Keywords

1 Introduction

Those who are given to the luxuries of the table are preparing for the pleasures of the operator’s table—(Hubbard 1927)

As evidenced by the quotation above, the effects of diet on human health have been considered for many years. The impact of nutrition in cancer is currently a major focus, for scientists as well as for the public. Accordingly, the influences of dietary fatty acids (FAs) on cancer prevention, promotion, and treatment have been extensively studied. Nonetheless, the groundbreaking de-orphanization of FFA1 and FFA4 as receptors for omega-3 FAs (n-3 FAs) and other long-chain fatty acids occurred in the context of inflammation and metabolic disease (Oh et al. 2010). FFA2 and FFA3 are receptors for butyrate and other short-chain fatty acids, which are produced by intestinal microbiota, and have long been known to be important in the physiology and pathophysiology of the digestive tract as well as in immune function (Ang and Ding 2016). Differences in the expression of FFARs between different tissues also directed early investigations into the physiologic roles of these receptors in particular organs. Thus, it is logical that the first studies of FFARs, as well as the development of pharmacologic ligands for these receptors, were focused on conditions other than cancer. Fortunately, it took only a few years for investigations of the newly identified FFARs to progress into the pathophysiologic realm of cancer. As the field moves forward, the roles of FFARs in neoplastic cells are beginning to provide intriguing new insights.

The situation with regard to cancer is particularly complex. While fatty acid metabolism in cancer cells has been studied for decades, epidemiologic studies examining fatty acid intake and cancer risk have yielded mixed results that are variable between different cancers. For example, omega-3 FAs are generally viewed as preventative or therapeutic for breast cancer (Chung et al. 2014; Hopkins et al. 2016; Yang et al. 2014), but may be detrimental for colon cancer (Wu et al. 2013). In the case of prostate cancer, most studies suggest that n-3 FAs lower risk; one analysis to the contrary generated a firestorm of response from scientists and the public (reviewed by Hopkins and Meier 2016). Clinical data indicate that n-3 FAs can increase response to chemotherapy (reviewed by Laviano et al. 2013). Dietary intake is one aspect of people’s lives over which they have a great deal of control. Thus, potential dietary interventions are of great interest to the public with respect to health maintenance and cancer prevention. The effects of dietary fatty acids have been frequently reviewed and are not the major focus of the current analysis. Not all research concerning fatty acids in cancer involves dietary intake, however, since variations in lipid metabolism within malignant cells can also affect cell signaling and tumor growth (reviewed by Currie et al. 2013).

Potential mechanisms for anticancer effects of n-3 FAs have been reviewed in detail (Gu et al. 2013). Until recently, mechanistic explanations for the effects of fatty acids on cancer cells have focused primarily on the metabolism of the FAs to other products that either enhance or inhibit cellular signaling. Foremost among these theories is the idea that n-3 FAs compete for metabolic pathways that convert n-6 FAs to pro-inflammatory products and that the n-3 products are either less inflammatory or anti-inflammatory. Since inflammation generally promotes cancer, this mechanism is logical, and much evidence has accumulated in its support. However, in the field of lipid signaling, it has historically been the case that non-receptor mechanisms have initially predominated, only to be overturned or reconsidered once a receptor for a lipid mediator is identified. The potential roles for G-protein-coupled receptors (GPCRs) (Audigier et al. 2013), and lipid GPCRS in particular (Van Jaarsveld et al. 2015), as druggable targets for cancer therapy have recently been reviewed.

The discovery of FFARs, and our evolving view of their cellular roles, is only beginning to provide perspective on the roles of these GPCRs in cancer cells. We have reached the point where it is important to distinguish receptor-mediated events from events mediated by fatty acid metabolism, in order to obtain a more complete picture of the actions of fatty acids on cancer cells. In this review, we will organize the discussion by receptor type, and then by cancer type.

