Pterostilbene and Probiotic Complex in Chemoprevention of Putative Precursor Lesions for Colorectal Cancer in an Experimental Model of Intestinal Carcinogenesis with 1,2-Dimethylhydrazine

Simple Summary The composition of the intestinal microbiota, chronic inflammation, and oxidative stress are factors related to the onset of colorectal cancer (CRC). Diet is the environmental factor most related to the development of non-hereditary CRC. Pterostilbene has great potential as an antitumor drug for CRC chemoprevention. Several strains of probiotics in multiple combinations, concentrations, and dosages have been studied for cancer prevention. The available evidence is insufficient to justify the chronic and simultaneous administration of pterostilbene and probiotics with Lactobacillus and Bifidobacterium to a population at high risk of CRC. In vivo studies may contribute to the incorporation of dietary supplementation with the ability to reduce the incidence of CRC. The combination of substances should be stimulated with a view to potentiating the expected effect through action on different targets. Abstract Dietary supplementation with pterostilbene (PS) and/or a probiotic (PRO) may ameliorate the intestinal microbiota in disease conditions. This study aims to evaluate PS and PRO for the chemoprevention of putative precursor lesions for colorectal cancer (CRC) in an experimental model of intestinal carcinogenesis with 1,2-dimethylhydrazine (1,2-DMH). Sixty male Wistar rats were equally divided into five groups: Sham, 1,2-DMH, 1,2-DMH + PS, 1,2-DMH + PRO, and 1,2-DMH + PS + PRO. PRO (5 × 107/mL) was offered in water, and PS (300 ppm) was provided in the diet ad libitum. 1,2-DMH (20 mg/kg/week) was administered for 15 consecutive weeks. In the 25th week, proctocolectomy was conducted. PRO alone and PRO combined with PS were the best intervention strategies to improve experimental 1,2-DMH-induced CRC regarding several parameters of carcinogenesis. Our findings may contribute to the development of novel preventive strategies for CRC and may help to identify novel modulators of colon carcinogenesis.


Introduction
Colorectal cancer (CRC) is the second neoplasm with the highest number of potentially preventable deaths, behind lung neoplasms. About 45.1% of cancer-related deaths in the of therapy, and the method of supplementation [30]. Although probiotics are attractive agents for the prevention of CRC, their effectiveness has not yet been fully established, and more studies need to be carried out [31].
Our study aims to evaluate the isolated or combined effect of PS and PRO supplementation on the CRC tumorigenesis markers, OS reduction, and histopathological scores after 1,2 -DMH-induced CRC. The potential protective effect of PRO and PS in preventing precursor lesions for CRC would be particularly beneficial for high-risk patients.

Animals
The sample consisted of 60 male Wistar rats (Rattus norvegicus albinus) from the Christus University Center (UNICHRISTHUS) vivarium, weighing 80 ± 10 g. The study protocol was approved by the Ethics and Animal Research Committee (nº 031/ 2018). The rats were kept in polypropylene cages in a temperature-controlled environment (22 ± 1 • C) with a 12-h light/dark cycle, free access to drinking water, and a standard chow diet. Animals that showed any signs of illness or died were excluded from the study and replaced.
Eight capsules with 150 mg of PS (Mental Refreshment, EUA) were mixed with 4 kg of the standard chow diet (Nuvilab CR-1, Quintia, BRA) to produce a diet with 300 ppm of PS or 3 mg of PS/10 g of diet. The PS molecular weight was 256.3 g/mol and the PS diet concentration was 1.17 mmol/kg. A rat weighing 200 g consumes on average 10 to 20 g of the diet with 3 to 6 mg of PS/day or 15 to 30 mg of PS/Kg/day. Tests were performed to verify the presence of PS in the rats' diet and liver (Supplementary Materials).

Experimental Design
The rats were randomly divided into five groups of 12 animals (control group or Sham, cancer group or 1,2-DMH, 1,2-DMH + PS, 1,2-DMH + PRO, and 1,2-DMH + PS + PRO). The rats in the experimental group received a weekly subcutaneous injection of 1,2-DMH (D161802; Sigma-Aldrich, St. Louis, MO, USA) at a dose of 20 mg/kg body weight for 15 weeks [32]. The carcinogen was dissolved in 0.9% NaCl (pH 6.5) and the control group received the equivalent dose of 0.9% NaCl without adding the carcinogen.
After dividing the groups, solutions with PRO (in the water) and/or PS (in the diet) were administered ad libitum from the first day of the research until euthanasia during the 25th week ( Figure 1).

