Tumor lipid metabolism: a mechanistic link between diet and cancer progression

Abstract

The potential for “anti-cancer” diets to markedly alter cancer risk and prognosis has captured the imagination of patients, physicians, and researchers alike, but many of these dietary recommendations come from correlative studies that attribute certain diets to altered cancer risk. While provocative, little is known about the molecular mechanisms behind how these dietary interventions might impact cancer progression. Within this context, however, changes in tumor lipid metabolism are emerging as a key contributor. In this review, we examine the current understanding of lipid metabolism in the tumor microenvironment, suggesting how diet-induced changes in lipid composition may regulate tumor progression and therapeutic efficacy. By dissecting various cellular pathways involved in lipid metabolism, we highlight how diet modulates the balance between saturated and unsaturated fatty acid species in tumors to impact cancer cell and stromal cell function. Finally, we describe how current cancer therapies may synergize with diet to improve therapeutic efficacy.

Graphical Abstract

Introduction

The possibility of using “anti-cancer” dietary interventions to prevent cancer onset or improve cancer prognosis is enthusiastically embraced by patients, clinicians, and researchers alike. Unfortunately, while dietary choices are known to contribute to cancer risk [1], the impacts of implementing dietary modifications after diagnosis to improve survival rates remains understudied. As many cancer patients are only motivated to change their diet after receiving a diagnosis, differentiating between prevention and prognosis is key. In addition, because any dietary advice must work in concert with standard-of-care cancer treatment, understanding how diet interacts with cancer therapies is also crucial [2,3]. Importantly, the heterogeneity of cancer (e.g., multiple cancer cell types, genetic alterations, and therapeutic options) makes individualized precision nutrition necessary to achieve optimal patient prognosis. Implementing this vision requires translational studies that test how diet-induced molecular changes synergize with different therapies to target specific tumor types. It is therefore critical to define the molecular mechanisms underlying the effects of diet on cancer progression and therapy responses.

Since environmental metabolite levels constrain cellular metabolic activity [4], recent studies have focused on how dietary modifications change metabolite levels within the tumor microenvironment (TME) to impact tumor metabolism, growth, and responses to therapy [2,5]. In particular, diet-induced changes to lipid metabolism have emerged as an important modulator of cancer progression [6,7]. Lipids contribute to key biological processes, including cellular membrane synthesis, energy generation, and signal transduction [8–10]. Fatty acids (FAs) – a major building block of complex lipids – are characterized by the number of double bonds they contain, including: saturated FAs (SFA, 0 double bonds), monounsaturated FAs (MUFA, 1 double bond), or polyunsaturated FAs (PUFA, 2+ double bonds). The degree to which cellular FAs are saturated is tightly regulated to maintain membrane fluidity and cellular function, with cancer cells altering FA desaturation pathways to improve their growth and survival [11–13]. In this review, we will discuss how various diets can impact tumor progression by controlling the amounts and types of FAs delivered to the TME, which directly alters lipid metabolism pathways within tumors. Specifically, we will focus on emerging evidence that diet can impact cancer progression and therapy by disrupting the balance between SFAs and MUFAs/PUFAs within tumors.

The degree of cellular fatty acid saturation is tightly regulated

Many cancer cells upregulate the synthesis of FAs during rapid cell division to support the production of complex lipids that make up cellular membranes (for a more in-depth review, see ref [8–10]). Mammalian cells use fatty acid synthase (FASN) for de novo synthesis of the SFA palmitate (16:0) from glucose and glutamine-derived acetyl-CoA and malonyl-CoA (Figure 1). 16:0 undergoes further elongation by FA elongases (ELOVL1–7) or desaturation by stearoyl-CoA desaturase (SCD) to produce palmitoleate (16:1(n-7)). Analogously, SCD converts the SFA stearate (18:0) into oleate (18:1(n-9)). These MUFAs can then be further elongated to generate longer-chain MUFAs. In contrast, omega-6 (n-6) and omega-3 (n-3) PUFAs cannot be de novo synthesized by mammalian cells and must be taken up from the environment. The most upstream PUFAs are linoleic acid (18:2(n-6)) and alpha-linolenic acid (18:3(n-3)), which, after being imported, are elongated by ELOVL2/5 and further desaturated by fatty acid desaturase 1/2 (FADS1/2) to produce long-chain, highly unsaturated PUFAs (Figure 1).

