The role of sphingosine-1-phosphate in the tumor microenvironment and its clinical implications

Abstract

Elucidating the interaction between cancer and non-cancer cells, such as blood vessels, immune cells, and other stromal cells, in the tumor microenvironment is imperative in understanding the mechanisms underlying cancer progression and metastasis, which is expected to lead to the development of new therapeutics. Sphingosine-1-phosphate is a bioactive lipid mediator that promotes cell survival, proliferation, migration, angiogenesis/lymphangiogenesis, and immune responsiveness, which are all factors involved in cancer progression. Sphingosine-1-phosphate is generated inside cancer cells by sphingosine kinases and then exported into the tumor microenvironment. Although sphingosine-1-phosphate is anticipated to play an important role in the tumor microenvironment and cancer progression, determining sphingosine-1-phosphate levels in the tumor microenvironment has been difficult due to a lack of established methods. We have recently developed a method to measure sphingosine-1-phosphate levels in the interstitial fluid that bathes cancer cells in the tumor microenvironment, and reported that high levels of sphingosine-1-phosphate exist in the tumor interstitial fluid. Importantly, sphingosine-1-phosphate can be secreted from cancer cells and non-cancer components such as immune cells and vascular/lymphatic endothelial cells in the tumor microenvironment. Furthermore, sphingosine-1-phosphate affects both cancer and non-cancer cells in the tumor microenvironment promoting cancer progression. Here, we review the roles of sphingosine-1-phosphate in the interaction between cancer and non-cancer cells in tumor microenvironment, and discuss future possibilities for targeted therapies against sphingosine-1-phosphate signaling for cancer patients.

Introduction

“The seeds of a plant are carried in all directions; but they can only live and grow if they fall on congenial soil.” First invoked by Paget ¹ in 1889, the seed and soil hypothesis has suggested that the vivid growth of cancer depends on the properties of cancer cells (seeds) and its interactions in their target organs with a congenial microenvironment (soil). ² Over 100 years after the seed and soil theory, the concept of the tumor microenvironment (TME), where not only the cancer cells but also the surrounding blood vessels, immune cells, other stromal cells with signaling molecules, and the extracellular matrix and interstitial fluid (IF) all participate in cancer progression, is under the spotlight.^3–5 Determining the interaction between cancer and non-cancer cells in the TME is imperative to understanding the mechanisms underlying cancer progression and metastasis, which may open new avenues for developing therapeutics. Cancer cells influence the surrounding microenvironment by releasing bioactive molecules, such as cytokines, chemokines, and lipid mediators.^6–8 Concurrently, non-cancer components in the TME, such as blood vessels, lymphatic vessels, and inflammatory cells, also provide bioactive molecules that influence cancer progression. These bioactive molecules secreted from both cancer and non-cancer cells in the TME have attracted attention as they are potential targets for molecular-targeted therapies or immunotherapies, such as anti-PD-1 therapy. ⁹

The most intensely studied bioactive molecules that mediate cancer and non-cancer cell–cell interactions in the TME are cytokines and chemokines. Cytokines are extracellular signal proteins or peptides that locally mediate cell–cell communication. In contrast, chemokines are small peptides that are structurally and functionally similar to growth factors, binding to G protein–coupled receptors to induce chemoattraction, inflammation, and/or angiogenesis. Compared with cytokines and chemokines, lipids have attracted less attention as the primary signaling molecules in the TME. Determining the quantity and the localization of lipids has been a challenge due to a lack of an established method. Because of these limitations, the importance of lipid mediators in cancer biology has often been overlooked. However, recent advances in lipidomics utilizing mass spectrometry allow for the detection and quantification of lipid mediators in biological samples.^10,11 Increasing evidence suggests that lipid mediators, including sphingolipids in the TME, play pivotal roles in cancer biology.^12–14

Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid mediator, which has emerged in the last decade as a player in TME and cancer progression. ⁷ S1P is secreted into the TME from non-cancer stromal cells and cancer cells.^5,7 Furthermore, the interaction of S1P with cancer cells and non-cancer stromal cells in the TME plays an important role in cancer progression. Here, we review the roles of S1P in the TME and discuss future possibilities for targeted therapy in cancer, including targeting S1P signaling.

S1P, a pleiotropic lipid mediator

S1P is a pleiotropic lipid mediator that regulates cell survival, migration, recruitment of immune cells, angiogenesis, and lymphangiogenesis, which are all factors involved in cancer progression. ⁵ S1P was first reported as a bioactive lipid mediator that affects cellular functions in many biological processes. ¹⁵ S1P is generated from sphingosine by two sphingosine kinases (SphK1 and SphK2; Figure 1) and then exported from the cell via S1P transporters on the cell membrane.^5,16–21 S1P can then stimulate any of five specific G protein–coupled receptors (S1P-specific receptors 1–5; S1PR1–5), ²² which regulate activation or inhibition of the downstream intracellular signaling involved in various cellular functions.

