Weicng Wng, Jinn Zhng, Guodong Zhng,b,?
a Department of Food Science, University of Massachusetts-Amherst, Amherst, MA, 01003, USA
b Molecular and Cellular Biology Program, University of Massachusetts-Amherst, Amherst, MA, 01003, USA
Keywords:Linoleic acid Polyunsaturated fatty acids ω-3 Fatty acids Colon cancer Colonic inflammation Cytochrome P450 Eicosanoids
A B S T R A C T Human consumption of linoleic acid (LA, 18:2ω-6, abundant in vegetable oils) is very high. Animal experiments showed that excessive LA intake increased azoxymethane-induced colon tumorigenesis,however, the impact of excessive LA on colon cancer in human is not conclusive, making it difficult to make dietary recommendations for optimal intake of LA. Understanding the molecular mechanisms of LA on colon tumorigenesis could help to clarify its health effect, and facilitate development of mechanismbased strategies for preventing colon cancer. Recent studies show that the previously unappreciated cytochrome P450 monooxygenase-mediated eicosanoid pathway is upregulated in colon cancer and plays critical roles in its pathogenesis, and could contribute to the effects of dietary LA, as well as ω-3 fatty acids, on colon tumorigenesis. In this review, we will discuss recent studies about the roles of cytochrome P450 monooxygenases in fatty acid metabolism and its roles in colonic inflammation and colon cancer, and how this information could help us to clarify the health impacts of dietary fatty acids.
The consumption of linoleic acid (LA, 18:2ω-6), which is abundant in vegetable oil products (such as corn, soybean, and canola oils, as well as fried food, salad dressing, and mayonnaise), is very high in the United States: the average intake of LA is as high as~12-17 g/person/day [1]. The potential issue with excessive intake of LA is exacerbated because the consumption of LA-rich vegetable oil has significantly increased in the United States in the last century [1], and could be even more prevalent in the future as dietary recommendation continues to encourage greater consumption of LA-rich vegetable oil products [2]. Indeed, the consumption of LArich soybean oil has risen more than 50% since 1980 and more than 1000-fold since 1909 [1]. The impact of high dietary intake of LA on human health is an intensively debated topic [3]. Multiple animal experiments have shown that excessive intake of LA exacerbated azoxymethane (AOM)-induced colon tumorigenesis,suggesting its potential adverse effect on colon cancer [4–7]. Epidemiological studies also support that high dietary intake of LA is linked with increased risks of colitis and colon cancer [8–14].Notably, the European Prospective Investigation into Cancer and Nutrition (EPIC) study has shown that high intake of LA doubled the risks of colitis, and could be responsible for ~30% of ulcerative colitis cases [8]. However, the impact of LA on colon cancer and colonic inflammation in humans remains inconclusive, making it difficult to make dietary recommendations for optimal intake of LA [3].Exploring the molecular mechanisms of LA on colon tumorigenesis could contribute to clarify its health effect, identify the individuals who are susceptible to excessive LA intake, and facilitate the development of mechanism-based strategies for preventing colon cancer.
Fig. 1. Multiple dietary PUFAs, including linoleic acid (LA, 18:2ω-6), α-linolenic acid (ALA, 18:3ω-3), dihomo-γ-linolenic acid (DGLA, 20:3ω-6), ARA (ARA, 20:4ω-6), eicosapentaenoic acid (EPA, 20:5ω-3), and docosahexaenoic acid (DHA, 22:6ω-3), are substrates of COX, LOX, and CYP enzymes. This leads to formation of a large array of metabolites with diverse chemical structures and distinct biological actions.