Figure 1. Schematic of cellular fatty acid metabolism pathways.

De novo fatty acid synthesis (top) uses carbons from acetyl-CoA, which is derived from nutrients such as glucose and glutamine, to produce saturated fatty acids (SFA) and monounsaturated fatty acids (MUFA). Essential fatty acid metabolism (bottom) elongates and desaturates diet-derived polyunsaturated fatty acids (PUFA). Fatty acids are designated X:Y(n-Z), where X indicates the number of carbons, Y indicates the number of double bonds, and n-Z (or omega-Z) indicates the position of the double bond closest to the methyl end of the fatty acid. ACC: acetyl-CoA carboxylase, FASN: fatty acid synthase, SCD: stearoyl-CoA desaturase, ELOVL1–7: elongation of very long chain fatty acids protein 1–7, FADS1/2: fatty acid desaturase 1/2, β-ox: β-oxidation.

These pathways are regulated, in part, by extracellular lipid availability. The presence of unsaturated FAs inhibits both the transcription and proteolytic activation of sterol regulatory element-binding proteins (SREBPs), which themselves control the expression of de novo FA synthesis and PUFA metabolism genes [14]. Conversely, environmental lipid limitation stimulates de novo FA synthesis [15,16]. Importantly, tight regulation of these pathways controls cellular FA saturation to maintain membrane fluidity and cellular function and to promote cell proliferation and survival [11]. For example, excess 16:0 is toxic to cells because increased cellular FA saturation disrupts the endoplasmic reticulum (ER) membrane, which causes ER stress and activates the unfolded protein response (UPR). How UPR signaling resolves lipotoxic stress remains poorly understood, but prolonged UPR activation induces apoptosis [11,17]. SCD is critical for disposing of 16:0 by desaturating it into 16:1(n-7), particularly when FASN is stimulated under lipid starvation conditions. Inhibiting SCD in these conditions leads to 16:0 accumulation, inducing lipotoxicity and cell death [17,18^••]. Moreover, because SCD is an oxygen-dependent enzyme, hypoxic stress also inhibits SCD and promotes increased FA saturation. Some cancer cells can overcome SCD inhibition by employing FADS2 to convert 16:0 into the 16:1(n-10) MUFA sapienate [12], while others, particularly clear cell renal cell carcinoma cells, counter this stressor by releasing MUFAs stored in lipid droplets into phospholipid pools [19]. Similarly, exogenous unsaturated FAs can protect cells against 16:0 toxicity by channeling 16:0 into triglyceride pools [20] or hindering de novo 16:0 synthesis [18^••]. However, excess levels of some PUFAs in the environment, such as arachidonic acid (20:4(n-6)), can also be toxic to cells [21,22], in part through oxidation to lipid peroxides that trigger a form of cell death known as ferroptosis [23]. These examples illustrate how cells, in conjunction with the specific FAs that are available, must carefully control the activity of intracellular FA metabolism pathways to maintain the required balance of saturated versus unsaturated FAs.

The dietary fat type affects lipid composition in the TME

The types of lipids that are available to tumors depends on the type of fats contained in the diet, in part because different fats have distinct amounts of SFAs, MUFAs, and PUFAs. The American Cancer Society (ACS) recommends higher consumption of MUFAs and PUFAs, relative to SFAs [24]. Similarly, we recently showed that among pancreatic ductal adenocarcinoma (PDAC) patients in the Nurses’ Health Study and Health Professionals Follow-up Study, a dietary pattern lower in carbohydrates and higher in plant-based, but not animal-based, fat was associated with longer survival [18^••]. Because plant-based fats contain more unsaturated FAs [25], these observations suggest that the quantity and composition of fats in low carbohydrate diets can influence their effects on PDAC progression.

