreased Expression of Pro-Inflammatory Markers In Vitro*

Macrophages are part of the first line of defense against injury, mediating the innate non-specific immune response via inflammation, and playing an important role not only in host defense but in tissue homeostasis, repair, and pathology development including IBD [21,22]. RAW 264.7 mouse derived macrophages have been widely used as in vitro model to study modulatory effects of several compounds using LPS for activation [23–25]. Treatment with BG-4 at concentrations up to 375 μg/mL did not affect the viability of macrophages (Figure 2A). It is known that LPS can activate NF-κB which acts as a master regulator of the inflammatory response by promoting the release of signaling and effector molecules such as pro-inflammatory cytokines (IL-6, TNF-<sup>α</sup>, IL-1β) and nitric oxide (NO) that act as mediators of inflammation during the host immune response [6]. The destruction of normal and healthy tissue as a result of chronic inflammatory condition can lead to IBD, rheumatoid arthritis and multiple sclerosis [26], hence reducing the level of these pro-inflammatory secreted molecules can lead to managemen<sup>t</sup> and alleviation of these diseases. BG-4 dose-dependently reduced the production of NO (Figure 2B) and IL-6 (Figure 2C) in LPS-activated RAW 264.7 macrophages. Moreover, the expression of iNOS and COX-2 (Figure 2D) was reduced by BG-4.

**Figure 2.** Effect of BG-4 on viability of activated lipopolysaccharide (LPS+) and non-activated (LPS-) macrophages (**A**), NO production (**B**), cytokine production (**C**), and iNOS and COX-2 expression (**D**), in vitro. Results are presented as mean ± SD for (**A**–**C**) and mean ± SE for (**D**). Different letters indicate significant differences (*p* < 0.05) among treatments.

These results sugges<sup>t</sup> that the mechanistic effect of BG-4 had a major impact on molecules particularly involved in the pathway associated with the specific expression of iNOS and IL-6, and consequently over their downstream products, but affected in a lower extent the expression of COX-2 and TNF-<sup>α</sup>. Hence, we selected 375 μg/mL concentration to study the ability of BG-4 to ameliorate DSS-induced colitis in mice with a translation of IP administration of BG-4 at 15 mg/kg bw assuming an average mouse weight of 25 g and a circulating blood volume of one milliliter. Two stages of DSS administration were performed in order to induce damage to the epithelial barrier in the colon to simulate colitis, allowing a period of recovery in between as presented in Figure 3A. This experimental design mimics the periods of remission and relapse that IBD patients usually experience [16]. BG-4 administration started at three days post DSS to mimic ingestion of BG-4 as a supplement for ameliorating established IBD (versus prevention) in human as there are no known risk factors to IBD. Hence, it is highly likely that dietary supplement will be consumed for managemen<sup>t</sup> and not for the prevention of IBD.

#### *3.3. IP Administration of BG-4 did not Alleviate Indicators of Colitis in Vivo*

The damage in the colonic mucosa that occurs in ulcerative colitis results in different symptoms and indicators such as diarrhea, presence of blood in the stool, increased frequency of bowel movements, mucus discharge, abdominal pain, weight loss, bloody diarrhea, and rectal bleeding irritation and fissures [27]. Daily monitoring of body weight, presence of visible blood and stool consistency were performed, in order to establish variation of external indicators of the disease. CG maintained a constant incremental increase in weight while the DSS group presented a significant decrease starting at day five, and minimum trough at day seven, with a 6% loss of original weight. BG-4 administration in DSS-treated mice showed a continued loss in weight until day nine and in days 14 and 15 with a 10% decrease in body weight at euthanasia as shown in Figure 3A. For CG, no visible blood in the stool

or anus and normal stool consistency were observed with a disease score of zero (Figure 3B) neither DSS nor BG-4 groups presented watery or bloody diarrhea, but changed in stool consistency, becoming much softer in comparison to CG group and BG-4 administration led to stool softening earlier than DSS group as shown in Figure 3B. Presence of blood in stool was observed for DSS and BG-4 groups mostly between days 3–6 and 10–14, but the frequency of mice presenting blood in the anus was higher for BG-4 than for DSS group as presented in Figure 3C.

**Figure 3.** Animal study design and effect of BG-4 treatment over relative weight (**A**), stool consistency (**B**), and presence of visible blood (**C**) on dextran sodium sulfate (DSS)-induced colitis in mice. Scores for stool consistency and presence of visible blood for CG were 0 for every observation. Results are presented as mean ± SE. Different letters indicate significant differences (*p* < 0.05) among treatments.

Overall, external indicators of the disease sugges<sup>t</sup> that DSS administration induced colitis with a mild grade of severity in this study. Since diarrhea or reduction in food intake were not observed, loss of body weight might be due to an increased frequency in bowel movements with a softened stool consistency and presence of blood, that upon removal of DSS was compensated by food intake, allowing partial or total recovery of the animals. However, the lower extent of recovery for BG-4 group, in addition to the higher frequency of animals with rectal bleeding may indicate that IP administration of BG-4 aggravated symptoms of the disease.

#### *3.4. IP Administration of BG-4 Aggravated DSS-Induced Colitis in Mice*

Colon shortening has been reported for several studies as an outcome of colitis in animal models [6,28–30]. BG-4 group presented a significant reduction in average colon length of about 2 cm in comparison with CG as shown in Figure 4A. H&E staining allows a comparison of the effects of DSS challenge and BG-4 treatment over the internal structure of colonic tissue. Micrographs of stained colonic tissue for the three groups show that animals under DSS-induced colitis exhibited immune cells infiltration (black arrows) in the lamina propria, and a distortion of crypts architecture (white arrows) accompanied by the loss of goblet cells (Figure 4B). In the case of BG-4, infiltration was extended to the muscularis region and to form cryp<sup>t</sup> abscesses and reach the glandular lumen of the crypts forming abscesses as shown in Figure 4B.