A widely recognized explanation for the health effects of dietary polyunsaturated fatty acids (PUFAs), including LA, is that PUFAs could be enzymatically metabolized to generate lipid mediators (LMs), which are lipid signaling molecules with potent biological actions and play critical roles in the pathogenesis of many human diseases [15–17]. After dietary consumption, the dietary PUFAs are partly incorporated into membrane phospholipids [15–17]. Upon cellular stimulation such as inflammation,the membrane-incorporated PUFAs are released mainly by phospholipase A2(PLA2) to generate free-form intracellular PUFAs,which are then converted by a series of enzymes to generate LMs [15–17]. There are three major enzymatic pathways participated in PUFA metabolism: cyclooxygenase (COX), lipoxygenase(LOX) and cytochrome P450. The COX pathway converts PUFA into prostaglandins, which are well-known mediators of inflammation and pain [16]. The LOX enzymes produce hydroxyl fatty acids and leukotrienes, which are predominately pro-inflammatory mediators playing critical roles in inflammatory disorders such as asthma[16]. The cytochrome P450 monooxygenases convert PUFAs to the corresponding epoxygenated fatty acids (EpFAs) [17,18]. For example, the metabolism of LA by cytochrome P450 monooxygenases generates two regioisomers of 9,10- and 12,13-epoxyoctadecenoic acids (EpOMEs), and the metabolism of arachidonic acid (ARA,20:4ω-6, a downstream metabolite of LA) by cytochrome P450 monooxygenases generates epoxyeicosatrienoic acids (EETs) that include four regioisomers of 5,6-, 8,9-, 11,12- and 14,15-EET [17].These EpFAs are then metabolism by soluble epoxide hydrolase(sEH, encoded by Ephx2) to generate the corresponding fatty acid diols [19]. Multiple dietary PUFAs, including LA, ARA, α-linolenic acid (ALA, 18:3ω-3), γ-linolenic acid (γ-LA, 18:3ω-6), dihomoγ-linolenic acid (DGLA, 20:3ω-6), eicosapentaenoic acid (EPA,20:5ω-3), and docosahexaenoic acid (DHA, 22:6ω-3), are substrates of these PUFA-metabolizing enzymes (Fig. 1). This leads to formation of a large array of LMs with diverse chemical and biological properties.
While previous mechanistic research of PUFA metabolism has focused on the COX and LOX pathways [21,22], the roles of cytochrome P450 pathway (the other major pathway besides COX and LOX branches) are not well understood. Recent research supports that the previously unappreciated cytochrome P450 monooxygenase pathway is upregulated in colon cancer, and play vital roles in its pathogenesis. Furthermore, recent research suggests that the cytochrome P450-mediated conversion of dietary LA to EpOMEs could contribute to the colon tumorigenesis-promoting effects of cytochrome P450 monooxygenases [23]. These results suggest a potential novel mechanistic explanation about dietary LA consumption and risks of developing colon cancer. In this essay, we will focus on recent advances in the roles of cytochrome P450 monooxygenases in fatty acid metabolism and its roles in colonic inflammation and colon cancer, and how this information could help us to better clarify the health impact of dietary PUFAs.
Recent studies support that cytochrome P450 monooxygenases are overexpressed in colon cancer in human. More than 70% of tested colon tumor samples have been found to have positive staining of CYP2C9, while it was not found in matched human benign samples [24]. Additionally, CYP2J2 was reported to increase in human colon tumor samples [25]. In addition to CYP2C/2 J enzymes, other cytochrome P450 enzymes, including CYP2W1, CYP1B1, CYP2S1, CYP2U1, CYP3A5, and CYP51, have been found to highly expressed in human colorectal cancer samples compared to normal colorectal samples [26,27]. Further studies are needed to verify and study each individual cytochrome P450 enzyme expression regulation during colon tumorigenesis.
Consistent with the human studies, cytochrome P450 monooxygenases were found to overexpress in colon tumor tissues in vivo and colon cancer cells in vitro, and the concentrations of cytochrome P450-derived LMs were increased in the circulation and colon of colon cancer mice [28]. Indeed, increased expression of human cytochrome P450 genes, such as CYP2C8,CYP2C9, CYP2C19, and CYP2 J2, was been found in human colon cancer cell lines, compared with the normal human colon cell line. In the well-established AOM/dextran sodium sulfate (DSS)-induced colon cancer model, the gene expression of a series of mouse Cyp genes, including Cyp2c38, Cyp2c39,Cyp2c55, Cyp2c65, Cyp2c70, Cyp2j6, Cyp2j9, and Cyp2j13, was increased in colon of AOM/DSS-induced colon cancer mice[28].