Despite these associations with dietary choices, the mechanisms underlying how dietary fat composition alters the metabolism of saturated versus unsaturated FAs within tumors remain poorly defined. By conducting studies in mouse cancer models, where both dietary fat intake and composition can be manipulated (e.g. ~10% in a low fat, high carbohydrate diet to ~90% in a ketogenic diet (KD)) (Figure 2), we can gain important insights into these effects. In Table 1, we highlight the FA compositions of common dietary fats, including SFA-enriched palm oil and cocoa butter, MUFA-enriched lard and olive oil, and PUFA-enriched fish and soybean oils. Interestingly, a recent study showed that a ~60% high fat diet (HFD), formulated with distinct types of fats, differentially impacted weight gain, blood lipid levels, and the gut microbiome composition in mice [26^•].

Figure 2. Dietary fat compositions can be altered to formulate different types of diets.

Macronutrient compositions of diets with different percentages of dietary fat, including caloric restriction (CR), a low fat diet (10% fat), a western diet (45% fat), a high fat diet (60% fat), and a ketogenic diet (KD, 90% fat).

Table 1.

Approximate fatty acid compositions of common dietary fats.

Fatty Acid	Milk Fat	Palm Oil	Cocoa Butter	Lard	Olive Oil	Fish Oil	Soybean Oil	Flaxseed Oil
4:0	2	-	-	-	-	-	-	-
6:0	1.5	-	-	-	-	-	-	-
8:0	1	-	-	-	-	-	-	-
10:0	2.5	-	-	-	-	-	-	-
12:0	3	-	-	-	-	-	-	-
14:0	11	1	<0.5	1	-	9	-	-
16:0	29	45	26	23	12	16	11	5
16:1(n-7)	1.5	<0.5	<0.5	2	1	12	-	-
18:0	13	4.5	35	13	3	3	4	4
18:1(n-9)	24	39	34	39	72	10.5	23	19.5
18:2(n-6)	3	9.5	3	18	9	2	53	15.5
18:3(n-3)	0.5	<0.5	<0.5	1	1	1.5	8	56
18:4(n-3)	-	-	-	-	-	3	-	-
20:4(n-6)	<0.5	-	-	<0.5	-	1.5	-	-
20:5(n-3)	-	-	-	-	-	13.5	-	-
22:5(n-3)	-	-	-	-	-	1	-	-
22:6(n-3)	-	-	-	-	-	10.5	-	-
SFA	63	50.5	62	37	15	28	15	9
MUFA	29	39	34	41	73	22.5	23	19.5
PUFA	4	9.5	3	19	10	33	61	71.5

Values expressed as % of total fatty acids. Data adapted from Envigo Teklad.

Using mass spectrometry-based metabolomics and lipidomics methods, the tumor interstitial fluid (TIF), which closely mirrors the nutrient milieu experienced by tumor cells, can be analyzed to provide a direct measurement of how different diets alter metabolite and lipid levels in the TME [27]. For example, a 60% lard-based HFD increases diacylglyceride (DAG) and triacylglyceride (TAG) levels but decreases free FA levels in the TIF [28^••]. Similarly, we demonstrated that a caloric restriction (CR) diet decreases, while a KD increases, FA availability in the TME [18^••]. Critically, changing the fat type alters the plasma FA composition in a manner that reflects the SFA, MUFA, and PUFA composition of the diet. For example, palm oil is enriched with both 16:0 and 18:1(n-9) (Table 1), and a KD with palm oil raises 16:0 and overall SFA levels in the blood and in tumors. Likewise, the decreased levels of circulating 16:0 and 18:1(n-9) induced by CR can be reversed by feeding a high fat CR diet formulated with palm oil; a similar CR diet with soybean oil, which is enriched in 18:2(n-6) (Table 1), does not increase these levels [18^••].