Surgical Procedure and Sample Preparation
At the end of 25 weeks, the animals were anesthetized with 10% ketamin hydrochloride (80 mg/kg/weight) and 2% xylazine hydrochloride (10 mg/kg/weight). Th

Surgical Procedure and Sample Preparation
At the end of 25 weeks, the animals were anesthetized with 10% ketamine hydrochloride (80 mg/kg/weight) and 2% xylazine hydrochloride (10 mg/kg/weight). The animals were positioned in dorsal decubitus on a wooden board and immobilized by fixing their limbs. Then, a laparotomy and proctocolectomy were performed. Subsequently, the surviving animals were sacrificed by hypovolemic shock after a section of the abdominal aorta was cut. Confirmation of death occurred by verifying the absence of respiratory movements (apnea), heartbeat (asystole), and pulse.
The proctocolectomy product was opened longitudinally at the antimesenteric border for intestinal lavage and extended with the exposed mucosa. After macroscopic evaluation and division of the colon into three equal segments, the specimens were randomly stored in a 10% buffered formalin solution (n = 30) and in a freezer at −80 • C (n = 30). The samples separated for histology and immunohistochemistry were embedded in paraffin using a conventional method and stained with hematoxylin and eosin (H&E). Then, the paraffin blocks were used to make new slides that were stained with methylene blue (MB) at a concentration of 0.1% [33]. In the frozen samples, the distal segments of the colon were used to measure oxidative stress.
To evaluate the slides stained with MB, 10 fields per bowel segment (distal, middle, and proximal) were photographed randomly at 400× magnification. The crypts were analyzed in cross-sections, and the factors analyzed were the number of aberrant crypt foci (ACF) and the location in the colon (distal, middle, and proximal). ACF was considered when the crypts had at least two criteria: an increased crypt size, a thicker epithelial layer, more intense staining (due to nuclear increase and mucin depletion) [33], an increased pericrypt zone, an elliptical shape [35], and a reduction of goblet cells greater than 50% [36]. The ACF were not classified as hyperplastic and dysplastic.

Immunohistochemistry by the Tissue Microarray Technique
Six cylindrical distal colon fragments were collected from each rat from paraffin blocks. The material was collected with a 2 mm-diameter needle (Quick-Ray UNITMA ® , Seongnam-si, Republic of Korea) to offer good sampling, ease the construction of the recipient block, and avoid damage to the donor block. Then, the material was included in three paraffin blocks with 70 wells. In the same blocks, tissues were included to serve as a positive control for immunohistochemical reactions. Sequential 3 µm-thick sections of the tissue microarray block were deposited on silanized glass slides for conventional H&E staining and immunohistochemistry reactions.
The images were captured using a light microscope coupled to a camera with a LAZ 3.5 acquisition system (Leica DM1000, Wetzlar, Germany). Ten fields were photographed per histological section (40× objective), trying to select the areas with the highest marking in each animal (hot areas). For counting positive cells marked by each field, the adobe photoshop 8.0 program was used to obtain the total tissue area and the immunostained area. Positive cells were considered with brown staining within the cytoplasm for iNOS, NF-kB, IL1-β, TNF-α, and Wnt3a and within the nucleus for Ki67 and both for P53 and β-catenin. The most evident labeling of the protein was considered, avoiding the background. Then, to measure the percentage (%) of the marked area, the following calculation was performed: marked area (%) = immunomarked area (pixels) × 100/total area (pixels) [39].

