Targeting stearoyl-CoA desaturase in solid tumors

Abstract

Cancer cells are demarcated from normal cells by distinct biological hallmarks, including the reprogramming of metabolic processes. One of the key players involved in metabolic reprogramming is stearoyl-CoA desaturase (SCD), which converts saturated fatty acids to monounsaturated fatty acids in an oxygen-dependent reaction that is crucial for maintaining fatty acid homeostasis. As such, SCD has been identified as a potential therapeutic target in numerous types of cancers, and its inhibition suppresses cancer cell growth in vitro and in vivo. This review summarizes the evidence implicating SCD in cancer progression and proposes novel therapeutic strategies for targeting SCD in solid tumors.

Introduction

Over the last decade, the extensive reprogramming of cellular metabolism has become widely recognized as a hallmark of cancer (1,2). This reprogramming is necessary to support the energetic and biomolecular needs of proliferating cells, as well as other oncogenic processes such as metastatic invasion and chemoresistance. An abundance of research over the last century has expanded our understanding of cancer metabolism. In the early 20^th century, Warburg and colleagues discovered that cancer cells exhibit increased levels of glucose uptake and lactate secretion independent of oxygen levels (3). Since then, many studies have aimed to better understand cancer cell metabolism in different contexts and identify vulnerabilities for developing novel therapeutic strategies. In recent years, lipid metabolic reprogramming has been identified as an important component of the substantial modulation that takes place during tumorigenesis (4).

Lipids are hydrophobic biomolecules that fulfill diverse roles in all cells. These roles include serving as major structural components of cellular membranes, modifying proteins, acting as signaling molecules, and storing energy. Proper regulation of lipid synthesis, breakdown, and uptake is therefore essential for maintaining normal cellular physiology. Consequently, it is unsurprising that reprogramming lipid metabolism is an important facet of tumor biogenesis, survival, and proliferation. Numerous studies have established that cancer cells exhibit elevated rates of de novo fatty acid biosynthesis, which has been reviewed comprehensively (5). Once synthesized, these fatty acids may be stored as energy in lipid droplets in the form of triglycerides, assembled into phospholipids for membranes, or used for signaling molecules such as diacylglycerol and phosphatidylinositol-(3,4,5)-triphosphate (PIP₃). Several fatty acid biosynthetic enzymes, including fatty acid synthase (FASN) and stearoyl-CoA desaturase (SCD) have been implicated in oncogenic proliferation and survival. In this mini review, we examine the relevance of SCD in metabolic reprogramming in hypoxic solid tumors and the potential chemotherapeutic benefits of targeting unsaturated fatty acid biosynthesis.

The Role of SCD in Fatty Acid Synthesis

SCD function and isoforms

Stearoyl-CoA desaturase is a delta-9 desaturase located in the ER membrane, where it introduces a cis double bond at the C9 position of saturated fatty acids (SFAs) palmitic acid (16:0) or stearic acid (18:0) to form monounsaturated fatty acids (MUFAs) palmitoleic acid (16:1) or oleic acid (18:1), respectively (6,7). This reaction requires NAD(P)H-cytochrome b₅ reductase, cytochrome b₅, and molecular oxygen and is the rate-limiting step of MUFA synthesis (8). SCD activity influences triglyceride and phospholipid composition within the cell and serves to balance the ratio between saturated and unsaturated fatty acids (9). This ratio in turn serves as an important determinant of membrane fluidity, organelle function, and cell-cell interactions (10).

Five SCD isoforms have been identified in mice and humans. Each of the SCDs shares delta-9 desaturase activity but has distinct tissue distribution. The first four, Scd1-4, are found in mice, while SCD1 (notated as SCD) and SCD5 are found in humans (11,12). SCD shares high sequence homology with Scd1, whereas SCD5 is unique to primates and is not an ortholog of any of the mouse Scd genes (12). In mice and humans, SCD1 is the dominant isoform expressed ubiquitously, while the other isoforms are localized in specific tissues. Mouse Scd2 and Scd3 are found in the skin and Harderian and preputial glands, while Scd4 is expressed in the heart (11). SCD5 is predominantly expressed in the human brain and pancreas (12). In this mini-review, we focus on SCD as a target for cancer therapy.

