The Association of ACSL1 and UCP2 3′ UTR Polymorphisms With the Clinicopathological Characteristics of Patients With Colorectal Cancer in Serbia

Colorectal cancer (CRC) is a disease that affects millions of people worldwide. It is the third most commonly diagnosed type of cancer, and the second most common cause of death in both genders [1]. It is well known that intensively proliferating cancer cells can have unique metabolic patterns. Altered lipid metabolism is increasingly recognized as an important feature of colorectal cancer (CRC). The lipid metabolic pathways affecting CRC cells are synthesis, desaturation, elongation, and mitochondrial oxidation of the fatty acids (FAs) [2]. Although CRC can depend on glycolysis, colon cancer cells adapt to low glucose conditions and survive through shifting from glycolysis to oxidative phosphorylation (OXPHOS) [3, 4]. Cancer cells rely upon fatty acids (FA) for proliferation, survival, and metastasis and use them for the production of energy and membrane maintenance [5]. Fatty acids can trigger signals necessary for tumorigenesis, enabling cancer cells to migrate and generate distant metastasis [6]. The serum levels of FAs can potentially be used as predictive biomarkers in colorectal cancer tumors [7].

Fatty acid biosynthesis and β-oxidation are found to be increased in many cancers [5]. Both pathways require a common initial step known as fatty acid activation by acyl-CoA synthetase (ACS) [8,9]. Fatty acid activation, catalyzed by acyl-CoA synthetases (ACSs), is key in metabolizing extracellularly derived and de novo-synthesized fatty acids [10]. It has been observed that ACSs are frequently deregulated in cancer cells, which leads to the altered distribution and number of intracellular FAs [11]. In colorectal cancer, three main acyl-CoA synthetase long chain (ACSL) isoforms, ACSL1, ACSL4, and ACSL6, are upregulated [10]. Overexpression of ACSL1 in colorectal cancer cells can increase cell invasion and proliferation, and it has been associated with poor clinical outcomes [12]. A functional polymorphism rs8086 has been detected in the 3′-UTR region of theACSL1 gene. It causes substitution of cytosine (C) with thymine (T), and it has been previously shown that the TT genotype correlates with higher ACSL1 mRNA levels [12].

Most cellular reactive oxygen species (ROS) in mitochondria are produced through OXPHOS. A significant increase in oxidative stress markers has been observed in patients with CRC, suggesting a potential role for ROS in colon tumorigenesis and tumor progression [2,4]. ROS can influence the process of tumorigenesis by inducing DNA damage and nuclear genome instability and by increasing proliferation, survival, and migration of cells. Conversely, ROS can also function as anti-tumorigenic agents, since they can induce cellular senescence and cell death. The mechanisms by which cells respond to ROS depend on the location of ROS production and the concentration of different ROS species [13].

Uncoupling protein-2 (UCP2) is a mitochondrial inner membrane anion carrier which acts as a sensor of mitochondrial oxidative stress and a negative regulator of ROS production [14]. Loss of UCP2 function may result in increased generation of ROS [14]. Also, UCP2 participates in several metabolic processes, such as regulating food in-take, insulin secretion, and immune responses, and acts as a metabolic switch that promotes fatty acid metabolism over glucose utilization. However, the precise mechanism by which UCP2 modulates these processes and mitochondrial functions remains unclear [15]. One of the polymorphisms of the UCP2 gene is a 45bp ins/del polymorphism located in the 3′UTR region of exon eight at position +3474. This polymorphism’s biological function is still unclear, but the location in the 3′UTR suggests that it may influence mRNA processing or stability of the transcript, affecting protein expression [16].

Considering the importance of discovering new bio-markers that could help predict disease progression, we aimed to investigate the association of UCP2 45bp ins/del and ACSL1 rs8086 polymorphisms with the clinicopatho-logical characteristics of the disease.

Our study included 183 patients with colorectal cancer, with an average age of 62.1. One-hundred-and-ten (60.1%) were men, and 73 (39.9%) were women. Their main demographic features and clinicopathological characteristics are presented in Table 1.

Genotype and allele frequencies of ACSL1 rs8086 and UCP2 45bp ins/del polymorphisms are presented in Table 2. The results of the association analysis between ACSL1 and UCP2 genotypes and clinicopathological characteristics are presented in Tables 3 and 4.