2 FFAR Subtypes

In the last decade, four orphaned receptors were classified into a new GPCR family, the free fatty acid receptor (FFAR) family. The FFAR family is comprised of free fatty acid (FFA) receptors 1–4, formerly known as GPR40, 43, 41, and 120, respectively. Activation of these receptors appears to have many health benefits, ranging from stimulation of insulin secretion, improvement of adipose function and pancreatic cell viability, and modulation of inflammatory responses and energy intake (Butcher et al. 2014; Dranse et al. 2013; Gotoh et al. 2007). In the last few years, there has been an intensification of interest in this family of receptors, mostly due to their promise as targets for metabolic diseases such as diabetes and obesity. While most of the research has been focused on metabolic diseases, several groups have begun to examine the role of FFARs in other areas and particularly with respect to inhibition of inflammation. Inflammation has roles in multiple diseases, including cancer, cardiovascular disease, multiple sclerosis, and rheumatoid arthritis (Butcher et al. 2014; Talukdar et al. 2011; Wu et al. 2013). The following discussion focuses on the roles of FFARs in cancer, which in some cases overlap with their roles in inflammation.

FFAs 1, 2, and 3 are closely related GPCRs, sharing 30–40% of their amino acid sequence, which are encoded in tandem on chromosome 19q13.1. FFA2 and 3 are activated by short-chain fatty acids, such as formate, acetate, butyrate, propionate, and pentanoate. The EC50 values of the endogenous ligands range from 0.1 to 1.0 mM, with physiological serum levels ranging from 50 to 200 μM (Dranse et al. 2013; Feng et al. 2012). However, FFA2 and 3 exhibit obvious differences in the rank-order potency for these ligands. For instance, acetate and propionate have similar potencies at FFA2, while propionate has greater potency than acetate for FFA3. Another difference between the two receptors concerns G-protein coupling: FFA2 couples to both the Gi/o and Gq/11 families, whereas FFA3 couples solely to the Gi/o family (Burns et al. 2014; Dranse et al. 2013; Hara et al. 2011; Hirasawa et al. 2008; Holliday et al. 2012; Hudson et al. 2013a). FFA2 can also signal via arrestins (reviewed by Bolognini et al. 2016).

FFA1, while sharing genomic similarities to FFA2 and 3, is more similar in function to FFA4. FFA1 and 4 are both receptors for long-chain fatty acids, with physiological levels of the endogenous ligands ranging from 200 to 500 μM in serum and EC50 values ranging from 1 to 30 μM (Dranse et al. 2013). Long-chain fatty acids include saturated FAs such as palmitic acid, monounsaturated FAs such as oleic acid, and polyunsaturated omega-3 fatty acids (n-3 FAs). FFA1 and 4 also share G-protein coupling mechanisms; they both couple through Gq/11, although FFA1 also signals through Gi/o (Burns et al. 2014; Dranse et al. 2013; Hara et al. 2011; Hirasawa et al. 2008; Holliday et al. 2012).

The discussion that follows will be organized by individual FFARs, beginning with the long-chain FFARs because there is more information available about their roles in cancer cells than for the short-chain FFARs.

3 FFA1/GPR40

3.1 Overview of FFA1

FFA1 was the first FFAR to be identified as a long-chain fatty acid receptor and, as such, has been the most studied of the four receptors in the FFAR family. Aside from being the first de-orphanized, FFA1 has also been studied extensively due to its potential as a target for the treatment of metabolic diseases such as diabetes. FFA1 activation enhances glucose-stimulated insulin secretion, which makes it a therapeutic target in type 2 diabetes (Butcher et al. 2014; Feng et al. 2012; Hudson et al. 2013a; Qian et al. 2014; Watterson et al. 2014).

While FFA1 is widely expressed throughout the body, it is most highly expressed in the pancreas and gut, contributing to the interest in FFA1 as a target for diabetes and other metabolic diseases (Butcher et al. 2014; Dranse et al. 2013; Hirasawa et al. 2008; Hudson et al. 2013a, b; Talukdar et al. 2011; Watterson et al. 2014). Currently, of the FFARs, FFA1 has the widest range of synthetic ligands, including both agonists and antagonists (Milligan et al. 2015). Synthetic agonists of FFA1 include GW9508; AMG 837; MEDICA16; TUG424; Cpd B, C and 37; AM-1638; and TAK-875, while synthetic antagonists include GW1100, DC260126, and Pfizer compound 15i (Briscoe et al. 2006; Du et al. 2014; Feng et al. 2012; Hara et al. 2014; Holliday et al. 2012; Hudson et al. 2013b, 2014; Kebede et al. 2012; Sun et al. 2010; Talukdar et al. 2011; Watterson et al. 2014).

