Introduction

Finding a disease, the course of which is not related to lipid alterations can be challenging. Lipids raise a growing interest as potential biomarkers in many clinical conditions. This highlights the importance of lipidomic studies in understanding, diagnosing and treating numerous human pathologies, among them cancer; the use of lipidomics could create an opportunity to design targeted therapies, prognostic or screening biomarkers [1]. In everyday clinical practice, lipid status is estimated based on serum concentrations of total cholesterol (TC), high density lipoprotein (HDL), low density lipoprotein (LDL) and triacylglycerols (TGs). While only a limited information can be obtained from the analysis of those lipid fractions, other currently available techniques, e.g. mass spectrometry, may provide a detailed insight into the structure and function of some specific lipid species. In this review paper, we discuss lipid alterations associated with colorectal cancer (CRC), with special emphasis on fatty acids (FAs) and their potential therapeutic and diagnostic applications in patients with this malignancy.

Most cancers found in the colon or rectum are adenocarcinomas arising from pathological lesions in the epithelial cells of colorectal mucosa [2]. Vast majority of CRCs are thought to evolve from conventional adenomas through as a result of several dozens of mutations; this process is referred to as the adenoma-to-carcinoma sequence [3]. Most CRCs are sporadic malignancies and are not associated with inherited mutations in established cancer-related genes [4]. However, about 20–30% CRC may be associated with inherited mutations [5]. A progressive accumulation of multiple genetic mutations contributes to transition from normal mucosa to benign adenoma, severe dysplasia, and eventually, a frank carcinoma. It is estimated that approximately 15% of sporadic colon cancers are a consequence of malfunction in mismatch repair genes, whereas other 80–85% are associated with mutations in adenomatous polyposis coli (APC) gene. Furthermore, colon cancer may develop as a consequence of inflammatory bowel disease, on a different, yet uncharacterized pathway. Malignant transformation requires further genetic alterations [6]. Less than 50% of colon cancers harbor mutated KRAS, a protein that is involved in intracellular signal transduction [7, 8]. Approximately 50% of colonic lesions with high-grade dysplasia and about 75% of frank cancers may carry p53 mutations [6, 7]. A neoplastic disease cannot be effectively managed without the understanding of distinctive characteristics of cancer cells that contribute to tumor development. One of them is enhanced proliferation [9]. Two main genetic defects found in CRC, KRAS and p53 mutations, are both associated with enhanced proliferation [10, 11]. Intensively proliferating cancer cells display some unique metabolic patterns due to which they may obtain enough energy for new biomass synthesis. Cancer cells have a unique ability to generate energy in a nutrient-deficient environment. Since the preference of cancer cells for glycolysis rather than oxidative phosphorylations (OXPHOs) when oxygen is not limited has been demonstrated by Otto Warburg [12], the aberrant glucose metabolism became one of the hallmarks of cancer. However there has been a paradigm shift towards so called reversed Warburg effect, since research showed that each cancer has its unique metabolic features, and some may synthesize ATP by means of OXPHOs [13]. A recent evidence suggests that CRC cells rely on the reversed Warburg effect [14, 15], which opened new perspectives for the identification of new molecular therapeutic targets, among them FA oxidation [34] and cardiovascular diseases [35]. A growing number of studies analyzed the relation between lipids and various malignancies: breast cancer [36, 37], prostate cancer [38, 39], ovarian cancer [177]. The authors of one lipidomic study demonstrated considerable alterations of several complex plasma lipids in CRC patients, and based on the analysis of receiver operating characteristic (ROC) curve proposed phosphatidylglycerol PG-18:0/16:0, sphingomyelin SM-d18:1/24:1 (42:2), ceramide Cer-d18:1/26:4 (elevated), LPC-18:3, LPC-18:2, phosphorylethanolamines PE-18:2/18:1, PE-18:1/20:2 and SM-38:8 (decreased) as biomarkers of this malignancy [89]. The use of biomarker clusters may have greater discriminative power than single molecules. In one study, patients with early CRC were identified accurately based on their serum levels of palmitic amide, oleamide, hexadecanedioic acid, 12-hydroxystearic acid, 20:3 n-3, 14:0, lysophosphatidic acid LPA-16:0, LPA-18:0 and LPC-16:0, with the area under the ROC curve equal 0.991, 0.981 sensitivity and 1.000 specificity [178]. Similar approach, with a panel of various metabolites, among them lipids, was also used to predict the recurrence and spread of CRC and survival in patients with this malignancy [206]. In another study, rectal cancer patients showed a significant increase in serum TG 56:6, 52:2 and 52:1, but it must be stressed that the study group was relatively small [89]. The authors of most studies analyzing TG levels in blood and tissues of CRC patients reported their overall concentrations but did not provide a detailed information about the content of specific FAs.