Figure 1.

Production of sphingosine-1-phosphate (S1P) and its export to the tumor microenvironment (TME). S1P is generated from sphingosine by two sphingosine kinases (SphK1 and SphK2) inside cells. SphK1 is located in the cytosol close to the cell membrane where its substrate sphingosine resides. S1P produced by SphK1 is exported to the extracellular space including the TME, where it exerts various functions associated with cancer progression. In contrast, SphK2 is localized in specific organelles, such as the nucleus and mitochondria, and S1P produced by SphK2 is thought to play important roles in intracellular functions.

S1P produced by SphK1, but not by SphK2, is exported to the TME

There are two isotypes of SphK, designated SphK1 and SphK2. SphK1 is located in the cytosol close to the cell membrane where its substrate sphingosine resides, ²³ while SphK2 is localized in specific organelles, such as the nucleus and mitochondria. Importantly, S1P produced inside the cell by SphK1, but not SphK2, is exported to the extracellular space, including the TME.^21,24 SphK1 is ubiquitously distributed in each organ, and is a critical regulator of S1P production that functions in “inside-out” signaling. “Inside-out” signaling refers to the process by which S1P produced inside cells is secreted by transporters and signals through its receptors (S1PRs) on the outside of cells.^5,7,25 Thus, S1P produced by SphK1-overexpressing cancer cells is a major source of S1P in the TME, which promotes cancer progression by “inside-out” signaling.

S1P, produced by SphK2, and a portion produced by SphK1, acts intracellularly, although this is not thought to be associated with the TME directly. S1P produced by SphK2 in the nucleus is an endogenous inhibitor of specific histone deacetylases, linking sphingolipid metabolism to epigenetic regulation of genes in the brain and liver.^26–29 SphK2 in mitochondria produces S1P that interacts with prohibitin 2 to regulate complex IV assembly and respiration. ³⁰ Furthermore, S1P produced by SphK1 can function as a cofactor for the E3 ubiquitin ligase activity of TRAF2 in nuclear factor-κB (NF-κB) signaling. ³¹ In sum, S1P produced by SphK2 or SphK1 can act inside the cell to regulate a variety of cellular functions, while S1P produced by SphK1 can act outside the cell to affect the TME through “inside-out” signaling.

S1P transporters export S1P from cells to the TME

Although a major source of S1P is produced inside cells by SphK1, S1P is unable to freely pass through plasma membranes to the TME because it possesses a polar head group. S1P export requires a carrier or transporter, and studies have suggested the involvement of ATP-binding cassette (ABC) transporters in this process from various types of cultured cells. ²² S1P is exported from mast cells via ABCC1 (also known as multidrug-resistant protein 1; MRP1), ³² from astrocytes via ABCA1, ³³ from endothelial cells via ABCA1 and ABCC1, ³⁴ and from thyroid carcinoma cells via ABCC1. ³⁵ Furthermore, we demonstrated that ABCC1 and ABCG2 (also known as breast cancer resistance protein; BCRP) are involved in estradiol-mediated transport of S1P and dihydro-S1P out of MCF-7 human breast cancer cells. ²¹ These studies suggest that ABC transporters are responsible for export of S1P from cancer cells and non-cancer cells that are part of the TME, all of which contribute to maintain levels of S1P in the TME.

Spns2, a member of the major facilitator superfamily without a typical ATM-binding motif, has recently been found to export S1P from cells.^36–39 Spns2 was identified independently by two groups, both of which revealed that it transports S1P through observations in zebrafish. They showed that a mutation in Spns2 caused abnormal development resulting in cardia bifida (two hearts) and that the phenotype could be rescued by exogenous S1P.^38,40 Interestingly, the same phenotype as in Spns2 knockout was seen in S1PR2 knockout zebrafish.^38,40 Spns2 exports endogenous S1P and dihydro-S1P from vascular endothelial cells, ⁴¹ and Spns2 is important for vascular development. ⁴² We also observed that lymph nodes of Spns2 knockout mice have aberrant lymphatic sinuses that appear collapsed, with a reduced number of lymphatic vessels. ⁴¹ Our data suggest that Spns2 is an S1P transporter that plays a crucial role in regulating S1P levels in the lymph nodes and lymphatic network. ⁴¹ Taken together, S1P transport and extracellular signaling are important as they have implications for the TME in cancer and immune cell trafficking. ⁷