To date, the roles of FFA1 in cancer have been investigated in several different types of cancer cells. Breast cancer has been of particular interest, in part, because of the known association between obesity and breast cancer. Obesity leads to hyperlipidemia and increases in free fatty acid levels in plasma (Felber and Golay 2002). In addition, there is considerable epidemiological evidence that dietary n-3 FAs are of benefit for breast cancer prevention and treatment (Fabian et al. 2015), although the mechanism has not been established.

3.2 FFA1 in Breast Cancer

The expression of FFA1/GPR40 mRNA in human breast cancer cell lines was first demonstrated in 2004 by Yonezawa et al. (2004). They showed that oleate and linoleate increased intracellular calcium levels in these cells in a manner that was partially pertussis toxin sensitive, consistent with a role for a GPCR. However, knockdown experiments were not performed to specifically investigate the role of FFA1 in this early study.

Another report, published in 2005, reported that the n-3 FA eicosapentaenoic acid (EPA) inhibits proliferation of MCF-7 human breast cancer cells, using mouse tumor xenografts perfused in situ (Sauer et al. 2005). This inhibitory effect, which was measured by thymidine incorporation, was pertussis toxin sensitive. The authors reference possible roles for FFA1 and/or FFA4, but did not directly test them within their study.

In one study, Hardy et al. (2005) showed that 100 μM oleate stimulates proliferation of serum-starved MDA-MB-231 cells. The response was blocked by pertussis toxin, suggesting GPCR involvement. The authors further demonstrated that MDA-MD-231, T47D, and MCF-7 cells express mRNA for FFA1/GPR40. When FFA1 expression was knocked down by siRNA in MDA-MB-231 cells, the dose-response curve for oleate-induced proliferation was shifted to the right, with the most profound decreases in response seen with 1 and 5 μM oleate. The maximal response (50–100 μM oleate) was also significantly reduced, although only partially. This study was conducted before pharmacologic agonists for FFA1 were commercially available. In summary, this early study established that FFA1 is expressed in breast cancer cells, and suggested that activation of this receptor enhances proliferation.

Two additional studies, published in 2008, showed that both FFA1 and FFA4 are expressed in breast cancer cells. In one of these reports, expression of FFA1 and FFA4 was demonstrated in both MCF-7 and MCF10A cells by flow cytometry using anti-receptor antibodies (Soto-Guzman et al. 2008). Interestingly, 400 μM oleate increased DNA synthesis in MCF-7 and MDA-MB-231 breast cancer cells, and not in MCF10A cells which are non-tumorigenic (Navarro-Tito et al. 2008). The potential roles for the FFARs were not investigated by knockdown studies. In the subsequent report, from the same group, the expression of FFA1 and FFA4 was again demonstrated by flow cytometry. The major point of this second manuscript was to characterize responses (FAK activation, migration) to arachidonic acid, which were shown to be pertussis toxin sensitive. However, these responses were presumed to be independent of FFA1 and FFA4 since these GPCRs are not activated by arachidonic acid.

Data concerning FFA1 function in breast cancer cells were presented as part of a 2014 study that focused on melanoma (Nehra et al. 2014). The authors showed that FFA1 mRNA was expressed in MCF-7 cells, although the major focus of the manuscript was that the FFA1 mRNA levels were much higher in melanoma cells than in breast cancer cells. The FFA1-selective agonist TAK-875 decreased cell numbers for melanoma cells, but not MCF-7 cells, as measured after 72 h in medium containing 1% serum. The cell loss was referred to as both an “inhibitory effect on growth” as well as a “toxic effect”; the protocol used did not distinguish between inhibition of proliferation and loss of viability.

More recently, our group showed that EPA (20 μM) and pharmacologic FFAR agonists inhibit proliferation and migration of MCF-7 and MDA-MB-231 cells in response to lysophosphatidic acid (LPA) or epidermal growth factor (EGF) (Hopkins et al. 2016). Dose-response studies comparing the inhibitory effects of the FFA1-selective agonist GW9508, and the FFA4-selective agonist TUG-891, suggested that FFA1 may play a major role in inhibiting growth in these cell lines. In other words, the IC50 values for the inhibitory effects of the agonists were most consistent with a role for FFA1. Expression of FFA1 was demonstrated at the mRNA level, and expression of FFA4 was demonstrated at the mRNA and protein levels in this study. However, knockdown studies were not performed to further test the relative roles of FFA1 and FFA4. The inhibitory effects of FFAR agonists on proliferation in this study are consistent with previous literature concerning effects of n-3 fatty acids on breast cancer cells (Manna et al. 2008; Zou et al. 2013; Xue et al. 2014; Pogash et al. 2015). The study design also investigated whether the inhibitory action of FFAR agonists on EGF response might be mediated by LPA receptors, which are also GPCRs. The results were consistent with such a mechanism, but not definitive.

