According to the Institutional Animal Care and Use Committee at the Faculty of Veterinary Medicine, University of Sadat City, Sadat City, Egypt (Ethical approval number: VUSC-025-1-22). We carried out the experimental procedure after the Animal Protocols Evaluation Committee’s affirmative opinion. Twenty-four male albino rats (180–200 g) of the Wistar strain were used. All animals in the study were housed in pathogen-free facilities under a 12-hour light/dark cycle at constant temperature and humidity and were fed standard rodent chow and water ad libitum (1).

For studies of liver tumor development, 15-day-old rats were treated with a single dose of DEN (Sigma-Aldrich) given dissolved in saline at a dose of 25 mg/kg body weight by i.p. injection. Rats in one randomly pre-assigned group were killed 4 months after DEN administration for histological and biochemical analyses (32).

In the current study, rats were divided into seven groups arranged as follows:

After the experimental period, rats fasted overnight, and then blood samples were taken from the heart under diethyl ether anesthesia. Rats were sacrificed by decapitation; the whole liver was dissected, and a small part from the right lobe of the liver was cut and fixed in 10% phosphate-buffered formalin (pH 7.2) for histopathological examination. The remaining liver was stored at -80°C. Liver tissue samples were fixed in 10% neutral buffered formalin. The fixed specimens were then trimmed, washed, and dehydrated in escalating grades of alcohol, cleared in xylene, embedded in paraffin, sectioned at 4-6 U thickness, and stained with hematoxylin and eosin.

The liver tissues of sacrificed mice were crushed in liquid nitrogen using phosphate-buffered saline (PBS) supplemented with 1% fetal bovine serum (FBS). Then the crashed tissues were homogenized, filtered using a 0.25µm filter, and collected in RNase- and DNase-free Eppendorf tubes. The homogenized and purified cells were used for total RNA isolation and flow cytometry staining.

The genomic DNA of liver tissue from the prepared rats was isolated using a DNA purification kit (Qiagen) according to the manufacturing protocol. To investigate the methylation activity in the coding sequences of DLC-1, TET-1, NF-kB, and STAT-3 genes, genomic DNA was digested for 4 hours at 37°C with either methylation-sensitive or methylation-dependent resection enzymes provided in the EpiTect Methyl II DNA Restriction Kit (Qiagen, USA). Quantitative real-time PCR (qRT-PCR) was used to achieve the methylation activity in the coding sequences of the indicated genes using the specific primers listed in Table 1 and cleaved DNA products. The following amplification program was used: 95°C for 10 min, 40 cycles (95°C for 45 sec, 62°C for 30 sec, and 72°C for 45 sec), and finally 72°C for 10 min. The mean cycle threshold (Ct) outcome using cleaved DNA by methylation-sensitive restriction was considered a methylated level, while the mean Ct value using cleaved DNA by methylation-dependent cleaved DNA was considered an unmethylated level. The methylation activities were calculated according to delta-delta Ct (ΔΔct) equations. Delta Ct (Δct) was equal to Ct for the methylated value and Ct for the unmethylated value. ΔΔct was equal to Δct for the sample and Δct for the control. Finally, the methylation fold change was equivalent to 2-ΔΔct (33).