Despite the importance of the fat type in the diet, many dietary studies using mouse cancer models do not take this factor into account. Nevertheless, it has become clear that diet-induced changes to lipid availability in the TME interact with tumor cell lipid metabolism to influence cancer progression. In the following sections, we highlight several recent studies that illustrate how dietary modifications alter the balance of saturated versus unsaturated FAs in tumors to impact tumor biology, arguing for better control of dietary SFA, MUFA, and PUFA compositions in future studies that explore the links between diet and cancer.

Dietary changes influence tumor progression by altering cancer cell lipid metabolism

Because cancer cells exhibit high glucose consumption (i.e., the Warburg effect) [29], low glycemic diets such as CR or the KD are commonly thought to inhibit tumor growth by limiting glucose availability [30]. However, we recently found that in subcutaneous mouse allograft models of PDAC and lung cancer, both CR and the KD lower blood glucose, but only CR inhibits tumor growth. Therefore, CR may act by altering the availability of other nutrients beyond glucose [18^••]. As discussed above, CR, but not the KD, lowers FA levels in the TME, creating a lipid-depleted environment under CR in which SCD is required for cancer cell proliferation. However, CR also suppresses tumor SCD activity. Thus, CR tumors lack access to environmental MUFAs as well as the ability to produce them via SCD, leading to reduced MUFA/SFA ratios that impair tumor growth. Either overexpressing SCD in these cancer cells or feeding a high fat CR diet that raises circulating 18:1(n-9) levels reverses these effects. Interestingly, the KD also impairs tumor SCD, but does not impair tumor growth because the high fat composition of the KD increases MUFA availability in the TME, thus maintaining the tumor MUFA/SFA ratios. This can be reversed by feeding a KD high in 16:0-enriched palm oil, thus increasing tumor SFA levels, which in conjunction with decreased tumor SCD activity, slows tumor growth. Therefore, it is the diet-induced changes to tumor MUFA/SFA ratios that determine whether a low glycemic diet will impair tumor growth (Figure 3A) [18^••].

Figure 3. Overview of dietary effects on cancer cell and stromal cell lipid metabolism.

(A) Dietary modifications alter lipid availability in the tumor microenvironment and drive imbalances in ratios of SFAs, MUFAs, and PUFAs, which can affect tumor growth. Caloric restriction (CR) decreases TME lipid availability and impairs tumor SCD activity, resulting in a disrupted MUFA/SFA ratio that inhibits tumor growth. Similarly, a PUFA-rich diet increases PUFA levels in tumors to induce ferroptosis. (B) Fatty acid availability contributes to cancer metastases. Melanoma cells metastasize more efficiently when circulating in the lymph versus the blood, in part because the lymph contains more oleate, which protects against oxidative stress and ferroptosis. Increased palmitate availability in pre-metastatic niches can also promote metastasis formation by promoting nuclear factor-kappaB (NF-kB) signaling. (C) The distribution of saturated or unsaturated fatty acids affects the function of tumor-associated immune cells. For example, to activate naïve T cells, phosphoinositides (PIs) enriched in PUFAs are required, whereas saturated PIs allows differentiated T cells to sustain proliferation and function. Fatty acid transport protein 2 (FATP2) regulates the immune suppressive function of polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs) via accumulation of arachidonic acid and synthesis of prostaglandin E2 (PGE2). (D) Dietary effects on lipid metabolism can mediate crosstalk between cancer cells and non-cancer stromal cells within a tumor. For example, in the context of high fat diet-induced obesity, cancer cells outcompete CD8⁺ T cells at uptaking fatty acids from the TME, leading to impaired T cell function and accelerated tumor growth.