In summary, expression of both FFA1 and FFA4 has been demonstrated in breast cancer cell lines by several different groups. The effects of FFA1 agonists have been variable, with different dose ranges and FAs tested in different studies. It is not yet clear that growth-stimulating effects of oleate are mediated predominantly by FFA1. There is a consensus between two groups that EPA inhibits breast cancer cell proliferation, with our group implicating FFA1 in this response. Further studies are needed to delineate the relative roles, whether stimulatory or inhibitory, of FFA1 and FFA4 in breast cancer cells.

3.3 FFA1 in Other Cancers

Ishii et al. (2015a) recently published a study of the role of FFA1 in migration and invasion in HT1080 fibrosarcoma cells. This report is related to another study published by the same group earlier in the same year, concerning FFA1 and FFA4 in WB-F344 liver epithelial cells (Ishii et al. 2015b). The study of the liver epithelial cell line indicated that FFA4 activation promotes cell motility, while FFA1 inhibits motility (Ishii et al. 2015b). In this study, the investigators also indicated that the effects might involve modulation of LPA receptors. In their study of the fibrosarcoma cell line (Ishii et al. 2015a), the investigators detected expression of FFA1 but not FFA4. Treatment of these cells with the FFA1-selective agonist GW9508 decreased motility and invasion. This study is another example of an inhibitory effect mediated by FFA1 in cancer cells.

The same research group examined the roles of FFA1 and FFA4 in pancreatic cancer cells (Fukushima et al. 2015). In this study, the investigators used shRNA-mediated knockdown to modulate levels of both receptors, which were endogenously expressed. Cells lacking FFA4 showed decreased motility and reduced colony formation, as compared to control cells. In contrast, cells lacking FFA1 showed enhanced motility, increased MMP-2 activity, and enhanced colony formation. Taken together, the results published by this group of collaborators provide interesting insights into potentially opposing effects of FFA1 and FFA4 within a single cell line and confirm that FFAR agonists can exert inhibitory actions on cancer cells.

FFA1 has been recently shown to be expressed at elevated levels in A2058, A375, and SKMel3, all of which are human melanoma cell lines, when compared to a control fibroblast cell line and to several human neuroblastoma and breast cancer cell lines (Nehra et al. 2014). In their cell culture studies, this group demonstrated that GW9508, an agonist for FFA1 and FFA4, and TAK-875, a selective FFA1 agonist, both inhibited the proliferation of the three melanoma cell lines, while showing no such inhibition in the control fibroblast cell line. In addition, the researchers tested the effects of treating mice with subcutaneous xenografts of human melanoma with 100 mg/kg of TAK-875. These mice developed tumors with significantly lower weights and volumes when compared to the tumors in the placebo control mice; there was no significant difference in total body weight between the two groups of mice. The studies with the FFA1 agonists were conducted in both cell lines (treatments with DHA) and animal models (diets with high n-3 fatty acid contents); the results of these studies mirrored one another, leading the authors to conclude that DHA is acting through FFA1 to inhibit the growth of malignant melanoma. This is another key example of the ability of long-chain FFAR agonists to inhibit cancer growth.

Results presented in a meeting abstract indicate that FFA1 is expressed in ~80% of high-grade serous ovarian carcinomas (Munkarah et al. 2016). The researchers report that exposure of ID8, A2780, C200, OVCAR3, and SKOV3 human ovarian cancer cell lines to adipocytes, a source of FFAs, results in increased expression of FFA1. The investigators further state that the FFA1 antagonist, GW1000, inhibits proliferation of ovarian cancer cell lines, while the agonist CAY10587 has no effect on proliferation. The proliferation study results appear to contradict results from previous reports, showing that n-3 FAs inhibit proliferation of ovarian cancer cells (Sharma et al. 2005, 2009). Further work is needed to resolve these contradictory findings.

Our group has shown that both FFA1 and FFA4 are expressed in human prostate cancer cell lines, PC-3 and DU145 (Liu et al. 2015a). As will be discussed further below, the inhibitory role of FFA4 is more prominent than that of FFA1 in these particular cells.