There was no statistically significant association between the analyzed polymorphisms and patient age at the time of diagnosis, stage of the disease, tumor grade and presence of lymph node metastases. Carriers of ACSL1 rs8086 CC genotype had rectal localization of cancer significantly more often than patients with CT or TT genotype (p=0.006, OR=2.390 95% CI 1.278–4.496). Also, in patients that are carriers of the UCP2 gene II genotype, colon cancer was statistically more frequent than in patients with DD or ID genotype (p=0.017, OR=5.535, 95% CI 1.411–21.711). Although the result is statistically significant, the number of patients with II genotype is small. Only three patients (27.3%) with II genotype developed rectal cancer while 8 (72.7%) had colon cancer. We have also found that the analyzed polymorphisms are associated with the presence of distant metastases. Multiple logistic regression analysis has shown that distant metastases were significantly more frequent in TT genotype carriers than carriers of CC or CT genotype (β=0.822, p=0.040). Age and gender were used as covariates based on the results of univariate logistic regression analysis (Table 5). In the case of UCP2 polymorphism, carriers of UCP2 45 I allele (II+ID genotype) had distant metastases more frequently than patients with DD genotype (p=0.040, OR=4.120, 95% CI 1.093–15.534). This result remained significant after logistic multiple regression analysis (β=1.510, p=0.036) (Table 5). Multiple regression analysis has also confirmed that the age and gender of patients can influence the presence of distant metastases. Patients older than 50 years of age had distant metastases less frequently than younger patients (p=0.023, OR=0.187; 95% CI 0.044–0.793), while women had distant metastases more frequently than men (p=0.04, OR=4.396, 95% CI 1.069–18.085).

Identifying gene polymorphisms that could represent new biomarkers of disease progression is of great importance for patients with colorectal cancer. It could significantly improve the prognostic ability and facilitate the development of effective targeted treatments, thus improving patients’ survival rates.

ACSLs, necessary for activating long-chain fatty acids, are often deregulated in cancer cells. They can promote ungoverned cell growth, evade programmed cell death, and facilitate tumor invasion and migration [8,19]. So far, it has been found that ACSL1 is upregulated in colon, breast, and liver tissue [8]. Also, it has been shown that ACSL1 knockdown suppresses anchorage-independent growth and reduces the migration of cancer cells [5,10]. Vargas et al. showed that polymorphism rs8086 in the 3′ UTR of the ACSL1 gene affects gene expression, possibly through the change of miRNA binding affinity and mRNA stability. Patients’ carriers of the TT genotype had higher gene expression than carriers of the CC or CT genotype [12]. The increased expression of the ACSL1 gene in patients with the TT genotype may promote growth and migration of cancer cells.

The frequency of genotypes and alleles obtained in our study did not differ from the frequency in the general European population (https://www.ncbi.nlm.nih.gov/snp/rs8086); based on this, we can assume that rs8086 is not a susceptibility factor for colorectal cancer [20]. However, our results show that ACSL1 rs8086 polymorphism is associated with a higher risk for the presence of distant metastases in patients with CRC. In the previous study of Vargas et al., TT genotype was significantly associated with the clinical outcome of patients with colon cancer that had stage II or III of the disease. Patients, carriers of the TT genotype, had significantly decreased disease-free survival, with a 3-fold higher risk of relapse [12]. To our knowledge, this is the only study dealing with the association of rs8086 polymorphism and clinical outcomes of patients with colon cancer that has been published so far.

The expression of UCP2, coding for a member of the mitochondrial anion carrier proteins family, changes during the course of the disease. While it is repressed during the process of tumorigenesis, leading to increase of oxidative stress in cancer cells, the expression of UCP2 increases in the later stages of cancer development. Due to the increased UCP2 expression, cancer cells are protected from apoptosis by negative regulation of mitochondrial ROS production [21]. Another possible mechanism of UCP2 action in tumor cells, besides regulating reactive oxygen species production, is the regulation of glucose and fatty acid oxidation metabolism, which is directly associated with tumor cell proliferation [4]. Pecqueur et al. have shown that UCP2 +/+ provides the persistence of FA catabolism, which, on one side, leads to reduced cell proliferation. On the other hand, it enhances resistance to glucose starvation [22]. So far, it is not clear how the UCP2 expression is affected by the ins/del polymorphism. The location of this polymorphism in the 3′UTR region of the gene suggests that it may influence mRNA processing or transcript stability [16]. Also, positive linkage disequilibrium between the -866G/A (rs659366) polymorphism minor allele and the 45bp I allele has been observed. The minor allele (A allele) possibly has higher transcriptional activity than the wt allele. Other functional variants also have modulating functions, and their interaction leads to a complex pattern of associations [23]. Li et al. have found the association of UCP2 -866G/A with the survival rate after surgery. Patients with the GG genotype had the highest survival rate. The G allele has lower UCP2 mRNA expression when compared with the A allele. This suggests that higher UCP2 expression may promote cancer growth and survival [24]. It is possible that the presence of the I allele causes an increase in UCP2 expression, which provides better modulation of colorectal cancer cells metabolism and enables their survival, proliferation, and migration.