Specific fatty acids changes in adipose tissue of CRC patients

Although available data on FAs esterified in TGs are generally limited, some studies provided an insight into this lipid group, based on the analysis of adipose tissue. The latter is the main reservoir of TGs, capable of releasing them into bloodstream, and thus, it may influence the lipid profiles of various tissues. Many studies documented a relationship between obesity and colorectal cancer risk [58, 59]. Abdominal fat deposits, which can be expressed as the waist-to-hip ratio, seem to be a predominant “measure” of colorectal adenoma risk in men and women [207]. Moreover, as outlined recently in the review articles published by Himbert [208] and Masoodi [209]; also multifaceted interactions between adipose microenvironment and tumor, especially those mediated by proinflammatory factors, raise a growing interest of researchers. Thus, adipose tissue is no longer considered a merely energy reservoir, but also a source of various signaling molecules, adipokines [210], and FAs with proinflammatory properties that can modulate immune cells [211] or activating autophagy [212]. Furthermore, adipose tissue is no longer analyzed as a single entity, but as two distinct compartments, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT). Furthermore, studies of SAT sometimes consider additional heterogenic nature of this tissue, with two distinct layers, deep and superficial one, that differ in terms of various parameters, e.g. the intensity of lipolysis [213,214,215,216]. Surprisingly, however, only few previous studies analyzed a link between CRC occurrence or progression and the content of some specific FAs in adipose tissue, showing some significant changes of their levels [92, 213, 217, 218].

The authors of one study published in 1988 found no significant intergroup differences in the content of seven major FAs determined by means of GLC-FID in SAT and RBCs from 49 CRC patients and 49 sex- and age-matched controls [218]. Also another case-control study conducted by Giuliani et al. [92] showed no significant differences in total SFAs or MUFAs content between SAT and VAT for both controls and patient. Total SFA content in VAT and total MUFA content in SAT turned out to be higher in CRC patients than in the controls (p < 0.001). Among specific PUFAs, CRC patients presented with higher levels of visceral 18:3 n-3 whereas lower 18:4 n-3 than the controls. Furthermore, the study showed that in CRC patients, the level of n-6 PUFA, 18:2 n-6, was significantly higher in SAT than in VAT. Finally, the total content of n-6 PUFAs (LA + GLA + DGLA + AA) in SAT was shown to be higher in healthy controls than in CRC patients.

A somehow different approach was presented by Cottet et al. [217], who analyzed subcutaneous adipose tissue FAs based on the estimated activity of the enzymes involved in their metabolism. Therein the putative marker for ELVOL5 + Δ-6-desaturase activities estimated by 20:3 n-6 to 18:2 n-6 ratio as well as ELVOL2/5 activity (ratio of 22:4 n-6 to 20:4 n-6 and 22:5 n-3 to 20:5 n-3) were positively associated with CRC risk. No such association with CRC risk was observed on the basis of 18:1 n-9 to 16:1 n-9 ratio.

One limitation of adipose tissue studies is the method of sample preparation, which has already been shown to influence FA concentration [219]. Furthermore, adipose tissue collection is an invasive procedure, and hence, is unlikely to be applicable to large-scale studies.

Conclusions

Despite a decrease in mortality, CRC still remains a serious public health burden [26]. A growing number of CRCs are diagnosed in patients younger than 50 years [220, 221]. The reason for this alarming tendency is yet to be elucidated, but it may be a consequence of greater exposure to environmental factors, lesser physical activity and unfavorable dietary changes. Analysis of lipid metabolism in cancer patients may provide a better insight into metabolic disturbances that contribute to carcinogenesis. The fact that cancer cells require lipids to proliferate [20], may open new therapeutic perspectives: perhaps some specific pathways involved in the synthesis and storage of fatty acids might be targeted to prevent cancer development [24]. Furthermore, some metabolites of fatty acids are important signaling molecules involved in the maintenance of proinflammatory and anti-inflammatory equilibrium. Probably these are proinflammatory factors which constitute a link between obesity and CRC [208]. Moreover, obesity is associated with lipidome changes [32] that may predispose to the development of some related conditions, among them cancer. Alterations of FAs, their metabolites and lipid species containing FA chains can be observed in tumor microenvironment as well (Table 1). Some of those alterations, such as accumulation of PC-16:0/16:1, may be considered as cancer biomarkers [176]. Lipid profile alterations, e.g. presence of cerotic acid [22] or a decrease in the concentration of hydroxylated, polyunsaturated ultra-long-chain fatty acids [169], can be also found in the sera of CRC patients, differentiating between early and advanced stages of this malignancy [178], or serving as a predictor of survival [179]. However, the development of clinically useful lipid biomarkers requires consistent research methodology, and previous studies were quite heterogenous in this matter. Another drawback of previous studies is limited sample size which may hinder generalization of their results onto the whole population of CRC patients. Nevertheless, understanding of lipid alterations associated with CRC may define new directions in the diagnosis and treatment of this malignancy.

Table 1 Colorectal cancer related changes of lipid species content in various biological samples
Full size table