Regulation of SCD expression

SCD is regulated primarily on the transcriptional level by several different hormones, nutrients, and transcription factors (13). Two hormones that have been found to regulate SCD are insulin and leptin. In human liver cells, insulin induced SCD gene expression, mediated by insulin response elements (IREs) in the SCD promoter (14). This upregulation supports liver anabolism after feeding when excess glucose is shunted to fatty acid synthesis pathways. Leptin, a hormone that negatively regulates appetite and fat storage, was found to have an opposite effect on Scd1 regulation in mouse liver and suppressed Scd1 expression (15). This downregulation, in turn, suppressed de novo fatty acid synthesis and instead upregulated fat storage pathways. In mice, Scd1 mRNA expression in the liver was induced upon feeding with a high carbohydrate, fat free diet but dampened upon a diet containing polyunsaturated fatty acids (PUFAs) (16). Cholesterol was found to increase expression of Scd1 in rats and also reversed the PUFA-mediated suppression in mice that were fed PUFA- and cholesterol-rich diets (17,18). Scd1 upregulation was thought to decrease toxic effects of free cholesterol and increase oleic acid for cholesterol esterification.

Several transcription factors regulate SCD, including SREBPs, LXR, and ChREBP. Sterol regulatory element-binding proteins (SREBPs) are transcription factors that respond to low lipid nutrient conditions and activate transcription of lipid metabolic genes by binding to a sterol regulatory element (SRE) within the target gene promoter (19). SCD has an SRE in its promoter, and its mRNA was induced in response to insulin by SREBP-1c (20). Liver X Receptor α (LXRα), a nuclear hormone receptor transcription factor and known SREBP activator, was shown to increase Scd1 expression in an SREBP-1c-dependent manner in mouse livers (21). Carbohydrate responsive element-binding protein (ChREBP), a transcription factor known to act in synergy with SREBP-1c to activate fatty acid synthesis genes, was also shown to induce Scd1 expression in mouse hepatocytes (22). A number of important growth factors, including PDGF and FGF, have also been shown to initiate SCD expression through PI3K and SREBP activation (23). Aside from regulation on the transcriptional level, SCD has also been shown to be regulated on the post-translational level through ubiquitination and subsequent degradation by the proteasome (24). It is unknown which E2 conjugating enzyme and E3 ligase(s) are involved in this process. This degradation also appears to be constitutive and independent of MUFA levels within the cell. As a result, SCD is naturally short-lived, with a half-life of about 3 hours (24).

SCD Protects Against Lipotoxicity and Programmed Cell Death

SCD inhibition triggers ER stress and apoptosis

It has long been understood that the addition of saturated free fatty acids, such as palmitate, to cells in 2D culture results in cell death, termed lipotoxicity (25,26). Accumulation of saturated fatty acids induces apoptosis that is triggered by endoplasmic reticulum (ER) stress mediated by the unfolded protein response (UPR) (27). Palmitate addition to cultured cells resulted in increased saturated fatty acid content in rough ER microsomes due to increased saturated fatty acid incorporation into phospholipid and triglycerides (9). This change in fatty acid composition resulted in substantial alterations in ER structure and integrity, indicating that these may result in ER stress and ensuing cell death (9). The canonical sensors of ER stress, protein kinase R-like endoplasmic reticulum kinase (PERK), activating transcription factor 6 (ATF6), and inositol-requiring enzyme-1 (IRE-1), act upstream of apoptosis activation (28). Palmitate increased markers of these ER stress pathways, which was reversed by the addition of oleate (27,29). This saturated fatty acid-induced apoptosis has been found to act through C/EBP-homologous protein (CHOP), a pro-apoptotic transcription factor (27). Lipotoxic cell death has been characterized by canonical apoptotic markers such as caspase-3 activation and cytochrome c release from the mitochondria into the cytosol (30). A pan-caspase apoptosis inhibitor was found to inhibit lipotoxic cell death in HeLa cells (31). PERK knockdown also inhibited palmitate-induced apoptotic cell death in human liver cells (27). Lastly, genetic and chemical inhibition of SCD induced UPR sensor proteins and initiated cell death that was ameliorated by the addition of oleate (31,32). Given that SCD is responsible for MUFA synthesis, it follows that inhibition of SCD results in an increase in saturated fatty acid content and resulting cell death. SCD, alongside uptake of fatty acids, regulates the ratio of saturated to unsaturated fatty acids within the cell. Therefore, SCD inhibition results in an imbalance between saturated and unsaturated fatty acids, inducing ER stress and apoptosis.

