Effect of Omega-3-Rich Fish Oil on TNF- a Expression in Mice's Colonic Tissue Induced with Azoxymethane (AOM) and Dextran Sodium Sulphate (DSS)


Elvan Wiyarta1, Kusmardi Kusmardi2,3,4, Yurnadi Hanafi Midoen5*

1Faculty of Medicine, Universitas Indonesia, Jl Salemba Raya No 6, Senen, Central Jakarta, 10430 Indonesia.

2Departement of Anatomic Pathology, Faculty of Medicine, Universitas Indonesia

Jl Salemba Raya No 6, Senen, Central Jakarta, 10430 Indonesia.

3Drug Development Research Cluster, Indonesia Medical Education and Research Institute, Faculty of Medicine, Universitas Indonesia, Jl Salemba Raya No 6, Senen, Central Jakarta 10430 Indonesia.

4Human Cancer Research Center, Indonesia Medical Education and Research Institute, Faculty of Medicine, Universitas Indonesia, Jl Salemba Raya No 6, Senen, Central Jakarta, 10430 Indonesia.

5Departement of Medical Biology, Faculty of Medicine, Universitas Indonesia,

Jl Salemba Raya No 6, Senen, Central Jakarta 10430 Indonesia.

*Corresponding Author E-mail: yurnadi.kes@ui.ac.id



Objective: Colorectal cancer (CC) is one type of cancer with a high incidence worldwide. Many therapeutic techniques have been for CC but have not yet yielded satisfactory results. Fish oil has potential as an alternative therapy for CC through its anti-inflammatory effects. Here, we want to investigate that effect on TNF-α expression using omega-3-rich fish oil (FO). Methods: FO was injected into Swiss mice that have been induced (2% in drinking water) with DSS for seven consecutive days. Animals were separated into six groups: normal, negative control, positive control, solvent control, and FO groups (3 and 6 mg/kg body weight/day). All animals were sacrificed, and the colons were collected then stained with anti-TNF-α. The stained sections were subsequently examined with ImageJ based on colour density. Results: Based on the H-Score of each group, FO 3 mg and 6 mg has significantly decreased the expression of TNF-α compared to the negative control (p=0.001 and p=0.009). Conclusion: FO administration was able to inhibit the expression of TNF-α on mice's colonic tissue induced with AOM and DSS.


KEYWORDS: Fish oil, TNF-α, Anti-inflammatory, Colorectal cancer, Mice's colonic tissue.




Colorectal cancer (CC) is one type of cancer with a high incidence globally, with the second rank for cancers affecting men and third for cancers that attack women1. In 2012, there were around 614,000 women and 746,000 men diagnosed with CC. Of this population, 694,000 people died from CC2-3.


Responding to the high incidence of CC, many treatment techniques were developed. Conventionally, there are several CC treatments, including radiotherapy, chemotherapy, and immunotherapy1. However, this treatment technique has not yet yielded satisfactory results. Radiotherapy and immunotherapy have a common therapeutic effect but a high risk of radiation exposure and toxicity4. Meanwhile, immunotherapy also shows toxicity and autoimmunity in patients5. Several new treatments also have been developed, such as laparoscopic surgery and resection of metastatic  disease1,5. However, its safety and effectiveness are still being investigated. Therefore, innovative strategies need to be found in order to improve the treatment of CC.



In recent years, various alternative therapies have been developed to treat CC, which is nutritional therapy6. Unlike therapy with drugs, nutritional therapy promotes the use of natural ingredients that patients can consume. One that is being developed is therapy using omega-3-rich fish oil (FO).


The content of omega-3 polyunsaturated fatty acid (n-3 PUFA) in FO is allegedly an anti-inflammatory agent through various inhibition mechanisms. It triggers the activation of the gamma peroxisome proliferator-activated receptor (PPAR-γ), which then inhibits the activity of nuclear factor kappa B (NFkB). NFkB is a gene that codes to release pro-inflammatory cytokines activated during cancer progression6-11. In its activity, NFkB secretes various cytokines such as tumour necrosis factor-alpha (TNF-α)12,13 and other pro-inflammatory cytokines. These cytokines will further increase the inflammatory cascade through further activation of the NFkB gene, further aggravating cancer progression6-13. Therefore, it can be assumed that n-3 PUFAs in FO can affect the expression of TNF-α on CC cells12,13. However, the evidence supporting this is still minimal.