4 FFA4/GPR120

4.1 Overview of FFA4

Like FFA1, FFA4 is a receptor for long-chain fatty acids, although it does not share homology with the other three receptors in its family, having only 10% identity with FFA1. It was included in the FFAR family after being identified by high-throughput screening. FFA4 has been described as a selective n-3 FA receptor (Dranse et al. 2013; Hirasawa et al. 2008). Similar to FFA1, FFA4 is expressed throughout the body, with high levels of expression in the GI tract, adipocytes, taste buds, immune cells, and lungs. The highest expression in humans is in the lungs (Zhang and Leung 2014). FFA4 has been reported to play a role in regulating inflammation, insulin sensitization, and obesity reduction, as well as regulating lipid and glucose metabolism by stimulating the release of hormones (Liu et al. 2015a). It is believed to regulate inflammation and metabolism through suppression of macrophage-induced tissue inflammation and β–arrestin-2-mediated augmentation of insulin sensitivity, respectively (Burns et al. 2014; Dranse et al. 2013; Gu et al. 2013; Hara et al. 2014; Hirasawa et al. 2008; Hudson et al. 2014; Oh et al. 2014; Watterson et al. 2014; Wu et al. 2013; Zhang and Leung 2014).

There are two FFA4 isoforms: FFA4 long (FFA4L) and FFA4 short (FFA4S). FFA4S is found in humans, as well as in other species, whereas FFA4L is believed to be present only in humans. FFA4S contains 361 residues, while FFA4L contains an additional 16 residues between positions 231 and 247 in intracellular loop 3. As stated above, FFA4 signals through Gq/11, which it does only through its short isoform. However, both isoforms signal through the arrestin pathway (Dranse et al. 2013; Holliday et al. 2012; Hudson et al. 2013b; Milligan et al. 2015; Oh et al. 2010; Oh and Walenta 2014; Watson et al. 2012; Zhang and Leung 2014). FFA4 is thought to signal through β-arrestin-2, as it is a Class A or rhodopsin-like GPCR. Class A GPCRs preferentially bind β-arrestin-2 (arrestin-3) over β-arrestin-1 (arrestin-2) (Luttrell and Gesty-Palmer 2010; Oh and Walenta 2014). In addition, the Milligan group has shown that FFA4 does recruit arrestin-3 in Chinese hamster ovary (CHO) cells when stimulated with ALA or a synthetic FFA4 agonist, TUG-891 (Butcher et al. 2014).

The growing interest in FFA4 has prompted the discovery of several synthetic agonists to FFA4. There are two selective synthetic agonists with high affinity thus far: TUG-891 (chemical name 4-[(4-Fluoro-4′-methyl[1,1′-biphenyl]-2-yl)methoxy]-benzenepropanoic acid), which was developed by the Ulven group (Butcher et al. 2014; Milligan et al. 2015; Shimpukade et al. 2012), and GPR120 cpdA (chemical name 3-[2-chloro-5-(trifluoromethoxy)phenyl]-3-azaspiro[5.5]undecane-9-acetic acid), developed by Olefsky and co-workers (Oh et al. 2014). Other synthetic agonists are available; however, they are far less selective and/or potent. GW9508, mentioned above as an agonist for FFA1, is also an agonist for FFA4 but has much higher potency for FFA1 than for FFA4. Additional early FFA4 synthetic agonists include NCG21, NCG46, and AH7614, all of which, like GW9508, have significant dual agonism for FFA1 (Hara et al. 2014; Hudson et al. 2014; Oh et al. 2014; Shimpukade et al. 2012; Sparks et al. 2014; Watterson et al. 2014; Milligan et al. 2014). Although TUG-891 is currently one of the best available synthetic FFA4 ligands, it is limited to human FFA4, as it has limited selectivity for the mouse ortholog of FFA4 over the mouse ortholog of FFA1 (Hudson et al. 2014; Zhang and Leung 2014). CpdA has been shown to recruit β-arrestin-2 with equal intensity in both human and mouse cells (Oh et al. 2014).