SCD activity promotes ferroptosis resistance

Ferroptosis, first described in 2012, is an iron-dependent, programmed cell death pathway that is distinct from canonical cell death pathways, like apoptosis and necrosis (33,34). This novel form of regulated cell death has recently been studied as an important regulatory process in several disease contexts, including cancer (35,36). The apoptotic pathway has been widely studied for the last few decades, but implementing anticancer drugs targeting apoptotic effectors has not been successful thus far. Moreover, resistance to apoptosis has been identified as a hallmark of cancer, so targeting ferroptosis may be an alternative therapeutic strategy (1,36). Interestingly, the ferroptosis inducer erastin, now known to be a system Xc- inhibitor, was found to selectively induce cell death in engineered tumor cell lines expressing oncogenic mutant RAS, but not wild-type RAS (37). Subsequent studies more clearly defined the ferroptosis signaling pathway.

Ferroptosis occurs when there is an imbalance in the levels of oxidative stress and antioxidant action (Fig. 1a). Two key signals, iron overaccumulation and lipid peroxidation, initiate membrane damage during ferroptosis (33). Ferroptosis can be induced by an extrinsic pathway or an intrinsic pathway (34,36,38). The extrinsic pathway is induced by activation of iron transporters, like the transferrin receptor, or inhibition of transporters such as system Xc-, a cystine/glutamine antiporter that feeds into glutathione synthesis. The intrinsic pathway is induced by blocking antioxidant enzymes. Glutathione peroxidase (GPX4) is one such enzyme that utilizes glutathione to detoxify lipid peroxides. Consequently, inhibiting cystine import through system Xc-results in restricted glutathione synthesis and blocking GPX4 activity disrupts the balance between oxidative stress and antioxidant action, triggering ferroptosis.

Fig. 1. Mechanism of ferroptosis.

(a) A simplified schematic of key players in the molecular mechanism of ferroptosis. Figure made at biorender.com. (b) An illustration of key modulators of lipid peroxide production and their respective chemical inhibitors.

Polyunsaturated fatty acids (PUFAs) are most prone to lipid peroxidation and are incorporated into membranes by ACSL4, an essential pro-ferroptotic enzyme (39). Exogenous MUFAs protect cells against ferroptotic cell death triggered by erastin by decreasing lipid peroxides generated from PUFAs in the plasma membrane (40). This prompted several groups to hypothesize that the MUFA-synthesizing activity of SCD protects cells against ferroptotic cell death. Treatment of pancreatic and ovarian cancer cells with an SCD inhibitor sensitized cells to treatment with erastin or RSL3, a GPX4 inhibitor, and cell viability was rescued by a ferroptosis inhibitor, Fer-1 (Fig. 1b) (41,42). Mice injected with pancreatic cancer cells and treated with both erastin and an SCD inhibitor showed reduced subcutaneous tumor volume in comparison to mice treated with either drug alone (41). In a separate study using multiple cancer cell lines, overactivation of the PI3K-AKT-mTOR signaling pathway was found to protect cells against ferroptotic cell death, upstream of SCD (43). In that study, multiple cell lines showed marked sensitivity to RSL3 treatment after inhibition of mTORC1, SREBF1, or SCD. Altogether, these results indicate that SCD activity, specifically MUFA production, protects cancer cells from ferroptotic cell death, prompting an opportunity for a novel combination therapeutic strategy, detailed below.

Hypoxia affects SCD activity in solid tumors

The rapid proliferation of cancer cells results in tumor growth that outstrips the existing vasculature within the tissue. For this reason, solid tumors often exhibit nutrient-deprived, hypoxic microenvironments (44). Cancer cells therefore face the challenge of continued growth under limiting oxygen and nutrient conditions. This nutrient deprivation presents a vulnerability that can be leveraged for new therapeutic strategies. Given that MUFA synthesis is oxygen-dependent, and that cancer cells may need to upregulate SCD in conditions where oxygen is scarce to maintain production of monounsaturated fatty acids, targeting this pathway may be an effective anticancer therapeutic strategy.