According to the manufacturer’s protocol, the genomic DNA was isolated from the obtained samples using a DNA purification kit (Qiagen, USA). Bisulfite analysis of the DLC-1 and TET-1 promoter regions was performed using the spin columns of the EpiTect Bisulfite Kit (Qiagen, USA). According to the manufacturer’s protocol, the purified genomic DNA (1 µg) was treated with 100 µl sodium bisulfite (10M) and 100 µl sodium chloride (1M). The reaction was thermally denaturized for 5 minutes at 95°C and incubated for 6 hours at 55°C (3 cycles) for complete conversion of unmethylated cytosine. The converted DNA was then applied to PCR amplification using the following unmethylated and methylated specific primers, as displayed for DLC-1 and TET-1 promoter regions; DLC1-Unmethylated-F-5- AAACCCAACAAAAAAACCCAACTAACA- 3, DLC1- Unmethylated -R-5- TTT TTTAAA GAT TGAAATGAGGGAGTG -3, DLC1-Methylated-F-5- CCCAACGAA AAAACC CGACTAACG-3, DLC1- Methylated-R-5-TTTAAAGATCGAAACGAGGGAGCG -3, TET1-Unmethylated-F-5- GGCTCGGGCCTTGACTGTGCTG -3, TET-Unmethylated-R-5-AGGTTTTGGTCGCTGGCCGGGT -3, TET1-Methylated-F-5-AACTCAAACCTTAACTATACTA -3, TET1-Methylated-R-5- CGCTAACCGAATCACATTCCCA-3 (27). The conventional PCR was used to amplify the methylated and unmethylated fragments in the bisulfite-converted DNA using the following parameters: 95 °C for 5 min and 40 cycles (95 °C for 45 sec, 62 °C for 45 sec, and 72 °C for 45 sec) (23). The PCR products were loaded in 1% agarose gel and electrophoresed using a 1X-TBE buffer, and the amplified fragments were monitored using the UV trans-illuminator with a long-wave (320 nm) UV and gel documentation system.

Flow cytometry was used to evaluate the protein expression profiles of DNMT1, MS, DLC-1, TET-1, NFkB, and STAT3 in the liver tissue of treated rats. Accordingly, drug-treated homogenized cells were centrifuged for 3 minutes at 1500 rpm. The supernatant was discarded; the pellet was resuspended in PBS for washing and then centrifuged and resuspended in cold methanol for fixation. The cells were resuspended in PBS for permeabilization, including Triton-X-100 (0.1%), and incubated for 3 min. For staining of DLC-1, the cells were resuspended and incubated overnight at 4°C in PBS supplemented with 1% BSA and diluted mouse monoclonal anti-DLC-1 (Sigma-Aldrich, Germany). After washing, the cells were centrifuged and resuspended in PBS that contains 1% BSA and 1-1000 secondary antibody donkey anti-mouse IgG (Alexa Fluor 488, Invitrogen, Germany). The same conditions were followed in the staining of TET-1 protein in treated cells using rabbit polyclonal anti-TET-1 (PromoCell, Germany) and goat anti-rabbit IgG (Alexa Fluor 594, Abcam, USA). The same procedures were considered for staining DNMT1, MS, NFkB, and STAT3 using specific antibodies: mouse monoclonal anti-DNMT1 (Sigma-Aldrich, Germany), rabbit monoclonal anti-MS (Sigma-Aldrich, Germany), mouse monoclonal anti-NFkB (Sigma-Aldrich, Germany), and rabbit monoclonal anti-STAT3 (Sigma-Aldrich, Germany), respectively. The same secondary antibodies, donkey anti-mouse IgG (Alexa Fluor 488, Invitrogen, Germany) and goat anti-rabbit IgG (Alexa Fluor 594, Abcam, USA), were used independently to achieve the kinetic protein expression of each target. Finally, the flow cytometry assay (BD Accuri 6 Plus) was used to assess the protein levels using a resuspended pellet in 500 µl PBS (34, 35).

Total RNA was isolated from rat-prepared tissues using TriZol (Invitrogen, USA), chloroform, and isopropanol. Then the total RNA was purified by using an RNA purification kit (Invitrogen, USA). According to the manufacturer’s protocol, the complementary DNA (cDNA) was prepared from total RNA using the QuantiTech Reverse Transcriptase Kit (Qiagen, USA). The relative gene expression of DNMT-1, MS, and the mRNA expression of the indicated genes was quantified using the QuantiTect SYBR Green PCR Kit (Qiagen, USA). Ct values of the housekeeping gene, GAPDH, were used for normalization. The following PCR parameters were used to assess fold changes in the indicated gene expression: 94°C for 5 min, 40 cycles (94°C for 30 sec, 60°C for 30 sec, 72°C for 45 sec). Delta-delta Ct equations have been used to determine the relative gene expression indicated by fold changes in quantified mRNA (36–38).