There are many possible explanations by which cytochrome P450 monooxygenases are increased in colon tumors. Hypoxia,which is commonly found in the tumor microenvironment, has been shown to increase the expression of cytochrome P450 monooxygenases [18,29]. It is feasible that the hypoxic microenvironment could contribute to the enhanced expression of cytochrome P450 monooxygenases in colon tumors. To support this notion, a previous study showed that CYP2 J2 was overexpressed in the majority (77%) of human samples from many tumor types, including breast, stomach, liver, colon adenocarcinoma, and pulmonary squamous cell carcinoma, adenocarcinoma, small cell carcinoma [25].
A recent study showed that genetic deletion or pharmacological inhibition of cytochrome P450 monooxygenases attenuated AOM/DSS-induced colon cancer in vivo [28], supporting a functional role of cytochrome P450 monooxygenase pathway in colon tumorigenesis. Compared with AOM/DSS-induced Cyp2c+/+mice,there was lower tumor number and total tumor burden, as well as reduced expression of proliferation marker such as PCNA,in the AOM/DSS-induced of Cyp2c+/?mice [28]. Consistently,dietary administration of cytochrome P450 inhibitor, SKF-525A or clotrimazole, suppressed colon cancer development in mice [28].Treatment of SKF-525A or clotrimazole reduced tumor burden,attenuated colonic expression of proliferation marker PCNA and increased expression of apoptosis marker cleaved caspase-3, further supporting the roles of cytochrome P450 monooxygenases in the development of colon cancer in mice [28]. Together, these results demonstrate that the cytochrome P450 monooxygenases play important roles in colon tumorigenesis. To verify and explore the biological significance of cytochrome P450 monooxygenases in colon cancer development, further studies are needed to study the role of cytochrome P450 monooxygenases in other colon cancer models.
Colonic inflammation has been recognized as the major driving force in promoting colonic tumorigenesis. Therefore, targeting inflammation would be a good strategy to inhibit colon carcinogenesis. Recent studies support the contribution of cytochrome P450 monooxygenases in colon tumorigenesis-related inflammation [23]. In the AOM/DSS-induced colon cancer model, there was decreased expression of pro-inflammatory cytokines Tnf-α and Il-1β in colon tissue of Cyp2c+/?mice, compared to Cyp2c+/+mice[23]. Consistently, treatment of the cytochrome P450 inhibitors led to reduced AOM/DSS-induced expression of pro-inflammatory cytokines Tnf-α, Mcp-1, Il-6, Il-1β, and Ifn-γ in colon tissue [23].These results demonstrate that inhibition of cytochrome P450 monooxygenases reduces colon cancer-associated inflammatory responses. However, it remains to determine whether cytochrome P450 inhibition or reduced colon tumorigenesis is the direct reason to cause attenuated inflammation.
Besides colon cancer, previous studies support that cytochrome P450 monooxygenases are involved in the tumorigenesis of other types of cancer. CYP2C8, CYP2C9 and CYP2 J2 have been found to be up-regulated in stomach, liver, ovarian cancer, renal carcinoma and testicular seminoma, compared to adjacent normal tissue [24,25].In a xenograft tumor model, overexpression of human CYP2C8 in endothelial cell (using Tie2-CYP2C8 Tr mice) led to increased tumor growth of B16F10 melanoma and T241 fibrosarcoma, and Lewis lung carcinoma (LLC) in C57BL/6 mice [30]. Endothelial overexpression of CYP2 J2 (using Tie2-CYP2 J2 Tr mice) also led to increased xenograft growth of B16F10 melanoma and T241 fibrosarcoma in mice [30]. CYP2C11 was found to be increased brain tumor tissue compared to the normal tissue, and treatment of cytochrome P450 monooxygenase inhibitor, 17-octadecynoic acid or miconazole, significantly inhibited glial tumor formation in Fischer rat RG2 cell-induced glioblastoma model, suggesting a potential role of cytochrome P450 monooxygenase in brain tumorigenesis [31].Human studies support that genetic polymorphisms of CYP2C8 and CYP2C9 are correlated with decreased survival in breast cancer patients [32]. Together, these results demonstrate a critical role of cytochrome P450 monooxygenases in many types of cancer.