S1P receptors play a role in “inside-out” signaling among cancer and non-cancer cells in TME

Although S1P is a relatively simple molecule, it can evoke diverse biological functions important in cancer progression due to five broadly expressed specific cell surface G protein–coupled receptors (S1PR1–5). Cancer and non-cancer cells in the TME express different combinations of S1P receptors that contribute to the cellular functions regulated by S1P. After intracellular production by SphK1, S1P is released and can activate S1PRs in an autocrine and/or paracrine manner. Here, we introduce the roles of each S1P receptor in the TME and in cancer progression.

The S1PRs display tissue-specific expression patterns and are coupled to various G proteins, enabling them to regulate a broad spectrum of downstream signaling pathways and numerous biological processes.^5,7 For example, S1PR1 is important for B and T lymphocyte egression from secondary lymphatic organs such as lymph nodes.^5,16 In endothelial cells, S1PR1 and S1PR2 are essential in vascular development.^42–45 Stimulation of S1PR1 and/or S1PR3 often promotes cell proliferation and migration in normal and cancer cells, while S1PR2 may inhibit the signaling that promotes cell proliferation and migration.^46–48 “Inside-out” signaling of S1P plays a pivotal role in cancer cells and in the TME by stimulating the S1P receptors, especially S1PR1–3, on each cell type.^49,50

Compared with S1PR1–3, the functions of S1PR4 and S1PR5 are less clear, although recent studies have revealed that they play a role in inflammation. ⁵¹ S1PR4 is related to the migration of neutrophils from blood to tissue. ⁵² S1PR5 is expressed predominantly by oligodendrocytes and/or fibrous astrocytes in the rat brain, and couples with Gi/oα proteins for migration and survival of those cells.^53–55 Patrolling monocytes express high levels of S1PR5 similar to natural killer (NK) cells. However, S1PR5 may regulate monocyte trafficking via a mechanism independent of S1P gradients. ⁵⁶ Taken together, S1PR4 and S1PR5 are important for inflammation, which is also closely related to the TME and cancer progression.

S1P levels in the TME are regulated by S1P lyase and phosphatase

S1P levels are tightly regulated by the balance between synthesis by SphKs from sphingosine that is made from ceramide, and reversible conversion to sphingosine by specific S1P phosphatases (SPP1 and SPP2) and irreversible degradation by S1P lyase (SPL).^7,25,57–59 S1P levels in the blood and the lymph are maintained at high levels, while S1P levels in tissue, such as lymph nodes and thymus, are kept at low levels by SPPs and SPL. It is expected that S1P levels in the TME are also regulated by SPPs and SPL. Importantly, this difference in S1P concentration between blood/lymph and tissue creates a concentration gradient crucial in cell trafficking. ¹⁶ Here, we will highlight the functions of SPPs and SPL in the TME and their involvement in cancer progression.

SPPs regulate levels of S1P and its metabolite ceramide in the TME. Overexpression of SPP1 results in elevation of ceramide that induces apoptosis.^60,61 SPP1 messenger RNA (mRNA) and protein levels in gastric cancer tissues were significantly decreased compared with adjacent normal gastric tissues. ⁶² Furthermore, weakly expressed SPP1, which indicates more S1P and less ceramide in the gastric TME, was positively correlated with lymph node metastasis and distant metastasis, and SPP1 positive expression resulted in a significantly better prognosis. These findings suggest that S1P levels regulated by SPP1 in the TME affect cancer progression and prognosis in patients with advanced gastric cancers.

SPP2 has also been implicated in regulating the levels of S1P in the TME. We recently revealed that SPP2 promotes disruption of mucosal integrity and contributes to ulcerative colitis in mice and humans. ⁶³ Deletion of SPP2, which is mainly expressed in the gastrointestinal tract, significantly reduced the severity of dextran sodium sulfate–induced colitis, possibly by suppressing intestinal epithelial cell apoptosis and improving mucosal barrier integrity. Finally, in ulcerative colitis patients, SPP2 expression was elevated in colitis tissues relative to that in uninvolved tissues. These results indicate that induction of SPP2 expression contributes to the pathogenesis of colitis by promoting disruption of the mucosal barrier function.