Our results show that the frequencies of UCP2 ins/del genotypes and alleles are in line with those published for other European populations [ 25, 26, 27], which suggests that this polymorphism does not represent a susceptibility factor for colorectal cancer. However, regarding the association with clinicopathological characteristics, we observed that carriers of the I allele have a higher risk of developing distant metastases. Marques et al. (2017) have shown that I allele could be associated with a decreased risk of CRC development. However, they did not find an association of the ins/del polymorphism with tumor location, TNM stage risk, relapse risk, and death risk [28]. To the best of our knowledge, no other studies are investigating this polymorphism’s association with the clinicopathological characteristics of patients with CRC. Kuai et al. investigated the association between the expression of UCP2 in colon cancer tissue and clinicopathological characteristics of the disease. They observed that UCP2 expression was significantly higher in patients with cancer stages III and IV in comparison with the patients with cancer stages I and II. Also, they have shown that UCP2 expression is associated with the development of colon cancer metastasis. Over-expression of UCP2 could be associated with tumor aggressiveness, which makes UCP2 a potential target for colon cancer therapy [29]. Horimoto et al. have shown that UCP2 expression is increased in colon cancer and correlates with the degree of neoplastic changes along the adenoma-carcinoma sequence [14]. In the study of Kyriazanos et al., 69 cases of gastric adenocarcinoma were analyzed, and it has been observed that tumors with high UCP2 expression were poorly differentiated, larger, had advanced T and N stages, increased number of infiltrated lymph nodes, lower apoptotic index and shorter disease-free survival [30].

Overview of the ACSL1 rs8086 and UCP2 45 bp ins/ del gene polymorphisms and gene expression functional relevance in colorectal cancer has been presented in Table 6.

We also have found that tumor site specificity depends on the UCP2 and ACSL1 genotype. Carriers of the UCP2 I allele had an increased risk for the development of colon cancer in comparison to rectal cancer while carriers of ACSL1 CC genotype had increased risk of developing rectal cancer. It is known that the biology and histopathology of the colon and rectum are distinct. Evidence shows that there is a difference in site-specific molecular pathways and enzyme expression patterns, but more data is necessary to better understand possible differences between the molecular biology of colon and rectal cancers [31, 32]. Our study has several limitations, however. This is a single center retrospective study with no prospective validation performed in the same or other populations at this moment. While the total number of patients is not small, when divided into groups based on genotype and factors analyzed, some of the groups appeared to be relatively small. Also, the lack of patient follow-up data after the surgery did not allow us to further explore the impact of the analyzed polymorphisms over time.

Gene polymorphisms analysis has a great potential for use in clinical settings since it is easy to perform, there is no need of invasive techniques and it is cost effective. Polygenic risk score (PRS) development may become a basis for personalized treatment. It may help clinicians to recognize high-risk patients earlier and provide personalized treatment, potentially with more aggressive strategy. However, it is still considered only investigational. Main challenges for application are small effect sizes, missing heritability and lack of knowledge regarding gene-gene and gene-environment interactions [33].

In conclusion, our results suggest that ACSL1 rs8086 and UCP2 45bp ins/del polymorphisms may contribute to the occurrence of distant metastases and tumor location in patients with colorectal cancer. Subsequently these polymorphisms, as components of a polygenic risk score, might become biomarkers for tumor location and occurrence of distant metastases. Additional studies on a larger number of patients, are needed to confirm the results obtained here and to further investigate the potential importance of these genetic variants for patients with colorectal cancer.