Until now, there has been no in-vivo research focused on the immune system investigating the inhibition of TNF-α expression after administration of FO to CC cells. Based on this background, it is necessary to investigate the inhibition of TNF-α expression by FO in CC that has been induced by azoxymethane (AOM) and dextran sodium sulphate (DSS). DSS and AOM are substances that can create chronic inflammation in colon epithelial cells that can imitate the state of CC cells14.



Experimental animal:

Swiss mice at Indonesia's National Institute of Health Research and Development (NIHRD) were acclimated and studied for a week before inducing dextran sodium sulphate (DSS). Monitoring the temperature of 25°C, 12 hours of light/dark cycle, 55% humidity and traditional food and drink comply with the Animal Care and Use Committee's Guide for the Care and Use of Laboratory Animals.


Chemical materials:

Reagents used in this study include dextran sodium sulphate BM 500,000 (Sigma Aldrich), Aspirin (Brataco Inc.), sodium carboxymethylcellulose/CMC Na (Brataco Inc.), anti-TNF-α (Abcam), formaldehyde (Brataco Inc.), ether (Brataco Inc.), xylol (Merck), absolute alcohol (Merck), 70% alcohol (Merck), paraffin sodium (Brataco Inc.).




Study design and colitis administration:

The Universitas Indonesia Faculty of Medicine's Ethics Committee gave its approval to the study's protocol. Dextran sodium sulphate (DSS) was added into the animals' drinking water for seven consecutive days, except the usual normal (N) group. It was divided into six groups: normal (N), negative control (C-), aspirin-positive control (C+), solvent control (Cs), FO 6mg/kg BW (D1), and FO 3mg/kg BW (D2), each contains five mice. No oral exposure to DSS for the N group, who received 0.9percent saline13. Randomization and acclimation of the subjects were placed one week prior to the DSS induction. All animals were sacrifice, and their colons were collected after treatment.



Immunohistochemistry staining in this study followed the staining method in previous studies15-17. Fixation in 10% phosphate buffer for 10 hours at 4oC was used for the TNF-α staining technique. Additionally, anti- TNF- α antibodies were incubated for two hours at room temperature in the humidity chamber with phosphate buffer solution (PBS). This is followed by 10 minutes of room temperature incubation with 3,3'-diaminobenzidine (DAB), followed by dehydration and mounting18,19. Using a procedure described by Amalia et al.20, the sample slices were counterstained with Lilie Mayer haematoxylin solution for 1-2 minutes before being washed with water. A second rinse followed a 60-second immersion in lithium carbonate in water. Xylol was used to clarify the sample after it had been dehydrated in ethanol. As a final step, the parts were coated with a liquid cover, which is water-based. The histological changes in the stained sections were then investigated.


Quantification of TNF-α expression:

An integrated camera and computer with Leica LAZ software and the Leica DM750 microscope were used to capture images of each preparation at the microscope's 400X magnification. During each preparation, ten visual fields were photographed randomly. A plugin application in Image J, Immunohistochemistry (IHC) profiler, was then used to compute the number of epithelial cells in an image. The H Score is the outcome of quantification. H-score quantification using Image J follows the quantification conducted by Kusmardi et al.21


Statistical analysis:

The mean and standard deviation are shown for every data. To examine the differences between treatments, data were analysed using an analysis of variance (ANOVA) in SPSS 20.0 and subsequently with a Tukey's Post Hoc test. There is statistical significance when the difference in p-values is less than 0.05.