Oxidative Stress Markers
Samples from the distal segment of the large intestine were thawed and homogenized in cold EDTA (0.02 M) or KCL (0.15 M) to prepare a 10% homogenized suspension and estimate glutathione (GSH) and malondialdehyde (MDA) levels. Tissue GSH levels were estimated by the Sedlak method [40], with minor modifications. Approximately 100 µL aliquots of tissue homogenate were mixed with 80 µL of distilled water and 20 µL of trichloroacetic acid (50%, w/v) and centrifuged at 4500 rpm for 15 min. The supernatant (100 µL) was mixed with 200 µL of TRIS buffer (0.4 M, Ph 8.9) and 10 µL of 5,5'-dithiobis (2-nitrobenzoic acid) (DTNB, Sigma-Aldrich, USA). The absorbance of GSH was read at 412 nm using a control reagent (without the homogenate). The concentration was expressed in mg/g of tissue.
To determine the level of MDA in the tissues, the 2-thiobarbituric acid assay was used, which assays the level of lipid peroxidation (LP) in biological samples [41]. Aliquots of 125 µL of tissue homogenized were mixed with 750 µL of 1% H 3 PO 4 and 250 µL of 0.6% 2-thiobarbituric acid and incubated for 1 h in a bath at 100 • C. Then, the solution was cooled on ice for 20 min, and 1 mL of n-butanol was added. The mixture was centrifuged (2000 rpm, 15 min at 4 • C), and 100 µL of supernatant was added to 96-well plates to read the absorbance at 535 nm, using a control reagent (without the homogenate). The MDA concentration was expressed in nmol/mg tissue.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism software, version 6.0. To assess normality, the Shapiro-Wilk test was used. Parametric data were evaluated by the one-way ANOVA test of variance followed by Tukey's multiple comparisons test, and non-parametric data were analyzed by the Kruskal-Wallis test followed by Dunns' multiple comparisons test. The significance level adopted was 0.05 (α = 5%), and descriptive levels (p) lower than this value were considered significant. All quantitative results were expressed as mean ± standard error of the mean (SEM), except for histopathological score data and the number of ACF that were presented as median, minimum, and maximum values. Figure 2 shows the representative images of colon sections stained with H&E and MB. In the Sham group, normal crypts were observed, showing preserved goblet cells and enterocytes and the absence of inflammatory infiltrate and edema in the mucosa, submucosa, and muscle layer. In the 1,2-DMH group, a marked loss of tissue architecture and a Cancers 2023, 15, 2401 6 of 16 reduction in goblet cells were seen. In addition, increased infiltration of inflammatory cells, pericrypt zone enlargement, and highly stained crypts were observed. The 1,2-DMH + PS group did not show significant improvements in histopathological scores. The PRO (1,2-DMH + PRO and 1,2-DMH + PS + PRO) groups showed more preserved goblet cells and reduced inflammatory processes. However, ACF were also observed in the PRO groups ( Figure 2I,J). Figure 2 shows the representative images of colon sections stained with H& MB. In the Sham group, normal crypts were observed, showing preserved goblet ce enterocytes and the absence of inflammatory infiltrate and edema in the mucosa, s cosa, and muscle layer. In the 1,2-DMH group, a marked loss of tissue architecture reduction in goblet cells were seen. In addition, increased infiltration of inflamm cells, pericrypt zone enlargement, and highly stained crypts were observed. The 1,2 + PS group did not show significant improvements in histopathological scores. Th (1,2-DMH + PRO and 1,2-DMH + PS + PRO) groups showed more preserved gobl and reduced inflammatory processes. However, ACF were also observed in th groups ( Figure 2I,J).  Table 1 shows the average of ACF identified by field in the three segments of th intestine. In the large intestine middle and distal segments, the 1,2-DMH group sho higher mean ACF per field compared to the Sham group. In the distal colon segm reduction in the number of ACF per field was found in the 1,2-DMH + PRO group in the medial colon segment, the reduction of ACF per field was seen in the grou DMH + PRO and 1,2-DMH + PS + PRO. The 1,2-DMH + PS group was not able to the number of ACF per field, in any colonic segment, compared to the 1,2-DMH The substances used were not able to reduce the number of ACF per field in the pr  Table 1 shows the average of ACF identified by field in the three segments of the large intestine. In the large intestine middle and distal segments, the 1,2-DMH group showed a higher mean ACF per field compared to the Sham group. In the distal colon segment, a reduction in the number of ACF per field was found in the 1,2-DMH + PRO group, while in the medial colon segment, the reduction of ACF per field was seen in the groups 1,2-DMH + PRO and 1,2-DMH + PS + PRO. The 1,2-DMH + PS group was not able to reduce the number of ACF per field, in any colonic segment, compared to the 1,2-DMH group. The substances used were not able to reduce the number of ACF per field in the proximal segment of the colon. The highest number of ACF per field was observed in the distal large intestine segment.