4.2 FFA4 in Prostate Cancer

Our group used two human prostate cancer cell lines, PC-3 and DU145, to test whether previously reported inhibitory effects of n-3 FAs on prostate cancer cells are mediated by FFARs (Liu et al. 2015a). Our initial observations were that treatment of DU145, a human prostate cancer cell line, with n-3 FAs (EPA or DHA) inhibited responses to growth factors (EGF and LPA). These responses included signal transduction events such as Erk and Akt activation, as well as proliferation and migration. The inhibitory effect occurred too rapidly to be easily attributed to alterations in lipid metabolism. Further experiments showed that these inhibitory responses on proliferation and migration were mimicked by the synthetic FFAR agonists TUG-891 and GW9508; TUG-891 exhibited much higher potency, suggesting a major role for FFA4. FFAR agonists had similar inhibitory effects in PC-3, another human prostate cancer cell line. The inhibitory effects on proliferation were sustained as long as the agonists were present, but were readily reversible. Since these data pointed toward an FFAR-mediated effect, we tested for FFA1 and FFA4 mRNA expression and found that both transcripts were present in both cell lines, albeit at different levels. When FFA4 knockdown was accomplished in the prostate cancer cell lines using siRNA, the inhibitory responses to EPA and TUG-891 were lost. In summary, this study showed for the first time that inhibitory effects of n-3 FAs on prostate cancer cells can be mediated by an FFAR.

4.3 FFA4 in Breast Cancer

The roles of FFA4 in breast cancer cells are only beginning to be addressed. FFA4 is expressed in MDA-MB-231 and MCF-7 breast cancer cells (Hopkins et al. 2016; Navarro-Tito et al. 2008; Soto-Guzman et al. 2008). Navarro-Tito et al. (2008) concluded that although FFA4 is present in MDA-MB-231 cells, it likely does not play a role in facilitating FAK activation in these cells.

Using an obese mouse model to study the effects of n-3 FAs on breast cancer, Chung et al. (2014) determined that, in a mouse model, expression of FFA4 in the host mouse (i.e., host immune system) does not mediate the inhibition of xenograft mammary tumor progression observed in response to n-3 FAs. These investigators showed direct inhibitory effects of n-3 FAs on breast tumor cells, but these effects were independent of FFA4 expression.

The authors’ research group recently reported the results of cell culture studies (MCF-7 and MDA-MB-231) indicating that the inhibitory effects of FFAR agonists on breast cancer cell proliferation and migration are most consistent with a role for FFA1, even though FFA4 was also expressed in these cell lines (Hopkins et al. 2016). In summary, the findings for breast cancer cells recapitulate the general situation reported for prostate cancer cells, but are more complex and raise the possibility that FFA1 can also mediate inhibitory effects on growth factor action.

Based on the studies published to date, FFA4 can be expressed in breast cancer cells, but its roles in these cells remain to be defined.

4.4 FFA4 in Other Cancers

FFA4 is highly expressed in colorectal cancer cell lines (Liu et al. 2015a; Navarro-Tito et al. 2008). In a very comprehensive study, Wu et al. (2013) showed that FFA4 is overexpressed in colorectal cancer tissue as compared to adjacent normal tissue. Using GW9508 as agonist, these investigators used two established colorectal cell lines, with and without FFA4 knockdown, to show that FFA4 mediates Akt activation and enhanced migration in these cells (Wu et al. 2013). Taken together, the results of this study indicate that (FFA4?) promotes tumor growth in colorectal cancer. Results from our group, using the Caco2 colorectal cell line, also indicate that FFAR agonists activate Akt in these cells and do not inhibit growth factor response (Liu et al. 2015a), in contrast to their inhibitory effects in prostate and breast cancer cells (Hopkins et al. 2016; Liu et al. 2015a).

The varying responses to FFA4 activation in different cancer cell types, tissues, and models demonstrate the need for further investigation into the role of FFA4 in cancer (Oh and Walenta 2014).

5 FFA2/GPR43

5.1 Overview of FFA2

FFA2 and FFA3 are activated by short-chain (less than six carbon) fatty acids (SCFAs) and are sometimes referred to as “butyrate receptors”. These receptors were “de-orphanized” in 2003 (Le Poul et al. 2003). FFA2 is expressed rather selectively within the body, with highest expression in the GI tract, adipocytes, monocytes, and neutrophils. The transcriptional regulation of FFA2 expression has therefore been investigated (Ang et al. 2015). Since butyrate is an anti-inflammatory molecule produced by fermentation of dietary fiber by colonic bacteria, and since dietary fiber reduces colon cancer, the roles of FFA2 and FFA3 are of particular interest with respect to colon cancer (reviewed by Bultman 2013). It is not only the direct effects of SCFAs on cancer cells that are of relevance. Both FFA2 and FFA3 can mediate pro-inflammatory responses to SCFAs in colon epithelial cells, which may play a role in protective immunity (Kim et al. 2013). In addition, FFA2/FFA3-independent effects of SCFAs on T-cells may regulate immune response (Park et al. 2015). Other physiological roles of FFA2 and FFA3 have been reviewed (Offermanns 2014).