In addition to the regulation of SCD by transcription factors and other nutrients, oxygen levels have been shown to affect SCD expression, which has been found to vary in different tumor contexts. Hypoxia-inducible factor (HIF) is a key transcription factor that allows cells to respond to hypoxic stress. HIF activity is especially relevant in the context of cancer because of it allows cancer cells to undergo metabolic reprogramming in the hypoxic tumor microenvironment (45). Under hypoxia in clear cell renal cell carcinoma (ccRCC) cell lines, HIF-2α expression was increased and correlated with an increase in SCD expression (46,47). Interestingly, the hypoxic tumor core was found to have an abundant fraction of oleic acid in ccRCC patient tumors (47). A study in glioblastoma cells also concluded that SCD expression was elevated in hypoxic conditions and that this was dependent on SREBP function (48). Conversely, in multiple cancer cell lines, it was found that tumor-relevant hypoxia limited SCD enzyme activity and that hypoxic cells rely on uptake of MUFAs rather than de novo synthesis (49,50). It was also observed that in ccRCC cells, lipid droplets prevented toxic buildup of saturated fatty acids under hypoxic conditions by releasing stored oleate (51). Taken together, these data indicate that hypoxia is an important regulator of MUFA synthesis and that hypoxic cancer cells adopt different strategies to maintain proper membrane fluidity. Therefore, targeting SCD may be a viable treatment strategy for hypoxic solid tumors given the need for MUFAs and their limited supply in nutrient poor conditions.

SCD Activity Promotes Cancer Cell Proliferation

SCD modulates oncogenic proliferation

As noted above, fatty acid synthesis has been identified as an important aspect of the vast metabolic reprogramming that occurs in cancer progression (52). Recently, SCD has been identified as an important player in several malignant processes, including tumorigenesis, tumor survival, and metastatic colonization (Fig. 2). Upregulation of SCD has been observed in numerous cancer types in different studies. Mass spectrometry imaging and immunohistochemistry across six different cancer tissue types (breast, colorectal, esophageal, lung, gastric, and thyroid) showed increased levels of MUFAs and overexpression of SCD in human patient tumors compared to adjacent normal tissue (53). Several independent studies also observed increased expression of SCD in different cancerous tissues compared to adjacent normal tissue, including breast, prostate, lung, ovarian, endometrial, bladder, thyroid, colon, and ccRCC (54–59). This upregulation was also linked to a poorer prognosis in both bladder cancer and ccRCC after classification of patient data (60,61). Given that SCD is the rate-limiting enzyme of de novo MUFA synthesis, it is expected that this upregulation is beneficial to highly proliferative cancer cells, for example, to form cellular membranes for cell division, or to provide lipids for storage in lipid droplets, which can act to prevent lipotoxicity and serve as an energy source (5). A recent study on the effects of diet on tumor growth found that SCD activity is required for tumor adaptation to a lipid-limiting environment and that the antitumor effect of SCD inhibition was exacerbated by a caloric restriction diet that lowered tumor lipid availability and a ketogenic diet that increased tumor levels of SFA (62). Diet has been established as a risk factor for various types of cancer (63), and this study presents SCD activity as a link between the diet and cancer. As metabolic activity and diet are closely related, this also suggests that SCD can connect metabolic disorders and cancer (64). It is also possible that other metabolic genes can fill the same role.

Fig. 2. Summary of SCD’s multifaceted role in tumor proliferation.

A schematic highlighting the different roles SCD activity plays in oncogenic proliferation. Figure made at biorender.com.

In addition to the upregulation of SCD, several studies using cells cultured in vitro and in vivo mouse models have shown that cancer cell growth is suppressed upon inhibition of SCD. siRNA and shRNA knockdown of SCD showed reduced proliferation in vitro in ccRCC, endometrial cancer, and bladder cancer cells (56–58). In vitro studies in endometrial, bladder, thyroid, colon, glioblastoma, prostate, and lung cancers showed that upon addition of SCD small molecule inhibitors, cell proliferation was inhibited (57–59,65–68). Further, oleic acid was shown to rescue the growth defect induced upon SCD inhibition, indicating that this phenotype is a result of MUFA depletion (56,59,65,68). Additionally, in vivo mouse xenografts illustrated that tumor growth is impeded by inhibition of Scd1 by shRNA knockdown and chemical inhibition in endometrial cancer, colon cancer, glioblastoma, and prostate cancer (56,57,65,67). Altogether, these studies indicate that SCD is important for cancer progression and is a potential therapeutic target for treating cancers.