Results
The 1,2-DMH group had a higher inflammatory score than the control group. In the middle and distal segments of the colon, the reduction in the inflammatory score was evident in the 1,2-DMH + PRO and 1,2-DMH + PS + PRO groups. The reduction in the inflammatory score was more intense in the distal segment of the colon. The 1,2-DMH + PS group was not able to reduce the inflammatory score, in any colonic segment, in relation to  (Table 2). Table 2. Histopathological score of inflammation in the three colonic segments. The scores are expressed as the median, minimum, and maximum values, where # p < 0.05 vs. Sham and * p < 0.05, ** p < 0.01 vs. 1,2-DMH group. The data were analyzed using the Kruskal-Wallis test followed by the Dunn's post-test.

Intestinal
MDA levels were higher in the 1,2-DMH group compared to the Sham group. 1,2-DMH + PRO and 1,2-DMH + PS + PRO supplementation significantly reduced MDA levels in the distal large intestine. In addition, GSH levels were significantly reduced in the 1,2-DMH group, an effect that was improved by either compounded 1,2-DMH + PS + PRO supplementation ( Figure 3). The 1,2-DMH group had a higher inflammatory score than the control group. In the middle and distal segments of the colon, the reduction in the inflammatory score was evident in the 1,2-DMH + PRO and 1,2-DMH + PS + PRO groups. The reduction in the inflammatory score was more intense in the distal segment of the colon. The 1,2-DMH + PS group was not able to reduce the inflammatory score, in any colonic segment, in relation to the 1,2-DMH group. The tested compounds were not able to reduce inflammatory scores in the proximal segment of the colon (Table 2). Table 2. Histopathological score of inflammation in the three colonic segments. The scores are expressed as the median, minimum, and maximum values, where # p < 0.05 vs. Sham and * p < 0.05, ** p < 0.01 vs. 1,2-DMH group. The data were analyzed using the Kruskal-Wallis test followed by the Dunn's post-test.

Intestinal
MDA levels were higher in the 1,2-DMH group compared to the Sham group. 1,2-DMH + PRO and 1,2-DMH + PS + PRO supplementation significantly reduced MDA levels in the distal large intestine. In addition, GSH levels were significantly reduced in the 1,2-DMH group, an effect that was improved by either compounded 1,2-DMH + PS + PRO supplementation ( Figure 3).  The immunostaining of iNOS, NF-kB, TNF-α, and IL1β proteins is present in all groups of the experiment ( Figure 4A). A statistically significant difference was found between the 1,2-DMH group and the Sham group regarding the expression of iNOS, NF-kB, TNF-α, and IL1β proteins. TNF-α and IL-1β immunolabeling in the treatment groups was not different from the Sham. The immunoexpression of iNOS and NF-kB proteins was reduced in the 1,2-DMH + PRO and 1,2-DMH + PS + PRO groups compared to the 1,2-DMH group. The 1,2-DMH + PS group was not able to significantly alter the expression of iNOS, and NF-kB proteins in relation to the 1,2-DMH group ( Figure 4B,C). The immunoexpression of TNF-α and IL1β proteins was reduced in the 1,2-DMH + PS, 1,2-DMH + PRO. and 1,2-DMH + PS + PRO groups compared to the 1,2-DMH group ( Figure 4D,E).
TNF-α, and IL1β proteins. TNF-α and IL-1β immunolabeling in the treatment groups was not different from the Sham. The immunoexpression of iNOS and NF-kB proteins was reduced in the 1,2-DMH + PRO and 1,2-DMH + PS + PRO groups compared to the 1,2-DMH group. The 1,2-DMH + PS group was not able to significantly alter the expression of iNOS, and NF-kB proteins in relation to the 1,2-DMH group ( Figure 4B,C). The immunoexpression of TNF-α and IL1β proteins was reduced in the 1,2-DMH + PS, 1,2-DMH + PRO. and 1,2-DMH + PS + PRO groups compared to the 1,2-DMH group ( Figure 4D,E). The immunostaining of Ki67, P53, β-catenin, and Wnt-3a proteins is more evident in the intestinal crypts and present in all groups of the experiment ( Figure 5A). There is a statistically significant difference between the 1,2-DMH group and the Sham group regarding the expression of Ki67, P53, β-catenin, and Wnt-3a proteins. The immunoexpression of Ki67 and P53 proteins was reduced in the 1,2-DMH + PS, 1,2-DMH + PRO, and 1,2-DMH + PS + PRO groups in relation to the 1,2-DMH group ( Figure 5B,C). The immunoexpression of β-catenin and Wnt-3a proteins was reduced in groups 1,2-DMH + PRO and 1,2-DMH + PS + PRO in relation to group 1, 2-DMH. The 1,2-DMH + PS group was not able to significantly alter the expression of β-catenin and Wnt-3a proteins in relation to the 1,2-DMH group ( Figure 5D,E). The immunostaining of Ki67, P53, β-catenin, and Wnt-3a proteins is more evident in the intestinal crypts and present in all groups of the experiment ( Figure 5A). There is a statistically significant difference between the 1,2-DMH group and the Sham group regarding the expression of Ki67, P53, β-catenin, and Wnt-3a proteins. The immunoexpression of Ki67 and P53 proteins was reduced in the 1,2-DMH + PS, 1,2-DMH + PRO, and 1,2-DMH + PS + PRO groups in relation to the 1,2-DMH group ( Figure 5B,C). The immunoexpression of β-catenin and Wnt-3a proteins was reduced in groups 1,2-DMH + PRO and 1,2-DMH + PS + PRO in relation to group 1, 2-DMH. The 1,2-DMH + PS group was not able to significantly alter the expression of β-catenin and Wnt-3a proteins in relation to the 1,2-DMH group ( Figure 5D,E).  groups. The graphs (B-E) represent the mean ± SEM of the percentage of the immunopositive area for Ki67, P53, β-catenin, and Wnt-3a in relation to the total area. # p < 0.05 vs. Sham and * p < 0.05, ** p < 0.01, *** p < 0.001 vs. 1,2-DMH group. (B,C) The data were analyzed using the Kruskal-Wallis test followed by the Dunn's post-test. (D,E) The data were analyzed by the one-way ANOVA test followed by the Tukey's post-test.