Qualitatively, the expression of TNF-α can be seen by comparing the brown colour-density in each field of view. The TNF-α expression is depicted by the brown colour-density depicted in the cytoplasm of cells. The difference in the brown colour-density in the cytoplasm of colon epithelial cells can be seen in Figure 1. Figure 1 (N) has a neutral colour density. In contrast, figures 1 (C-) and (Cs) have a high brown colour-density intensity. Meanwhile, figure 1 (C+), (D1), and (D2) have almost the same brown colour density. All the results of this colour intensity are grouped and statistically tested by IHC profiler in ImageJ and SPSS, respectively.



Figure 1. Expression of TNF-α on mice colonic epithelial cell in various group with 400X magnification [normal (N), negative control (C-), aspirin-positive control (C+), solvent control (Cs), FO 6 mg/kg BW (D1), and FO 3 mg/kg BW (D2)].


Quantification of all TNF-α expressions in the images above is assessed by calculating the sum and average of Histo Score (HScore)22 for the group. The results of quantification were converted into H-Score based on the formula. H-Score = (% low positive x 1) + (% positive x 2) + (% high positive x 3). The results can be seen in Table 1.


Table 1. Average and standard deviation of HScores in Each group




Mean (± SD)



186,31 ± 8,01



195,05 ± 9,00



228,25 ± 9,96



220,22 ± 1,84



200,86 ± 6,51



206,31 ± 14,17


All data from Table 1 have normal distribution based on the Shapiro-Wilk normality test (p= 0.537). Based on Levene's test, the data groups are homogeneous (p = 0.333). Because the data groups were normally distributed and homogeneous, One-Way ANOVA statistical analysis was carried out and continued with the post hoc Tukey's test. ANOVA showed significant results (p= 0.001). Meanwhile, Tukey’s test results showed there were significant differences between N and C- (p= 0.001), N and D2 (p=0.02), N and Cs (p=0.001), C+ and C- (p<0.001), C+ and Cs (p=0.002), C- and D1 (p=0.001), C- and D2 (p=0.009), D1 and Cs (p=0.026) as shown in Figure 2.



Figure 2. Mean difference between groups based on Tukey's test. *p<0.05; **p<0.01; ***p<0.001 [normal (N), negative control (C-), aspirin-positive control (C+), solvent control (Cs), FO 6 mg/kg BW (D1), and FO 3 mg/kg BW (D2)].



Colorectal cancer model with AOM and DSS:

From the study results, there was a significant difference in the expression of TNF-α between negative control and normal colon tissue (p<0.001). This significant difference in TNF-α expression indicates that the induction of AOM and DSS can create a pro-inflammatory environment as measured by elevated levels of TNF-α compared to normal colonic tissue.


As pro-inflammatory agents, AOM and DSS can activate the NFkB activation pathway, which in turn increases TNF-α expression. This activation is associated with dissociation of IkB protein which causes activation of NFkB. In addition, TNF-α will activate the IkB kinase protein, which will dissociate IkB from NFkB complexes. This will cause more NFkB to activate, and more TNF-α will also be produced. NFkB and TNF-α have mutually reinforcing relationships23.


However, as seen in Figure 1 (N), normal colon tissue is not entirely blue (counterstain colour). This shows that TNF-α is also still expressed in tissues that AOM and DSS do not induce. This happens because the factors that activate NFkB are not only from AOM and DSS. Toll receptor activation, IL-1, DNA mutation, and hypoxia can also promote NFkB activation24.


On the other hand, this study also shows significant differences between negative and positive control colon tissue. Aspirin has been known to be an inhibitor in the inflammatory process. As in Amalia, et al.20, administration of aspirin at a dose of 150mg/kg for four weeks reduced the incidence of dysplasia in the colon tissue, which was lower than in tissue given AOM and DSS. The same thing is also shown in the results of this study. The results showed a significant difference between the negative and positive control groups (p<0.001). This difference indicates that aspirin given to positive controls has significantly decreased TNF-α expressions compared with negative controls.


Effect of corn oil as solvent control on TNF-α expression:

In this study, researchers also provided research coverage in a group called solvent control (Cs). This was done to avoid research bias. Bias can occur because FO must be dissolved in corn oil to be injected into mice. If this solvent turned out to carry significance to the test results, the results of the test groups of doses 1 and 2 were biased.