Selective agonists and antagonists for FFA2 have been under development, as discussed in a review that included the patent literature (Ulven 2012). Chloro-α-(1-methylethyl)-N-2-thiazolylbenzeneacetamide (CMTB) is a selective agonist for FFA2 (Milligan et al. 2009). A more recent report detailed the synthesis and characterization of new FFAR agonists that are phenylthiazole-carboxamido acid derivatives (Ma et al. 2016).

5.2 FFA2 in Cancer

In an early study of the roles of FFARs in the response to short-chain fatty acids in breast cancer cells, one group showed that both FFA2 and FFA3 are expressed in MCF-7 cells (Yonezawa et al. 2006). These investigators found that the short-chain fatty acids (10 mM acetate, butyrate, or propionate) induce an acute increase in intracellular calcium in MCF-7 cells, and also caused activation of p38 MAPK. Knockdown of FFA2 resulted in a loss of both the calcium and p38 responses. All three SCFAs (10 mM) decreased cell numbers in the absence of serum; only butyrate inhibited proliferation in response to serum. Effects on proliferation were not a major focus of this study.

As mentioned earlier, short-chain fatty acids, which are products of bacterial fermentation in the gut, are known to play a protective and preventative role against colon cancer (reviewed by Bindels et al. 2013). This represents one aspect of the burgeoning interest in the roles of gut microbiota in health and disease (reviewed by Louis et al. 2014). Tang et al. (2010) therefore examined expression of FFA2 in human colorectal cancer and found it to be markedly reduced in colon adenocarcinomas as compared to normal colon tissue. These investigators proceeded to restore FFA2 expression to HCT8, a colon cancer cell line, and then showed that propionate and butyrate-induced apoptosis in the FFA2-expressing cells. The authors concluded that FFA2 functions as a tumor suppressor in the colon.

In a particularly interesting study, Bindels et al. (2012) tested the hypothesis that production of SCFAs by gut microbiota, in response to consumption of nondigestible carbohydrates, might inhibit tumor growth by activating FFARs. Mice were injected with Ba/F3 transfected with Bcr-Abl, a leukemic cell line that expresses FFA2 and invades the liver. The reasoning was that the liver is directly exposed to SCFAs that are taken up by the gut, and thus an effect of SCFAs might be detected in this xenograft model. Accordingly, they found that treatment of the mice with nondigestible carbohydrates decreased the infiltration of the tumor cells into the liver. The authors also showed that acetate, propionate, and butyrate inhibit Ba/F3 cell proliferation in culture; this effect was mimicked by CMTB, a synthetic FFA2 agonist. In summary, this study provides another example of an antiproliferative effect on cancer cells, mediated by an FFAR.

Another important story concerning FFA2, with contrasting conclusions, is described in work published by Hatanaka et al. (2009). This study concerned gallbladder cancer. Retroviral expression screening was used to identify FFA2 as a transcript with oncogenic potential. Overexpression of FFA2 mRNA and protein enhanced gallbladder carcinogenesis; FFA2 mutations were not correlated with oncogenesis. Moreover, overexpression of FFA2 in 3T3 fibroblasts increased acetate and butyrate-induced proliferation of these cells. Evidence for upregulation of FFA2 in digestive tract tumors was also presented. In summary, this study showcases FFA2 as a transforming receptor with respect to digestive cancers, rather than an inhibitory receptor.

As pointed out in a review by Bindels et al. (2013), published studies have provided contradictory results, albeit with different model systems. Further work is needed to resolve the roles of FFA2 in cancer cells.

6 FFA3/GPR41

6.1 Overview of FFA3

Before GPR41 was de-orphanized as an FFAR, and assigned the alternative nomenclature FFA3, it was identified as a “hypoxia-induced apoptosis receptor” (Kimura et al. 2001). This work was done in H9c2, a rat cell line derived from embryonic heart. Although the study did not concern cancer, it is worthy of mention because overexpression of FFA3 resulted in increased p53 and apoptosis. A deletion mutant of FFA3 that was unable to couple to G proteins caused the opposite effect, inhibiting hypoxia-induced apoptosis. The authors also noted that transcript for FFA3 increased within 2 h after ischemia followed by reoxygenation, indicating that FFA3 is inducible.