Discussion
The concept of cancer prevention is to delay, regress, or eliminate precancerous lesions. The ACF number can be used as a reliable biomarker of preneoplastic and cancerous lesions in the large intestine [42]. This information contributes to developing different preventive strategies for CRC and helps in identifying modulators of colon carcinogenesis [35]. As expected, our study confirmed a higher number of ACF in the 1,2-DMH-induced CRC group compared to the Sham rats. The PRO and PS-PRO combination improved the number of microscopic lesions, including the number of ACF.
The histopathological inflammation scoring by MacPherson and Pfeiffer has been used in experimental research to quantify the intensity of the acute inflammatory process of the colon [34]. However, we showed an interesting association between the number of ACF and the intensity of the inflammatory process after 1,2-DMH-induced CRC. Our data indicated that both PRO and the PRO and PS combination improved the inflammatory score, indicating that this score is a valuable tool to assay CRC histopathological remediation.
Furthermore, our data support prolonged inflammation being associated with higher levels of ROS and CRC severity [43]. GSH is one key antioxidant of the tissue arsenals for balancing OS. Importantly, reduced levels of GSH in the 1,2-DMH-induced CRC were found in our study and were closely associated with increased intestinal MDA (a surrogate marker of OS). Zińczuk et al. observed a reduction in GSH levels and an increase in MDA levels in the blood of patients with CRC compared to healthy controls [44]. ACF from CRC patients were compared to healthy tissues, confirming a pro-oxidative crypt milieu. Only PRO and PS combined supplementation could improve intestinal GSH and MDA levels in our cancer-challenged rats. Of note, B. lactis A6 was found to attenuate OS by lowering MDA levels and increasing GSH levels in colonic tissues. It may also attenuate the inflammatory response via the downregulation of TNF-α, IL-1β, and IL-6 in colonic tissues [45]. Intestinal MDA levels have been consistently related to tumor invasion depth and lymph node metastasis [44].
Our data showed that PS and PRO alone or in combination improved TNF-α and IL-1β immunolabeling, two critical pro-inflammatory cytokines related to CRC pathogenesis [46], in the colon tissue compared to the 1,2-DMH-challenged controls, reaching the level of the sham group. Sustained high levels of TNF-α and IL-1β may precede putative CRC precursor lesions [47].
Although our findings with PS improved Ki67, p53, IL-1β, and TNF-α intestinal tissue expression, it did not improve Wn3a and β-catenin expression. Interestingly, a low dose of PS (40 ppm) in the diet of rats for 45 weeks lessens colon tumorigenesis with reduced Wnt/β-catenin signaling and with cyclin D1 expression inhibition (a well-known Wnt downstream effector) [48]. This discrepancy may be due to different time periods of PS supplementation between both studies. Corroborating our data, PS was also found to reduce the expression of proinflammatory cytokines TNF-α (by 51%) and IL-1β (by 47.7%) in mucosal scrapings derived from the colon of the Azoxymethane (AOM)-injected rats [48].
Interestingly, we found some P53-positive cells showing cytoplasmatic labeling, especially within the colonic crypt cells. Jansson and colleagues have highlighted that P53 cytoplasmatic accumulation could indeed be related to tumorigenesis and that both nuclear and cytoplasmatic expression of P53 could be associated with poor cancer prognosis [49]. Increased P53 colonic expression is an early genetic event in the process of CRC tumorigenesis. P53 mutations have been detected in the colonic mucosal tissue even with a negative or low-grade dysplasia and later surrounding regions of high-grade dysplasia and carcinoma. The mutation of a single P53 allele leads to tumoral growth advantage, with tumor cell clonal expansion with disrupted cellular DNA repair mechanisms [50]. It is recognized that such mutations may occur in the absence of any morphological changes [51]. Neoplasia may arise within different populations of cells in separate areas of the same colon [52].