Previous studies support that EETs, which are cytochrome P450 metabolites derived from ARA, contribute to the pro-cancer effects of cytochrome P450 monooxygenases. Panigrahy et al. found that treatment with 14,15-EET, at a dose as low as 15 μg/kg/day,increases primary tumor growth in several models, including xenograft LLC model, transgenic adenocarcinoma of the mouse prostate (TRAMP) mouse model, human liposarcoma in SCID mice,and orthotopic PC3M-LN4 human prostate cancer in SCID mice[30]. These results demonstrate that EETs have extremely potent pro-cancer effects in vivo, supporting that these metabolites are critical in mediating the biological activities of cytochrome P450 monooxygenases on tumorigenesis.
The troll said that he would always be welcome; he had served him faithfully for the three years they had agreed upon, and he could make no objections to his leaving now
Angiogenesis, which is the process of formation of new blood vessels from pre-existing vessels, is a hallmark of tumorigenesis [33]. Previous studies have shown that cytochrome P450 monooxygenases and their derived metabolites enhance tumor angiogenesis, resulting in increased tumorigenesis. In a xenograft mouse model, endothelial overexpression of human CYP2C8 or CYP2 J2 led to increased formation of CD31-positive vessels in B16F10 tumors, illustrating enhanced tumor angiogenesis [30]. Similarly, in a corneal tumor angiogenesis model,overexpression of CYP2C8 in endothelial cells increased neovascularization in the cornea [30]. Systemic treatment with the cytochrome P450 metabolites EETs also increased tumor angiogenesis. Administration of 14,15-EET in the TRAMP mice increased the formation of blood vessels in prostate tumor [30]. These results demonstrate that cytochrome P450 monooxygenases and their metabolites have potent biological activities to promote tumor angiogenesis, which could contribute to their pro-cancer effects.
Cytochrome P450 monooxygenases also contribute to tumor metastasis by increasing tumor cells to spread from its primary location to other organs, which is the reason for ~90% of cancerrelated death [34]. Panigrahy et al. showed that cytochrome P450 monooxygenases and metabolites EETs play critical roles in promoting tumor metastasis, and therefore could be important therapeutic targets to treat metastasis [30]. Indeed, using a LLCresection metastasis model, endothelial overexpression of human CYP2C8 or CYP2 J2 led to increased surface and invasive metastases in lung tissues [30]. In a hematogenic metastasis model, overexpression of CYP2C8 also led to increased lung and liver metastases[30]. Consistent with the pro-metastatic effect of cytochrome P450 monooxygenases, treatment with EETs also promoted tumor metastasis. In the LLC-resection metastasis model, administration of 14,15-EET at a dose of 15 μg/kg/day increased axillary and inguinal lymph node metastasis, and finally lung metastasis in C57BL/6 mice [30]. These results support that the cytochrome P450 monooxygenase eicosanoid pathway plays critical roles in regulating tumor metastasis. Considering the critical roles in metastasis in cancer patient survival, it is of critical importance to study the roles of cytochrome P450 monooxygenases in metastasis and develop targeted therapies.
Recent studies support that EpOMEs, which are the metabolites of LA produced by cytochrome P450 monooxygenases, play critical roles in inducing inflammatory responses and enhancing colon tumorigenesis. In the AOM/DSS-induced colon cancer model,increased concentrations of 9,10- and 12,13-EpOMEs were been found in the circulation and colon tissue of colon cancer mice. These two metabolites are also among the most abundant lipid metabolites detected in the circulation and tissues of mice, which could be at least in part due to the high abundance of LA in tissues [28].Treatment with low-dose 12,13-EpOME via mini-pumps increased tumor number, size, and total burden, and enhanced colonic expression of tumorigenic markers in colon tissue of C57BL/6 mice in colon cancer model. Consistent with the animal data,treatment with 12,13-EpOME at a concentration of 100 nmol/L increased production of pro-inflammatory cytokines and induced JNK phosphorylation in both macrophage and colon cancer cells[28]. Together, these results demonstrate the pro-inflammation and pro-cancer effects of EpOMEs in colon tumorigenesis.
These results are largely in agreement with previous studies which showed that EpOMEs have been associated with a series of detrimental effects, including chemotaxis, inflammation, cardiovascular diseases, and pulmonary injury [35–40]. In human studies,EpOMEs have been found to contribute to respiratory distress syndrome and even multiple organ failure in severe burn patients,and were termed as “l(fā)eukotoxins” [36,38,41]. In Wistar rats, treatment of EpOME led to alveolar epithelial and endothelium damage,promoted neutrophil infiltration into lung tissues, and induced pulmonary edema and acute lung injury [37]. In cell culture studies,EpOMEs enhanced neutrophil chemotaxis at the dose of 10 nmol/L,demonstrating the potent pro-inflammatory effects of EpOMEs[40].