The data obtained by q-RT-PCR reveal that the cycle threshold values (Ct) were analyzed using the previously described ΔΔCt equations. Hence, the relative gene expression of targeted genes on the RNA level was indicated by fold changes that equaled 2-ΔΔct (23, 33, 34). The student’s two-tailed t-test was used to determine the differences in the analyzed data. P<0.05 was considered statistically significant (*), while P<0.01 was considered highly significant (**).

Hepatocarcinogenesis of the rats was induced by using DEN; then, the animals were treated with different agents such as GA, DOX, an association of GA with DOX, and a combination of GA and probiotics. As shown in Figure 1 , GA decreases body weight by 25% as compared to positive controls (HCC animals). While the DOX drug contributes to increasing the body weight of the animal by 20% as compared to the positive control, Figure 1B displays the effect of the different agents on the organ’s weight. GA has a significant effect on decreasing the liver weight by 33.3% as compared to positive control animals (HCC animals), while GA increases the weight of the lung by 22% as compared to positive control animals. Also, the GA has a significant effect on reducing the kidney weight by 16% as compared to the positive control (HCC animals). There is no significant difference in the heart weight of all the groups. The Supplementary Figure 1 indicated the effect of the association of GA with probiotics and showed normal liver size and color compared with untreated rats with HCC.

Figure 2 shows that the negative control animals (normal animals) group showed normal-textured tissues of hepatic lobules with regular hepatic cords and a prominent central hepatic vein. The hepatic cells were joined in the organizing plate’s grade (0) ( Figure 2A ). The hepatic carcinogenic group (positive control) showed neoplastic areas with moderately differentiated grade II large hepatic cells with deeply basophilic cytoplasm and hyperchromatic nuclei, which indicated malignancy in the form of cellular and nuclear pleomorphism, scanty cytoplasm, and frequent mitotic The hepatic lobule displayed disorganization of hepatic cords and focal areas of necrosis and fatty change with hyperplasia of Kupffer cells. Figure 2B . The carcinogenic animal group treated by GA showed necrobiotic changes in hepatic carcinoma cells with hyperplasia of Kupffer cells. A partial 51%–99% of tumor necrosis was seen. The necrotic areas showed nuclear pyknosis and deeply eosinophilic cytoplasm. Disorganization of carcinoma cells with widening of hepatic sinusoids was noticed in Figure 2C . The hepatocarcinogenesis animal group treated with DOX showed a mild swelling of hepatic carcinoma cells with hyperplasia of Kupffer cells. Less than 50% of tumor necrosis was seen. There are apoptotic cells in the small focal necrotic areas in a small number of hepatocytes ( Figure 2D ). The liver tissue section of the carcinogenic group treated by bacteria showed vacuolar degeneration of hepatic carcinoma cells with hyperplasia of Kupffer cells. Poor (<50%) tumor necrosis was seen. The necrotic cells showed nuclear pyknosis and deeply eosinophilic cytoplasm ( Figure 2E ). The hepatocarcinogenesis animal group treated with a combination of DOX and GA showed mild swelling of hepatic carcinoma cells with hyperplasia of Kupffer cells. Partial (51%–99%) tumor necrosis was seen. The small focal necrotic areas showed apoptosis of a small number of hepatocytes ( Figure 2F ). The hepatocarcinogenesis animal group treated with GA associated with probiotics showed necrobiotic changes in hepatic carcinoma cells with hyperplasia of Kupffer cells. Partial 51%–99% tumor necrosis was seen. The necrotic areas showed nuclear pyknosis and deeply eosinophilic cytoplasm. The disorganization of carcinoma cells with the widening of hepatic sinusoids was noticed in Figure 2G . Finally, the cell morphology of HCC-developed rates treated with GA, a combination of GA with bacteria, and GA with Dox showed regular statements of liver cells. This observation suggests that treatment with the indicated compounds can restore liver cancer cells.