Cytoplasm and nuclear β-catenin expression is a useful marker for premalignant lesions of rat colon carcinogenesis [53,54]. In our study, we found β-catenin immunolabeling in colonic crypts, even in the ones with no clear ACF alteration. Early cytoplasmic or nuclear localization of ß-catenin may be a meaningful proxy of CRC precursor lesions that may corroborate with the more laborious genetic analyses [55].
Notably, a diet with added PS (40 ppm) for eight weeks was found to suppress ACF (57% inhibition) and reduce iNOS expression [56]. In another study, when PS was added to the diet (50 or 250 ppm) of mice for 6 to 23 weeks, PS reduced the number of ACF which was more pronounced when the supplementation was given for a longer time, with diminished iNOS and COX-2 levels and declines of Wnt/β-catenin signaling, VEGF, and cyclin D1 expression, along with high apoptosis rates in the colon [57]. Previously reported studies had intestinal carcinogenesis induced with the carcinogen AOM. There are no reports of experimental studies with the addition of PS in the diet to prevent the development of CRC.
PS was recognized to modulate IM with anti-inflammatory effects in an experimental study with dextran sodium sulfate (DSS)-induced colitis. PS attenuated the severity of colitis by reducing IL-2 and IL-6 and shifted the IM composition toward a healthier profile, increasing the concentration of Bifidobacterium and reducing the concentration of harmful bacteria in the gut [58].
Our beneficial findings with PS could be mainly linked to pinostilbene, the primary metabolite of PS in the colon. Pinostilbene may play important roles in the anti-colon cancer effects elicited by orally administered PS [59]. Pinostilbene bio-availability may be influenced by the IM through demethylation [60]. PS-deconjugated metabolites are biologically more active than conjugated ones [61]. The combination of PRO with PS may favor a composition of microbiota that facilitates the metabolization of pinostilbene, thus improving its anti-cancer effects that may not be seen with PS alone.
In support of our data, the chemopreventive effects of PRO Dahi (20 g/day) with two types of Lactobacillus (2 × 10 9 CFU/g Lactobacillus acidophilus LaVK2 and Lactobacillus plantarum Lp9) have been highlighted following intestinal carcinogenesis (IC) with four applications of 1,2-DMH (40 mg/kg body weight) in Wistar rats [62]. In addition, preconditioned rats (for 1 week) with a diet containing 0.2% or 4% lyophilized cultures of L. acidophilus were later challenged with AOM (5 mg/kg/week) for two weeks to induce IC and kept on the same diet for 10 weeks and showed a significantly lower number of ACF compared to non-supplemented controls [63].
Another study has shown that PRO with L. acidophilus (2 × 10 9 CFU), orally administered (three times a week), significantly reduced the incidence and multiplicity of ACF after the induction of CRC with 1,2-DMH [64]. The preventive administration of PRO (07 CCM7766) with L. plantarum to rats that received applications of 1,2-DMH was able to reduce the inflammatory process in the colon through the negative regulation of proinflammatory cytokines (IL-2, IL-6, IL-17, and TNF-α,), the elevation of the anti-inflammatory cytokine IL-10, and the suppression of NF-κB activity in mucosal cells [65]. Furthermore, a PRO composed of L. casei BL23 improved CRC induced in mice by a single application of AOM (8 mg/kg) followed by four cycles of DSS (2.5%) in drinking water. PRO had an immunomodulatory effect through a reduction of the IL-22 cytokine and an antiproliferative effect measured through the upregulation of caspases 7 and 9. There were no significant differences between IFN-γ, TNF-α, IL-6, and IL-10. None of the mice in the group that received L. casei BL23 developed macroscopic tumors, while 67% of the mice in the cancer group did. Ki-67 immunolabeling was lower in the group that received PRO [66].