Additional studies have highlighted how specific FAs can promote tumor progression. For example, breast cancer cells generate a pre-metastatic niche in the lung characterized by increased 16:0 levels, which is then oxidized to acetyl-CoA to promote p65 acetylation and pro-metastatic nuclear factor-kappaB (NF-kB) signaling. Therefore, a HFD that is high in 16:0 further increases 16:0 levels at metastatic sites, promoting the growth of breast cancer metastases (Figure 3B) [31^••]. Recent studies also showed that circulating melanoma cells in the blood experience oxidative stress, leading to lipid peroxides that induce ferroptosis and limit their metastatic potential [32^•,33]. To avoid this fate, more melanoma cells circulate in the lymph, which contains higher levels of 18:1(n-9) that protect against ferroptosis [34]. As such, melanoma cells in the lymph survive longer while in circulation, allowing them to form distant metastases (Figure 3B) [32^•]. While diet was not explored in this study, determining whether dietary fats that raise 18:1(n-9) levels in the blood promote metastases in certain cancer types will be important going forward.

While 18:1(n-9) generally protects against ferroptosis, increased uptake of PUFAs promotes ferroptosis because oxidized PUFA-containing phospholipids are responsible for triggering this form of cell death. Indeed, compared to an 18:1(n-9)-rich olive oil diet, a 22:6(n-3) fish oil diet increases omega-3 FA levels in blood and tumors, thereby inducing ferroptosis and impairing tumor growth (Figure 3A). Moreover, the 22:6(n-3)-rich diet enhances the effects of ferroptosis inducers, such as sulfasalazine [35^••]. This suggests that using dietary modifications that increase PUFA levels and decrease MUFA availability in the TME may be an effective strategy for promoting ferroptosis for cancer therapy.

Stromal cells and cancer cells compete for lipid species within the TME

Interactions between cancer cell and non-cancer stromal cell lipid metabolism also influence tumor progression. For example, FAs released from subcutaneous adipocytes can promote the proliferation and invasion of adjacent melanoma cells [36]. Dietary factors can also influence these cancer cell-stromal cell interactions. A recent study demonstrated that HFD-induced obesity accelerates the growth of some mouse tumor models by reducing the number and function of CD8⁺ T cells within tumors. The HFD increased FA uptake and oxidation in cancer cells, in part by lowering the expression of prolyl hydroxlase 3 (PHD3), a repressor of FA oxidation. As a result, decreased amounts of free FA were available in the TME for use by CD8⁺ T cells. By overexpressing PHD3 in these cancer cells, FA levels in the TME were restored, increasing T cell infiltration and blunting the growth promoting effects of the HFD. These data suggest that in the context of HFD-induced obesity, cancer cells outcompete T cells for FAs in the TME, which leads to impaired T cell function and accelerated tumor growth (Figure 3D) [[28^••].

How dietary intake of saturated versus unsaturated FAs influences these interactions remains poorly explored; however, emerging evidence suggests that FA saturation regulates the function of tumor-associated immune cells. For example, naïve T cells are enriched in PUFA-containing phosphoinositides (PI) that are necessary for initiating T cell activation. Activated T cells then stimulate de novo synthesis of SFA-containing PI, which are required for signaling events that sustain T cell differentiation, proliferation, and cytokine production (Figure 3C) [37^•]. In another case, polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs), pathologically activated neutrophils that impair anti-tumor immunity, were found to elevate expression of fatty acid transport protein 2 (FATP2). FATP2 facilitates the uptake of 20:4(n-6) for conversion to the proinflammatory cytokine prostaglandin E2, thereby supporting the immunosuppressive activity of PMN-MDSCs. Pharmacological inhibition of FATP2 blocks 20:4(n-6) uptake and synergizes with immunotherapy to impair tumor growth (Figure 3C) [38^•]. Together, these results suggest that it will be important to test whether diet-mediated changes to SFA versus PUFA availability in the TME influence anti-tumor immunity, revealing whether dietary interventions could enhance immunotherapy efficacy.