EpOMEs is biological unstable and could be further converted into corresponding fatty acid diols termed dihydroxyoctadecenoic acids (DiHOMEs) by sEH [18]. DiHOMEs have also been shown to induce chemotaxis, tissue injury, and cause mortality in animal models [41,42]. In both zymosan-induced and Freund’s Adjuvant-induced inflammatory pain models, the concentrations of 12,13-DiHOME were increased in peripheral nervous tissue in mice [43]. Injection of 12,13-DiHOME led to thermal hyperalgesia caused inflammatory pain via transient receptor potential vanilloid 1 (TRPV1)-dependent mechanisms [43]. In addition, administration of DiHOMEs could cause massive bronchiolar, peribronchiolar,and alveolar edema in the lungs of Swiss Webster mice37. The toxic effect of DiHOMEs might due to cellular mitochondrial dysfunction.DiHOMEs was found to change mitochondrial inner membrane permeability and induce mitochondrial membrane leakage in living cells, resulting in the release of cytochrome c and causing cell toxicity [44]. Together, these results show that not only EpOMEs but also their down-stream metabolites contribute to a series of detrimental biological effects.
As discussed above, recent research has shown that: (1) the concentrations of 9,10- and 12,13-EpOME, which are metabolites of LA produced by cytochrome P450 monooxygenases, are significantly increased in the plasma and colon of AOM/DSSinduced colon cancer mice; (2) treatment with 12,13-EpOME, at nM doses, increases production of pro-inflammatory cytokines and induced JNK phosphorylation in both macrophage and colon cancer cells; (3) treatment with low-dose 12,13-EpOME, via mini-pumps,exacerbates AOM/DSS-induced colon tumorigenesis in mice; (4)the enzymes that produce EpOMEs, cytochrome P450 monooxygenases, are overexpressed in colon tumor tissues in vivo and colon cancer cells in vitro; and (5) pharmacological inhibition or genetic ablation of cytochrome P450 monooxygenases attenuates AOM/DSS-induced colon tumorigenesis in vivo [28]. Together, these results support that the previously unappreciated cytochrome P450/EpOME axis could play a critical role in the carcinogenesis of colon cancer, and could mediate the colon cancer-enhancing effects of dietary LA.
Based on these findings, overconsumption of LA could increase tissue concentrations of EpOMEs, which could enhance colon inflammation and tumorigenesis, therefore could result in enhanced colonic inflammation and colon cancer. A recent study by Taha et al. [45] showed that the circulating concentration of EpOMEs, but not other metabolites, was dose-dependently increased with increasing levels of LA in the diet: after a 15-week treatment with diet containing 0.4%, 5.2%, and 10% calories LA,the plasma concentrations of EpOMEs in rats were ~27.9, 175.4,and 346.3 pmol/mL, respectively. These results support a critical role of EpOMEs in mediating the biological actions of dietary LA.Based on these studies, the polymorphisms in the genes encoding cytochrome P450 monooxygenases, as well as its downstream fatty acid-metabolizing enzyme sEH, could affect the metabolism of LA to generate EpOMEs [46–52]. This could in part explain the inconsistent results from human studies. Further study of the biological significance of cytochrome P450/EpOME pathway in colon tumorigenesis could establish a new mechanistic linkage between excessive LA intake and increased risks of colon cancer. The obtained results could provide the fundamental knowledge that is critical to perform better designed human studies to clarify the health impact of dietary LA, and broadly enhance our understanding of how diet-gene interactions affect human health.
A well-known theory to explain the effects of ω-3 PUFAs is that they suppress the enzymatic metabolism of ω-6 PUFAs(such as ARA) to generate ω-6-series LMs that are predominately pro-inflammatory and pro-angiogenic, or they act as alternative substrates to form ω-3-series LMs which usually have less detrimental or beneficial effects [53–56]. For example, EPA can effectively compete with ARA for metabolism by COX-2, leading to the reduced level of ARA-derived prostaglandin E2(PGE2)which potently induces inflammation and tumor progression, and increased levels of EPA-derived prostaglandin E3(PGE3) which is less detrimental [56,57]. The 4-hydroxy-docosahexaenoic acid (4-HDHA), the metabolite of DHA from 5-lipoxygenase (5-LOX), has potent anti-angiogenic effect; while the corresponding metabolite from ARA stimulates angiogenesis [55].