It has been documented that B. lactis inhibits the NF-kB pathway and regulatory genes in intestinal epithelial cells after the induction of IC with AOM and DSS [67]. The impairment of the intestinal epithelial barrier can be considered one of the first events that occurs in intestinal inflammation since it facilitates the entry of antigens from the intestinal lumen to the mucosa, which can lead to an uncontrolled and exacerbated immune response [68].
An experimental model of CRC induction with 1,2-DMH (30 mg/kg, twice a week for three weeks) investigated the effect of a diet supplemented with a combination of L. casei, B. bifidum (10 8 CFU/mL) and sphingomyelin (0.05%) and showed improvements in the number of ACF in the colon [69]. Another study showed that long-term (24 weeks) consumption of PRO (B. longum and L. gasseri) resulted in significant inhibition of ACF formation after the induction of CRC with 1,2-DMH [70].
In addition, PRO with De Simone formulation (L.casei, L. plantarum, L. bulgaricus, L. acidophilus, B.longum, B. breve, B. infantil, and Streptococcus thermophilus) showed anticancer and anti-inflammatory activity in an experimental induction model of the CRC [71]. PRO enriched in L. delbrueckii UFV-H2b20 or B. animalis var. lactis Bb12 resulted in a reduction in the total number of ACF (55.7% and 45.1%, respectively). However, a synergistic effect of Lactobacillus and Bifidobacterium supplementation on ACF inhibition was not observed [72]. L. acidophilus has been found to have a more important effect than B. bifidum in the chemoprevention of CRC. The administration of PRO inhibited the incidence of colonic lesions by about 57% with L. acidophilus and 27% with B. bifidum compared to the group that had the induction of IC with AOM and did not receive PRO [73]. Future studies are warranted to better balance the PRO composition of selected bacteria with more anti-tumoral effects. Table 3 summarizes the main selected studies with PRO, discussed in this article. All of the studies are from experimental CRC models. The current literature supports the benefit of PRO to intestinal health in improving colonic histopathological scores, OS, and inflammation biomarkers. Table 3. Summary of animal model studies of PRO supplementation following CRC induction.
Progression from this precursor lesion to CRC is a multistep process, accompanied by alterations in several suppressor genes that result in abnormalities of cell regulation, and it has a natural history of 10-15 years. The 10-15-year time frame of this process provides an opportunity for both primary and secondary prevention [74]. An advantage of chemoprevention over the current secondary prevention strategy of routine colonoscopy is the potential to intervene early in the carcinogenic sequence to reduce CRC risk, allowing for less frequent surveillance exams and a reduction in the number of invasive cancers.

Conclusions
New combination therapies are desirable for CRC prevention and adjuvant treatment. This was the first study to evaluate both PS and PRO supplementation in the chemoprevention of precursor lesions for CRC. Further studies should be performed to clarify which PRO-selected microorganism composition and concentration are more synergistic to improve PS biological effects and CRC chemoprevention.
Altogether our findings suggest that PRO alone and PRO combined with PS were the best intervention strategies to improve experimental 1,2-DMH-induced CRC regarding several parameters of carcinogenesis. Our findings highlight the importance of preventive measures to control intestinal microbiota and minimize CRC incidence in high-risk populations.
We acknowledge, due to the relatively low number of results obtained and experiments performed, we could not dissect in-depth the fine downstream inflammatory and early tumorigenesis crosstalk pathways (including the canonical Wnt signaling) that could shed light to find novel and promising pharmacological targets to halt CRC precursors lesion progression. Such results may guide future clinical trials in large populations worldwide to prevent/slow the occurrence of CRC.