According to the test results, there is a significant difference between the Cs with positive and normal controls. This shows that even though mice have been injected with AOM and DSS, the addition of corn oil will not cause significant changes. This was further strengthened by the insignificant difference between Cs and negative control (p = 0.725). This shows that the provision of corn oil is not much different from the negative control.


The effect of corn oil administration on inflammatory progression has an association with the n-6: n-3. This is in line with Yang et al which state that a good n-6: n-3 ratio is below 4:125. Meanwhile, if examined, the average corn oil has an n-6: n-3 ratio of 54.25. This high ratio shows that corn oil cannot influence the results of the fish oil therapy group test by reducing the inflammation process. This happens because the high content of n-6PUFA (pro-inflammatory agent) in corn oil exacerbates the inflammatory process. This finding is strengthened by Kirpich et al.26 study, which shows that corn oil causes intestinal inflammation and impaired intestinal barrier defense. In this study, corn oil tends to increase inflammation.


Effect of FO on TNF-α expression:

Fish oil is sometimes associated with the anti-inflammatory process. Fish oil contains many lipids that are useful for the body, one of which is n-6 PUFA and n-3 PUFA. The content of n-6 PUFA and n-3 PUFA varies, depending on the type of fish processed into fish oil. For example, salmon, sardines, and cod oil each have n-6: n-3 ratios of 0.12, 0.19, and 0.09. These ratios show that the n-3 PUFA content in fish oil tends to be the same or more than the n-6 PUFA content. This ratio is undoubtedly related to the inflammatory process because n-6 PUFA and n-3 PUFA are respectively pro-inflammatory and anti-inflammatory agents.


In this study, FO became a test material to assess its relationship regarding the expression of TNF-α, a pro-inflammatory cytokine. FO was administered in two doses: D1 (6mg/kg BW/day) and D2 (3mg/kg BW/day). Both D1 and D2 had significant TNF-α expression differences towards C- group. It shows that FO in mice colon tissue can reduce the expression of TNF-α. This statement is further strengthened by the data showing no significant difference between the D1 and D2 groups with the C+ group (p = 0.908 and p = 0.388). This shows that the administration of FO at D1 and D2 will affect TNF-α expression as good as the administration of aspirin. This finding related to n-3 PUFA as an anti-inflammatory agent. n-3 PUFA affects signalling pathways that control gene expression in inflammatory cells7. EPA and DHA (the result of n-3 PUFA metabolism) decrease the expression of adhesion molecules and the production of inflammatory cytokines in the COX-2 pathway6. In addition, EPA also reduces LPS-induced activation in the NFkB signalling pathway. This is related to the decrease in IκB phosphorylation. Likewise, DHA reduces the activation of NFkB in response to LPS in macrophages so that it has an effect that drives the decrease in IκB phosphorylation8,9. In addition, DHA and EPA affect TLR4 through their ability to interfere with the formation of rafts (raft) in the membranes of inflammatory cells11.


On the other hand, different results were obtained when the FO group were compared with the normal group. Significant differences were found between N and D2 but not between N and D1 (p = 0.151). This shows that the administration of FO at D2 cannot reduce the expression of TNF-α to customary conditions, unlike D1. These findings indicate a tendency for FO to be dose-dependent. However, there is no significant difference between D1 and D2 which does not strengthen the dose-dependent statement. This can occur because the range of differences in doses of D1 and D2 is relatively narrow, so that it cannot assess changes in effect at several doses.



Omega-3-rich fish oil (FO) has been widely studied as having anti-inflammatory effects. This effect was also related to the expression of TNF-α in cells as in this study. It was inversely proportional. The relationship between FO and TNF-α still need to be investigated. In addition, FO may have potential dose-dependent characteristics. Further research is needed that involves testing various FO doses to explore their dose-dependent characteristics.



The authors would like to thank Universitas Indonesia for funding this research through PUTI Grant with contract number NKB-2235/UN2.RST/HKP.05.00/2020.