To investigate the possible involvement of DNA methylation in restoring liver cancer cells, the relative gene expression of both DNMT1, as the enzyme responsible for DNA methyl ion activity and methionine synthase (MS), as the enzyme facilities performing methyl groups was detected in liver tissue. Interestingly, the relative gene expression of DNMT1 significantly increased in rates with well-differentiated HCC (positive control). Meanwhile, its relative gene expression strongly downregulated in rats treated with the GA, a combination of GA with bacteria, and GA with Dox ( Figure 3A ). The relative gene expression of MS increased in positive control rats. While, its expression was significantly reduced in rats treated with GA, a combination of GA with bacteria, and GA with Dox GA. Notably, MS gene expression showed neglected differentiation in treated rates with probiotic bacteria compared with other treated rats ( Figure 3B ). Interestingly, the protein expression profile of DNMT1 and MS was markedly increased in rats with HCC (positive control) up to 45% and 40% of stained cells compared with the negative control rats indicated by flow cytometry ( Figure 3C ). In addition, their expression profile was notably reduced to a percentage of about 15% of stained cells in rats treated with GA or Dox. While combining GA and Dox during treatment, the rats successfully reduced the expression of DNMT1 and MS protein to about 7% of stained cells ( Figure 3C ). This result demonstrates the hypermethylation activity in HCC-developed rats indicated by DNMT1 and MS expression profile which have been reduced upon treatment with the GA, combination of GA with Dox.

To investigate the possible epigenetic alteration in the coding sequence of TET-1 and DLC-1 genes, methylation-dependent and methylation-sensitive enzymes have been used to digest the purified genes using qRT-PCR. Interestingly, the methylation activity in TET1 increased up to a 16-fold change in rats with well-differentiated HCC, while it increased up to a 50-fold change in DLC1 as presented in ( Figure 4A ; Supplementary Table 1 ). Notably, GA treatment sufficiently decreased the methylation activity in both genes to only a 10-fold change. Furthermore, the combination of GA with DOX or bacteria successfully avoids the methylation activity in both genes ( Figures 4A, C ; Supplementary Table 1 ). Based on the principle of sodium bisulfate treatment indicated that all unmethylated cytosine is transferred to uracil after treatment with sodium bisulfate (39). Thus, the methylated and unmethylated primers that could be complementary with converted and non-converted cytosine were used to detect the unmethylated and methylated fragments, respectively. According to the gel electrophoresis presented in ( Figures 4B, D ), the methylated band of TET-1 and DLC-1 disappeared in the negative control while showing the detectable band in the positive control. Interestingly, GA treatment showed a faint band of methylated and unmethylated TET1 promoter while showing a strong band of unmethylated DLC1 promoter, indicating the role of GA in decreasing the methylation activity in the promoter region of both TET1 and DLC1. The combination of GA and DOX showed a faint band of the methylated and unmethylated promoter region of DLC1 compared to the promoter region of TET1. The combination of GA with the bacteria determined a faint band of the methylated promoter region of both TET1 and DLC1. Together, these data indicate the anti-methylation activities of GA, in addition to DOX during HCC treatment.