Recent research supports that the cytochrome P450 monooxygenase pathway could play critical roles in mediating the health-promoting effects of ω-3 PUFAs. Animal and human studies support that ω-3 PUFAs are mainly metabolized by the cytochrome P450 pathway [58–62], and they are known to be relatively poor substrates of other PUFA-metabolizing enzymes, such as COX and LOX [63–65]. In addition, cytochrome P450-derived ω-3 LMs,notably EDPs derived from DHA, have been shown to potently inhibited tumor growth, angiogenesis and metastasis [18,66,67].A recent study showed that treatment with EDPs via mini-pumps at a dose of 0.5 mg/kg/day inhibited xenograft colon tumor growth in mice [68]. These findings are supported by other studies, which showed that EDPs had potent actions to suppress angiogenesis in the mouse model of laser-induced macular degeneration [69,70].Treatment with 19,20-EDP inhibited pathological angiogenesis during macular degeneration by reducing the size of choroidal neovascularization lesions and suppressing vascular leakage [69].Moreover, administration of 19,20-EDP inhibited choroidal neovascularization progression by regulating leukocyte recruitment in local microenvironment of choroidal neovascularization lesions and reducing the expression of adherent molecules on both leukocytes and endothelial cells [70]. These results suggest a potential role of the cytochrome P450 monooxygenase pathway in mediating the beneficial effects of ω-3 PUFAs.
Colon cancer is a major public health issue: it is estimated to cause 140,250 new cases and 50,630 deaths per year in the United States, making colon cancer the third most common type of cancer and the second leading cause of cancer death in the United States[71]. A better understanding of the molecular linkage between dietary PUFA consumption and colon cancer risks could help to develop safe and effective strategies for preventing colon cancer,and therefore leads to significant and positive impact on public health.
Recent research supports that the previously unappreciated cytochrome P450 monooxygenase pathway could play a critical role in mediating the opposite actions of ω-6 PUFAs versus ω-3 PUFAs on colon cancer. Indeed, studies show that the ω-6-series cytochrome P450 metabolites, including EETs derived from ARA and EpOMEs derived from LA, promote tumorigenesis of colon cancer and potentially other types of cancer [28,30]; in contrast,the ω-3-series cytochrome P450 metabolites, notably EDPs derived from DHA, potently inhibits angiogenesis, tumor growth, and tumor metastasis [66,68]. Further exploration of the cytochrome P450 monooxygenase pathway could help us to clarify the health impacts of dietary PUFAs.
Recent advances in molecular pathological epidemiology (MPE)demonstrate that various exogenous and endogenous factors from dietary, immune response and gut microbial will work together to influence colon tumorigenesis [72–74]. Indeed, ω-3 PUFA intake has been associated with a lower risk of colorectal carcinomas only with increased regulatory T cell in tumor tissue instead of lower regulatory T cell density [75]. Additionally, dysregulation of the gut microbial could also contribute to colon tumorigenesis via increasing pro-inflammatory response. For example, intake of Western diets rich in red and processed meat is associated with increased risk of colorectal cancer when there is the presence of the intestinal Fusobacterium nucleatum in tumor tissue, while not with a risk in F. nucleatum-null cancer [76]. F. nucleatum has been found to promote colorectal carcinogenesis through downregulating antitumor T cell-mediated adaptive immunity and activation of E-cadherin/β-catenin signaling pathways in colon cancer cells[77,78]. More studies are needed to further elucidate the interaction between host cytochrome P450 monooxygenase pathway and gut microbiota, such as F. nucleatum, in regulating the immune response and development of colon cancer.
Declaration of Competing Interest
The authors declare no conflict of interest.
Acknowledgments
This research is supported by a new faculty start-up from the University of Massachusetts Amherst, USDA NIFA2016-67017-24423, USDA NIFA 2019-67017-29248, USDA/HatchMAS00492,and NIH/NCIR03 CA218520 (to G.Z.).