All the authors had contributed equally.



There are no conflicts of interest to declare.



1.      Kuipers EJ. Grady WM. Lieberman D. Seufferlein T. Sung JJ. Boelens PG. van de Velde CJH et al Colorectal cancer.  Nature Reviews Disease Primers. 2015; 1:1-25. doi:10.1038/nrdp.2015.65

2.      GLOBOCAN. Estimated cancer incidence, mortality and prevalence in 2012 [Internet]. International Agency for Research on Cancer–WHO; 2012 [cited 2020 May 20]. Available from: http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx 

3.      Widjaja S. Yo H. Colorectal cancer in Indonesia - a centre report. Annals of Oncology. 2016 Jun; 27(Suppl 2): ii97. doi: 10.1093/annonc/mdw201.46

4.      Khairina D. Suzanna E. Triana D. Kadir A. Widyastuti TH. Sulistyowati LS. Rosalina I et al Profile of colorectal cancer in 14 provinces in Indonesia. Journal of Global Oncology. 2018 4: Supplement 2, 66s-66s. doi:10.1200/jgo.18.64300

5.      Bhutia SK. Maiti TK. Targeting tumors with peptides from natural sources. Trends in Biotechnology. 2008 Apr;26(4): 210–17. doi:10.1016/j.tibtech.2008.01.002

6.      Calder PC. Omega-3 fatty acids and inflammatory processes: from molecules to man. Biochemical Society Transactions. 2017 Oct 15; 45(5): 1105-15. doi: https://doi.org/10.1042

7.      Volpato M. Hull MA. Omega-3 polyunsaturated fatty acids as adjuvant therapy of colorectal cancer. Cancer and Metastasis Reviews. 2018 Sep;37(2-3): 545-55. doi:10.1007/s10555-018-9744-y.

8.      Morland SL. Martins KJB. Mazurak VC.  n-3 polyunsaturated fatty acid supplementation during cancer chemotherapy. Journal of Nutrition & Intermediary Metabolism. 2016 Sep; 5 (1): 107-16. doi.org/10.1016/j.jnim.2016.05.001

9.      Patterson WL. Georgel PT. Breaking the cycle: the role of omega-3 polyunsaturated fatty acids in inflammation-driven cancers. Biochemistry and Cell Biology. 2014 Oct;92(5):321-8. doi: 10.1139/bcb-2013-0127

10.   Vergili-Nelsen JM. Benefits of fish oil supplementation for haemodialysis patients. Journal of the American Dietetic Association. 2003 Sep;103(9):1174-7. doi: 10.1016/s0002-8223(03)00984-2.

11.   Lee JY. Sim TB. Lee J. Na HK. Chemopreventive and chemotherapeutic effects of fish oil derived omega-3 polyunsaturated fatty acids on colon carcinogenesis. Clinical Nutrition Research. 2017 Jul; 6(3): 147–160. doi:10.7762/cnr.2017.6.3.147

12.   Kusmardi. Priosoeryanto BP. Harlina E. Cornain S. Inhibition activities of fish oil in inos, cox-2, and β-catenin expressions in colorectal preneoplasia of mice induced by azoxymethane and dextran sodium sulphate. Journal of Applied Biotechnology. 2014; 2 (2): 91-9. doi:10.5296/jab.v2i2.6294

13.   Zins K. Abraham D. Sioud M. Aharinejad S. Colon cancer cell-derived tumour necrosis factor-alpha mediates the tumour growth-promoting response in macrophages by up-regulating the colony-stimulating factor-1 pathway. Cancer Research. 2007 Feb 1; 67(3): 1038-45. doi:10.1158/0008-5472.CAN-06-2295

14.   Thaker AI. Shaker A. Rao MS. Ciorba MA. Modelling colitis-associated cancer with azoxymethane (AOM) and dextran sulphate sodium (DSS). Journal of Visualized Experiments. 2012 Sep 11;(67):4100. doi:10.3791/4100