Clinical Implications and Conclusions

Even though some diets can slow tumor growth on their own, it is unlikely that any single diet will be sufficient to induce tumor regression. Moreover, dietary recommendations must be made within the context of each patient’s specific tumor and the therapies they are receiving. Therefore, it is important to identify how dietary interventions can be leveraged to enhance cancer therapies. For example, CR and fasting can improve tumor sensitivity to various chemotherapies and immunotherapies [30,39,40], and the KD can improve the efficacy of PI3K inhibitors in some contexts [41]. Recent studies have also shown that a KD enhances the efficacy of cytotoxic chemotherapies, such as nab-paclitaxel, gemcitabine, and cisplatin, on PDAC tumors in mice. These effects were associated with enhanced ketone body utilization, increased redox stress, inflammatory gene expression, and altered FA composition within tumor cells [42^•,43^•]. Conversely, other diets, such as a HFD, decrease the efficacies of chemotherapies in various mouse models [44,45].

How these diets influence responsiveness to drug therapies, especially through effects on lipid metabolism, remains undefined. However, some studies suggest that cancer cell FA composition can indeed influence chemotherapy sensitivity. For example, high FASN expression in PDAC cells is associated with poor responsiveness to gemcitabine, and pharmacological inhibition of FASN synergizes with gemcitabine to impair PDAC tumor growth [46]. Interestingly, tumors with high FASN activity exhibit increased FA saturation, while inhibiting FASN shifts cells towards increased polyunsaturation. In the latter case, the PUFA-rich cellular membranes exhibit altered membrane dynamics that promote the uptake of some chemotherapeutics, thereby increasing cellular sensitivity to these drugs [47]. A recent study demonstrated that in fasting mice and humans, erythrocyte membranes exhibited decreased SFA and MUFA, but increased PUFA, compositions, and the degree of these changes were correlated with increased sensitivity of normal tissues to chemotherapy toxicity [48^••]. Finally, altering the fatty acid composition of a KD was recently shown to influence the response of a human head and neck cancer xenograft to radiation therapy in mice [49^••]. These initial studies emphasize the important role that FAs play in tumor responses to chemotherapies, but much still remains unknown.

Finally, it is important to note that differences in lipid metabolism between sexes as well as among different species may have implications for future translational studies. For example, triglyceride and free fatty acid levels are reported to be higher in human males compared to females under basal conditions [50,51]. Further, while the enzymes that participate in fatty acid metabolism are largely similar between humans and mice, cholesterol is generally transported by LDL in humans, whereas rodents use HDL for cholesterol transport [52]. These observations suggest that further study is warranted to determine how these and other biological variables impact lipid metabolism before clinically relevant dietary interventions will be possible.

In summary, cellular FA saturation is an important regulator of tumor cell biology that influences disease progression and therapy responses. As we define how diets containing distinct fat types modulate TME lipid availability, tumor lipid metabolism, and the balance between saturated and unsaturated FAs in tumors, we will be able to identify strategies for using dietary interventions to enhance the efficacies of cancer therapies. These studies will also impact the development of lipid metabolism inhibitors, including those for FASN and SCD, as a treatment for various cancers [10]. Elucidating the molecular mechanisms underlying potential synergistic dietdrug combinations will provide scientific evidence that can benefit patients and physicians, providing guidance on how best to incorporate diet and nutrition into cancer therapy.

Highlights.

• Dietary changes can perturb cancer progression by altering tumor lipid metabolism

• Changes to dietary fat type can impact the lipid composition in the tumor microenvironment

• Stromal cells and cancer cells compete for lipid species within the tumor microenvironment

• Diet-induced dysregulation of saturated versus unsaturated fatty acid metabolism influences tumor growth and therapy responses

• Dietary interventions may improve therapeutic efficacy in cancer patients