To investigate the possible involvement of DNA methylation in restoring liver cancer cells, the relative gene expression of both DNMT1, the enzyme responsible for DNA methyl ion activity, and methionine synthase (MS), the enzyme facility performing methyl groups, was detected in liver tissue. Interestingly, the relative gene expression of DNMT1 significantly increased in rates with well-differentiated HCC (positive control). Meanwhile, its relative gene expression was strongly downregulated in rats treated with GA, a combination of GA with bacteria, and GA with Dox ( Figure 3A ). The relative gene expression of MS increased in positive control rats. While its expression was significantly reduced in rats treated with GA, a combination of GA with bacteria, and GA with Dox GA, Notably, MS gene expression showed neglected differentiation in treated rates with probiotic bacteria compared with other treated rats ( Figure 3B ). Interestingly, the protein expression profile of DNMT1 and MS was markedly increased in rats with HCC (positive control) up to 45% and 40% of stained cells, respectively, compared with the negative control rats, as indicated by flow cytometry ( Figure 3C ). In addition, their expression profile was notably reduced to a percentage of about 15% of stained cells in rats treated with GA or Dox. While combining GA and Dox during treatment, the rats successfully reduced the expression of DNMT1 and MS protein to about 7% of stained cells ( Figure 3C ). This result demonstrates the hypermethylation activity in HCC-developed rats indicated by the DNMT1 and MS expression profiles, which have been reduced upon treatment with GA or a combination of GA and Dox.

NF-kb and STAT3 play an essential role in developing and regulating liver cancer. Accordingly, compared to healthy rats, we examined the methylation activity in the coding sequences of NF-kb and STAT-3 in rats with HCC (negative control). We found that the methylation activity in the coding sequence of both NF-kb and STAT3 was significantly reduced in rats with HCC while significantly increasing upon GA treatment and its combination with DOX or the bacteria ( Figures 5A, C ; Supplementary Table 3 ). Notably, the relative gene expression of NF-kb and STAT-3 significantly reduced upon GA treatment, and it indicated a combination, while their expression dramatically increased in rats with HCC ( Figures 5B, D ; Supplementary Table 4 ). These results demonstrate the potential methylomic changes in the coding sequence of NF-kb and STAT3 in rats with HCC and suggest the possible regulatory role of GA in their expression during HCC treatment. Furthermore, the protein expression profile of both NF-kB and STAT3 was markedly increased in rats with HCC (positive control) in almost 45% of stained cells, as indicated by the flow cytometry assay ( Figure 5E ). Meanwhile, their expression profile almost retched the same level of negative control upon treatment with indicated agents, indicating the role of each agent and indicated combination in regulating the expression of oncoprotein, NF-kB, and STAT3 ( Figure 5E ).

TET-1 and DLC-1 gene expression was quantified using RT-PCR. The relative gene expression of TET-1 was significantly downregulated in rats with well-differentiated HCC (positive control) (almost 90% inhibition). In contrast, the rats treated with different agents, such as GA, bacteria, a combination of GA with bacteria, and GA with DOX, showed significant upregulation in the TET-1 and DLC-1 genes, as Figure 6A and Supplementary Table 2 displayed. The relative gene expression of DLC-1 was significantly reduced in rats with well-differentiated HCC (approximately 90% inhibition). However, the relative gene expression of DLC-1 was significantly upregulated in rats treated with GA, bacteria, a combination of GA with bacteria, and GA with DOX ( Figure 6C ; Supplementary Table 2 ). Flow cytometric analysis was used to check the kinetic protein expression, as shown in Figure 6B . The protein profile of TET-1 in treated rats with GA showed detectable expression in more than 13% of stain cells, while its protein expression was dramatically reduced to less than 3% in positive rats. The protein expression of DLC-1 in rats treated with GA was markedly detected in almost 18.5% of stain cells, whereas less than 3% of stain cells were found in positive rats. Interestingly, the combination of GA with DOX showed potent expression of both TET-1 and DLC-1 in approximately 19.5% and 30% of stained cells, respectively. Likewise, the combination of GA with the bacteria showed competent expression of both TET-1 and DLC-1 in approximately 20.2% and 29.8% of stained cells, respectively. The results demonstrate the activity of GA and its combination with DOX or bacteria to increase the expression of both the TET-1 and DLC-1 genes.