**Figure 4.** Colon length (**A**), H&E staining of colonic tissue (**B**), MPO activity (**C**), and fecal hemoglobin (**D**) on DSS-induced colitis in mice. Results are presented as mean ± SD (**A**) and (**C**) or SE (**D**). Different letters and \* indicate significant differences (*p* < 0.05) among treatments. Immune cells infiltration and distortion of the cryp<sup>t</sup> architecture are indicated by black and white arrows, respectively.

MPO is a peroxidase enzyme that plays an important role during infections by modulating the production of hypochlorous acid and other non-selective reactive species that mediate pathogens killing, and can induce host tissue damage by promoting inflammation [31]. It is mainly present in specialized granules of neutrophils hence elevated MPO levels and neutrophil infiltration have been correlated to inflammatory autoimmune disorders [32]. DSS and BG-4 groups exhibited an MPO activity four and 14 times higher, respectively, than CG (Figure 4C), corroborating observations in H&E staining micrographs where a higher infiltration of immune cells was detected for both groups with elevated severity upon BG-4 treatment.

Quantification of hemoglobin in collected feces was used to estimate fecal occult blood. Figure 4D indicates the presence of hemoglobin in feces for BG-4 and DSS groups followed the same trend as loss in weight. At the end of the study, the presence of hemoglobin in feces from BG-4 group was significantly higher than in DSS group, being consistent with the observations and scoring for the visual presence of blood in stool or rectum presented in Figure 3B,C.

#### *3.5. IP Administration of BG-4 Exerted a Di*ff*erential Systemic and Localized E*ff*ect in Pro-Inflammatory Cytokines In Vivo*

Cytokines are signaling proteins that contribute to the modulation of the immune and inflammatory response during tissue damage and repair. The levels of pro-inflammatory cytokines IL-6, TNFα, and IL-1β in colonic extracts (A) and in serum (B) are presented in Figure 5. No significant e ffect was observed upon DSS challenge and/or BG-4 treatment over the expression of IL-6 and TNFα in colonic extracts, but both compounds were upregulated in serum in the DSS group and attenuated under BG-4 treatment (Figure 5A,B). IL-1β, in contrast, was upregulated in the DSS group and attenuated in the BG-4 group in colonic extracts but remained high in serum (Figure 5A,B). These results sugges<sup>t</sup> that DSS and BG-4 were capable to exert a systemic (in circulating blood) but not localized (in colonic tissue) effect. A recent study has demonstrated dysregulation of pro-inflammatory cytokines in DSS-induced colitis mouse model with similar di fferential outcomes at the serum and colonic level observed in our study [33]. The particular di fferences in the systemic and localized e ffect over cytokines seem to indicate that IP injection allowed BG-4 to reach blood circulation, as expected for the high bioavailability reported for this administration route but post the question if the appropriate concentration was, in fact, capable to reach the intestinal barrier, where the tissue damage was induced via DSS and the inflammatory response was concentrated.

**Figure 5.** Effect of BG-4 on the level of pro-inflammatory cytokines in colonic extract ( **A**) and serum (**B**). Results are presented as mean ± SE. Di fferent letters indicate significant di fferences (*p* < 0.05) among treatments.

Our results showed that BG-4 exerted a differential effect over inflammation in vitro and in vivo. In LPS-stimulated macrophages, treatment with BG-4 showed a dose-dependent reduction of NO, IL-6, COX-2, and iNOS, suggesting 375 μg/mL as an optimal concentration for significant reduction of inflammatory markers. However, IP administration of BG-4 in a DSS-induced colitis mouse model, aggravated disease symptoms when compared with untreated mice, accompanied by the dysregulation of pro-inflammatory cytokines. The systemic circulation of BG-4 due to IP administration involved a possible effect over other immune and epithelial cells that can be more or less sensitive, in terms of cytotoxicity [34] and alteration of signaling and regulatory pathways, than the tested macrophages. For instance, it has been recently demonstrated the induction of NO metabolism in enterocytes alleviates colitis and CACC [29], suggesting a controversial role of NO in the regulation of the inflammatory response associated with its origin. According to this perspective, suppression of NO production exerted by BG-4 could result in a counteracting effect by simultaneously suppressing NO production on macrophages and enterocytes.

The outcomes observed in vivo under BG-4 treatment sugges<sup>t</sup> that the mechanistic effect by which BG-4 suppresses NO production may be associated with enterocyte than macrophages. Under this circumstance, the reduced expression of NO by the enterocyte allows neutrophils infiltration, generation of cytotoxic compounds by MPO, tissue damage, and a consequently increased inflammatory response. One strength of our study is the first report on the comparative anti-inflammatory properties of BG-4 peptide from *Momordica charantia* using both in vitro and in vivo models. In addition, both mouse cell line and mouse model of colitis were utilized in the study. On the other hand, the observed contrasting result of in vitro and in vivo models on the anti-inflammatory effect of BG-4 is a potential weakness. The intraperitoneal injection of BG-4 may have resulted in the toxic effect of BG-4 in mouse model resulting in its pro-inflammatory effect which is contradictory in the observed anti-inflammatory effect in vitro. Hence the next step on the potential application of BG-4 for managemen<sup>t</sup> of inflammatory diseases must be observed via oral route to determine if it can be used as a dietary supplement to mitigate diseases such as colitis.