The major route of S1P irreversible degradation is through cleavage of the C2-C3 bond by SPL, and SPL affects S1P levels in normal tissues and cancer tissues. ⁵⁹ SPL expression in HEK293 cells potentiated apoptosis in response to stimuli including DNA damage. ⁶⁴ Importantly, SPL expression was significantly down-regulated in human colon cancer tissues in comparison with normal adjacent tissues. Down-regulation of SPPs was also observed, suggesting that colon cancer cells manifest a block in S1P catabolism. In addition, SPL expression and activity were down-regulated in adenomatous lesions of the multiple intestinal neoplasia mouse model of intestinal tumorigenesis.^64,65 Taken together, these results indicate that endogenous SPL plays a physiological role in stress-induced apoptosis and provide an example of altered SPL expression in a human tumor affecting the TME.

Sources of S1P in the TME

S1P levels in the components of the TME, such as blood, lymph, and IF in the connective tissue, are tightly regulated by its formation by SphK1, and its degradation by SPL and SPPs. ⁶⁶ The concentration gradient of S1P between the blood, lymph, and tissues (IF) is important for lymphocyte egression from the lymphoid tissues and the thymus.^67,68 S1P is abundant in blood, where most is contained in erythrocytes, and some is bound to albumin and high-density lipoprotein in serum.^69,70 S1P in the blood can stimulate S1PR1 on the surface of endothelial cells and lymphocytes, modulating vascular permeability and lymphocyte circulation. In contrast, S1P levels in tissues are considerably lower, although tissues with high blood content, such as spleen, are exceptions. ⁷¹ In this section, we will describe the sources of S1P that maintain blood and lymph at higher S1P levels than in tissues.

The source of plasma S1P was originally assumed to be platelets.^72,73 Platelets are characterized by high SphK activity and a lack of SPL, which allows them to accumulate large amounts of S1P.^74,75 However, the activity of serine palmitoyltransferase, the rate-limiting enzyme of the de novo sphingolipid biosynthesis pathway, in platelets is very low. Therefore, in order to synthesize S1P, platelets have to incorporate sphingosine from the plasma. Alternatively, sphingosine can be generated on the outer leaflet of the plasma membrane. ⁷⁶

The major source of S1P in the blood is now considered to be erythrocytes. Similar to platelets, erythrocytes obtain sphingosine by incorporation from the plasma or by generation on the outer leaflet of the cell membrane. ⁷⁴ Release of erythrocyte-generated S1P is stimulated by high-density lipoprotein and serum albumin in blood plasma. ⁷⁷ Furthermore, mice with erythrocyte-specific deletion of SphKs showed embryonic lethality due to defects in vascular development. ⁷⁸ These findings imply that plasma S1P supplied from erythrocytes is essential to regulate vascular and immune cells, as well as embryogenesis.

Vascular endothelial cells are a source of S1P in the circulation, ⁷¹ and can synthesize and release endogenous S1P more efficiently than human colon cancer cell lines. ³⁴ This indicates that stromal cells may also supply S1P in the TME. Similarly, lymphatic endothelial cells are a major source of S1P in lymph, with the transport of S1P from these cells mediated by Spns2.^41,79 Furthermore, knockdown of both SphKs in lymphatic endothelial cells resulted in the absence of S1P in the lymph. ⁶⁷ Recently, we showed that Spns2 plays a role in regulation of blood S1P and lymph node and lymph S1P levels. ⁴¹ Mast cells also play a role in regulating S1P concentration in lymph, with S1P export mediated by ABC transporters independently of their degranulation. ³² Transport of S1P by ABCC1 influenced migration of mast cells toward antigen but not degranulation. Taken together, lymphatic endothelial cells, as well as stromal cells such as mast cells, are a source of S1P, which regulates its concentration gradient and influences immune functions.

S1P in tumor IF, as a component of TME

The high levels of S1P in blood and lymph are considered to affect the S1P levels in the TME. In addition, we have recently shown that high levels of S1P secreted from cancer cells themselves affect the S1P levels in the TME, such as the IF.^24,80,81 Importantly, the levels of S1P in IF have never previously been measured due to a lack of established methods for collection and quantification. We developed an improved centrifugation method to collect IF and to measure sphingolipids by mass spectrometry in IF, blood, and tissue samples. ⁸¹ In mice with a deletion of SphK1, but not SphK2, levels of S1P in IF from the mammary glands were greatly attenuated, indicating that SphK1 plays a major role in S1P secretion to the IF in the mammary glands. Of note, sphingosine and S1P levels were significantly higher in human breast cancer tissue IF than in normal breast tissue IF, suggesting that the cancer cells are the source of the sphingolipids in breast cancer tissue IF. Thus, in addition to blood and lymph, cancer cells provide S1P in the TME, which tightly regulates cancer biology and progression.