In the studies described above, the effect of SCD inhibition was examined in solid tumors. Interestingly, the opposite effect was observed in chronic myeloid leukemia (CML), a liquid tumor (69). SCD was observed to be downregulated in leukemia stem cells compared to normal hematopoietic stem cells in a CML mouse model and in CML patients. When SCD was deleted, CML development was accelerated, supporting the conclusion that SCD functions as a tumor suppressor in CML, contrary to the studies described above (69). These contrasting results suggest that SCD plays different roles in different tumor contexts.

Several canonical oncogenic signaling pathways (70) require SCD activity, including the epidermal growth factor receptor (EGFR) (67,71), Wnt/β-catenin (67,72–75), and Notch (73) pathways. EGFR is a receptor tyrosine kinase that phosphorylates downstream targets, including phosphoinositide kinase (PI3K), to promote cell proliferation and survival. PI3K phosphorylates PIP₂ to PIP₃, the latter of which recruits Akt to the membrane to facilitate its activation through phosphorylation. SCD inhibition resulted in a decrease in PIP₃ and a decrease in activated phosphorylated Akt in prostate cancer and glioma cells (67,72). Moreover, in lung cancer cells treated with EGFR inhibitor, SCD overexpression was sufficient to restore cell death induced by EGFR inhibitor treatment (71). The Wnt signaling pathway is activated by secreted Wnt proteins that stabilize the transcription factor β-catenin, which promotes tissue development and homeostasis. Upon SCD inhibition β-catenin and Wnt protein secretion were decreased in colon, prostate, and glioma cancer cells (67,72,73). SCD was shown to have an essential role in the production of MUFA-modified Wnt proteins, which attributes the defect in signaling capability to an absence of MUFAs (75). SCD inhibition also decreased levels of key Notch pathway signaling molecules, responsible for regulating developmental genes (73). These pathways, including Wnt and Notch, are also relevant to regulating cancer cell stemness, which has been shown to contribute to cancer stem cell (CSC) phenotypes. CSCs are thought to initiate metastases and drive tumor formation. Given the relationship between these pathways and SCD, it follows that SCD also plays a role in supporting cancer cell stemness. Evidence in multiple cancer types show that SCD activity contributes to the maintenance of CSC phenotypes in ovarian, glioblastoma, gastric, breast, and lung cancer cells (66,74,76–79). Overall, SCD activity supports multiple branches of oncogenic signaling by contributing to the synthesis of lipid signaling molecules and protein modifications, ultimately affecting the activation of several pathways.

Perspectives on SCD as an Anticancer Therapeutic

SCD inhibitors as potential drug treatments

Activation of lipid synthesis has long been recognized as a cancer cell growth requirement. A specific fatty acid synthase (FASN) inhibitor, TVB-2640, recently started testing in several clinical trials (NCT03179904, NCT03808558, NCT02980029). Prior to the development of TVB-2640, FASN inhibitors showed whole body toxicity and had limited success (80). Thus, understanding the potential consequences of body-wide inhibition of SCD is critical for determining whether SCD is a viable drug target. Mouse knockouts of SCD have been previously studied. Scd1^−/− mice exhibited an altered phenotype of the skin, including alopecia and changes in cutaneous lipids (81). These mice also had increased energy expenditure and showed resistance to obesity and insulin resistance (81). Should these effects also occur in humans, the potential antitumor benefit of SCD inhibition may outweigh these side effects observed in knockout mice. It should also be noted that mice have four isoforms of Scd, while humans only have two. This may mean that the knockout mice phenotype does not necessarily reflect the phenotype in humans, given that only Scd1 and SCD share sequence homology, and SCD5 does not share homology with any of the mouse Scds.