15.   Kusmardi K. Wiyarta E. Estuningtyas A. Sahar N. Midoen YH. Tedjo A. Potential of Phaleria macrocarpa Leaves Ethanol Extract to Upregulate the Expression of Caspase-3 in Mouse Distal Colon after Dextran Sodium Sulphate Induction. Pharmacognosy Journal. 2021;13(1):23-29. doi: 05530/pj.2021.13.4

16.   Kusmardi K. Wiyarta E. Estuningtyas A. Sahar N. Midoen YH. Tedjo A. Pakpahan A. Potential Inhibition by Phaleria macrocarpa Leaves Ethanol Extract on Ki-67 Expression in Distal Colon Mouse. Pharmacognosy Journal. 2021;13(2):443-449. doi: 10.5530/pj.2021.13.56

17.   Rustamadji P. Wiyarta E. Bethania KA. Kusmardi K. Potential of AKT2 expression as a predictor of lymph-node metastasis in invasive breast carcinoma of no special type. Journal of Pathology and Translational Medicine. 2021 Jul;55(4):271-278. doi: 10.4132/jptm.2021.04.26

18.   Maharani R. Kusmardi. Elya B. Inhibitory activity goblet depletion and focal inflammatory phaleria macrocarpha leaves ethanol extract on crypta mouse after dextran sodium sulphate induction. International Journal of PharmTech Research. 2019; 12(1): 37-48. doi:10.20902/IJPTR.2019.120106

19.   Ramos-Vara JA. Technical aspects of immunohistochemistry. Veterinary pathology. 2005; 4: 405-26. doi: 10.1354/vp.42-4-405

20.   Amalia AW. Kusmardi. Elya B. Arsianti A. Inhibition of carcinogenesis by seed and soybean meal extract in colon of mice: Apoptosis and dysplasia. Asian Journal of Pharmaceutical and Clinical Research. 2017 Apr 1; 10(4): 123-128. doi:10.22159/ajpcr. 2017.v10i4.16276

21.   Kusmardi K. Wiyarta E. Rusdi NK. Maulana AM. Estuningtyas A. Sunaryo H. The potential of lunasin extract for the prevention of breast cancer progression by upregulating E-Cadherin and inhibiting ICAM-1. F1000 Research. 2021 Sep 8; 10:902. doi: 10.12688/f1000research.55385.1

22.   Fardet L. Galicier L. Lambotte O. Marzac C. Aumont C. Chahwan D. Coppo P et al Development and validation of the HScore, a score for the diagnosis of reactive hemophagocytic syndrome. Arthritis & Rheumatology. 2014 Sep; 66(9): 2613-20. doi:10.1002/art.38690

23.   Hayden MS. Ghosh S. Regulation of NF-κB by TNF Family Cytokines. Seminars in Immunology. 2014 Jun; 26(3): 253–266. doi:10.1016/j.smim.2014.05.004

24.   Oeckinghaus A. Ghosh S. The NF-κB family of transcription factors and its regulation. Cold Spring Harbor Perspectives in Biology. 2009 Oct; 1(4): a000034. doi:10.1101/cshperspect.a000034

25.   Yang LG. Song ZX. Yin H. Wang YY. Shu GF. Lu HX. Wang SK et al Low n-6/n-3 pufa ratio improves lipid metabolism, inflammation, oxidative stress and endothelial function in rats using plant oils as n-3 fatty acid source. Lipids. 2016 Jan; 51(1): 49-59. doi:10.1007/s11745-015-4091-z

26.   Kirpich IA. Feng W. Wang Y. Liu Y. Beier JI. Arteel GE. Cameron K et al Ethanol and dietary unsaturated fat (corn oil/linoleic acid enriched) cause intestinal inflammation and impaired intestinal barrier defence in mice chronically fed alcohol. Alcohol. 2013 May;47(3):257-64. doi: 10.1016/j.alcohol.2013.01.005




Received on 21.12.2020           Modified on 11.07.2021

Accepted on 22.11.2021         © RJPT All right reserved

Research J. Pharm. and Tech. 2022; 15(7):3179-3184.

DOI: 10.52711/0974-360X.2022.00532