6.2 FFA3 in Cancer

As noted earlier, Yonezawa et al. (2006) showed that FFA3 was expressed in MCF-7 human breast cancer cells. However, since mRNA for FFA3 was expressed at much lower levels than that for FFA2, these investigators subsequently focused on the role of FFA2.

Wu et al. (2012) stably transfected CHO cells with FFA3 and then examined responses to butyrate. They reported that butyrate was antiproliferative and pro-apoptotic in untransfected cells, but that these effects of butyrate were inhibited in cells transfected with FFA3. The roles of histone acetylation and cell cycle progression were also examined. However, it is not clear whether the potential role of endogenously expressed FFA2 was considered in this study.

As can be seen from the limited studies of FFA3-mediated effects on proliferation in both tumor and non-tumor cells, the roles of FFA3 in cancer cells are under-studied relative to those of the other three FFARs.

7 Conclusions

At the time when this review was compiled, there were relatively few studies published examining the roles of FFARs in cancer cells. However, several conclusions emerge from the available data.

First, several studies demonstrate that FFAR agonists, either dietary or pharmacologic, exert inhibitory effects on cancer cell proliferation. The best-established examples for long-chain fatty acid receptors are the inhibitory role of FFA4 in prostate cancer cells (Liu et al. 2015a, b), of FFA1 and/or FFA4 in breast cancer cells (Hopkins et al. 2016), and FFA1-mediated inhibition in fibrosarcoma (Ishii et al. 2015a), melanoma (Nehra et al. 2014), and pancreatic cancer (Fukushima et al. 2015) cells. For short-chain fatty acid receptors, FFA2 activation is inhibitory in colon cancer cells (Tang et al. 2010) and in a leukemia cell line (Bindels et al. 2012).

Second, FFARs have been shown to enhance proliferation in some types of cancer cells. The most prominent examples are FFA4-mediated proliferation in colorectal tumor lines (Wu et al. 2013), and FFA2-mediated carcinogenesis in gallbladder (Hatanaka et al. 2009). Results from the published studies that have been discussed in this review are summarized in Fig. 1.

Fig. 1
figure 1

Scheme summarizing the published reports discussed in this review. The terms “inhibits” and “stimulates” refer to proliferation, migration, and/or motility of the cancer cells. The term “likely” refers to studies in which agonist effects were consistent with the indicated action at a specific receptor, but not yet proven

Full size image

The third comment concerns potential mechanisms of action. Many GPCRs, when activated, can enhance the growth of tumor cells (Dorsam and Gutkind 2007; O’Hayre et al. 2013). Thus, it is atypical to see GPCR agonists under consideration for cancer therapy; typically GPCR antagonists are used to block pro-mitogenic signaling. The mechanisms for GPCR-mediated proliferation that are usually invoked involve activation of canonical signaling pathways such as Erk, Akt, and EGFR; these pathways are typically activated by Gq/11-coupled GPCRs. It is strikingly unusual that FFARs such as FFA1 and FFA4 can inhibit proliferation, despite their known coupling to Gq/11. There are other published examples of FFA4-mediated inhibition of signaling; these examples concern inflammation (e.g., Li et al. 2013; Mobraten et al. 2013; Yan et al. 2013). The key to the inhibitory mechanism may lie in the very prominent role of arrestins in mediating responses to some FFARs, and particularly to FFA4 (Burns et al. 2014; Hudson et al. 2014; Prihandoko et al. 2016). Arrestins can mediate diverse responses to GPCR activation, and can interact with non-GPCR signaling proteins, but the full range of their signaling activities has not been determined (Kaparianos et al. 2013; Luttrell 2013). It is therefore possible that some or all of the inhibitory effects noted for FFARs are arrestin dependent. From a GPCR perspective, this would be a very intriguing mechanism that could involve receptor-receptor interactions.

Further studies are needed to better establish the roles of FFARs in diverse types of cancer cells, and to pinpoint their mechanisms of action. With additional knowledge, we will eventually be able to progress from dietary interventions to pharmacologic ones to fully explore the potential of FFAR ligands as preventative or therapeutic agents in cancer.