Several SCD inhibitors are commercially available and have been tested in vitro and in vivo, including CAY10566, A939572, and MF-438. Such inhibitors have been tested extensively in preclinical studies which have been detailed above. Mice treated with these inhibitors show minor side effects such as increased eye discharge and hair loss (82). However, little SCD inhibitor pharmacokinetic and pharmacodynamic data are publicly available from human trials. One SCD inhibitor, MK-8245, was tested in Phase I clinical trials for Type 2 diabetes, and no severe adverse events were reported (83). Aramchol, another SCD inhibitor, is currently being tested in Phase III clinical trials for nonalcoholic steatohepatitis (84). It is possible that these inhibitors may also be beneficial in treating patients with solid tumors. Recently, a tumor-specific SCD prodrug was shown to impede tumor growth in cancer cell lines expressing CYP4F11 or CYP4F12, which are required for drug activation (85). This kind of inhibitor would limit side effects and may be studied further in different cell lines for targeting cancers specifically expressing these enzymes. Given the relatively mild phenotypes of SCD knockout mice, in vivo treatment data, and the availability of multiple small molecule inhibitors, using SCD inhibitors as chemotherapeutics is promising for future clinical trials.

Combination therapy strategies

Combination therapy is a treatment approach designed to optimize the additive or synergistic effect of two or more drugs targeting interconnected pathways. This may be an attractive anticancer treatment strategy for several reasons (86). First, compounds may be administered at reduced doses to minimize any negative side effects. Additionally, the use of multiple drugs may avoid chemoresistance issues that arise. Moreover, synergistic effects can be surveyed for among existing FDA-approved drugs to reduce research time and costs associated with developing and testing multiple new drugs.

The existing research implicating the role of SCD activity in multiple cell death pathways raises the potential for utilizing SCD inhibitors in novel combination therapeutic strategies. For example, because SCD activity protects cells against ferroptosis, it is possible that combined treatment with an SCD inhibitor and a ferroptosis inducer will result in more effective cancer cell killing (Fig. 1b). Targeting ferroptosis thus far has shown several challenges (35), including the absence of potent inhibitors and off-target toxicity, that might be overcome by leveraging such a combination therapy strategy. By impeding two arms of ferroptotic-protective pathways, cancer cells may be rendered more susceptible to drug treatment. In addition, SCD inhibition results in ER stress and apoptosis. Therefore, by similar logic, targeting the UPR while simultaneously inhibiting SCD may also result in more effective cell killing. The UPR serves as an important pathway for cancer cells to survive in stress conditions, which may confer more sensitivity to malignant cells in comparison to normal cells (87). Currently, ferroptosis inducers and UPR inhibitors have not progressed past the preclinical stages. This is largely attributed to the toxicity observed upon treatment with such drugs (82). Therefore, examining a synergistic effect with an SCD inhibitor could prove useful and provide support for administering both drugs at lower doses. A more recent study on dietary effects on tumor growth, described above, raises the additional possibility of a novel combination therapy by mimicking the effects of a calorie-restricted or ketogenic diet to sensitize tumor cells to SCD inhibition (62). This suggests that an SCD inhibitor may be coupled to a specific diet to starve tumors. These combination strategies may be an effective way to limit whole body toxicity issues that are responsible for limited success in targeting fatty acid synthesis in cancer.

Conclusion

SCD activity is critical for preserving an appropriate ratio of saturated to unsaturated fatty acids and has been found to be regulated by several distinct mechanisms. Abrogation of SCD activity results in lipotoxicity and cell death. Accumulating evidence presented in this mini review strongly supports the pursuit of SCD as a chemotherapeutic target. This idea is backed by many studies demonstrating that SCD is overexpressed in tumors and its inhibition results in reduced cell proliferation in both in vitro and in vivo models. SCD activity supports oncogenic proliferation that contributes to lipid metabolic reprogramming within the tumor and promotes growth by providing MUFAs in oxygen-limiting, nutrient-poor solid tumor microenvironments and protecting toxic buildup of saturated fatty acids. Moreover, SCD activity protects against apoptotic and ferroptotic cell death. This evidence provides a strong rationale for testing SCD inhibitors in anticancer treatment and reveals opportunities for novel combination therapy strategies.