Roles of S1P in the interaction between cancer cells and TME

S1P produced by cancer cells or TME may promote cancer progression in three ways (Figure 2.): (1) S1P secreted from cancer cells promotes cancer progression by stimulating proliferation, migration, and survival of cancer cells. ⁵ (2) S1P secreted from cancer cells affects their microenvironment by inducing angiogenesis/lymphangiogenesis, modulating tumor immunity, or evoking inflammation. For instance, we demonstrated that SphK1-produced S1P is a crucial mediator of breast cancer–induced angiogenesis and lymphangiogenesis in culture and in vivo. ²⁴ (3) S1P plays an indirect role in the interaction between cancer cell and non-cancer cell components in the TME, where cytokines such as interleukin-6 (IL-6), tumor necrosis factor α (TNFα), NF-κB, platelet-derived growth factor (PDGF), and fibroblast growth factor (FGF) produced by cancer cells promote S1P production from stromal cells in the TME. ⁴⁹ The S1P produced by blood and lymphatic endothelial cells can attract cancer cells to move toward the vessels, and S1P from stromal cells in the TME can stimulate cancer cell proliferation/survival.

Figure 2.

Roles of sphingosine-1-phosphate (S1P) in the interaction between cancer cells and the tumor microenvironment (TME). S1P produced by cancer cells or the TME is involved in cancer progression in three ways: (a) S1P secreted from cancer cells stimulates proliferation, migration, and survival of cancer cells in an autocrine and/or paracrine manner; (b) S1P secreted from cancer cells affects the TME by inducing angiogenesis/lymphangiogenesis; and (c) S1P has an indirect effect on the interaction between cancer cell and non-cancer cell components in the TME, where cytokines secreted from cancer cells promote S1P production from stromal cells, such as fibroblasts, immune cells, and endothelial cells in the TME. The S1P produced by blood and lymphatic endothelial cells can attract cancer cells to move toward the vessels, and S1P provided from stromal cells in the TME can stimulate cancer cell proliferation/survival.

We have shown that S1P links inflammation and cancer in colitis-associated colon cancer. ⁴⁹ S1P produced by upregulation of SphK1 links chronic intestinal inflammation to colitis-associated cancer, and both are exacerbated by deletion of SphK2. S1P is essential for the production of the multifunctional NF-κB-regulated cytokine, IL-6, persistent activation of the transcription factor, STAT3, and consequent upregulation of the S1P receptor, S1PR1. The prodrug, FTY720, decreased SphK1 and S1PR1 expression and eliminated the NF-κB/IL-6/STAT3 amplification cascade and development of colitis-associated cancer, even in Sphk2^−/− mice, and may be useful in treating colon cancer in individuals with ulcerative colitis. Thus, the SphK1/S1P/S1PR1 axis is at the nexus between NF-κB and STAT3, and connects chronic inflammation and colitis-associated cancer. Considering that S1P is one of the key molecules in the interaction between cancer cells and stromal cells in the TME, there is an opportunity to treat cancer by targeting S1P from cancer cells and/or stromal cells, potentially interrupting the interaction between cancer cells and the TME.

The potential of targeting S1P in the TME for cancer therapy

Recently, the TME has attracted attention as a therapeutic target. Good examples of therapeutic agents for the TME are antiangiogenic agents targeting vascular endothelial growth factor (VEGF) or its receptors.^82,83 Growing evidence implicates inside-out signaling of S1P as a key regulator of angiogenesis. Expression of S1PR1 is upregulated in tumor vessels, and its down-regulation was effective in inhibiting angiogenesis and tumor growth in vivo, suggesting that S1PR1 is a critical component of tumor-induced angiogenesis. ⁸⁴ The expression of S1PR1, and levels of S1P in IF, can be suppressed by orally administering FTY720, which has been approved by the US Food and Drug Administration (FDA) for multiple sclerosis.^5,81 SphK1 can also be a target of tumor-induced angiogenesis. We showed that the specific SphK1 inhibitor, SK1-I, suppressed angiogenesis and lymphangiogenesis with reduced levels of S1P in the blood, resulting in suppression of metastases to lymph nodes and lungs, and decreased overall tumor burden. Furthermore, neutralization of extracellular S1P with an anti-S1P antibody significantly inhibited angiogenesis, tumor growth, and metastasis, further confirming the dominant role that extracellular S1P plays in this process. ⁸⁵

Anti-S1P therapy can also target inflammation as components of the TME. We have recently discovered that the SphK1/S1P/S1PR1 axis plays a critical role in inflammation-associated cancer progression utilizing a colitis-associated colon cancer mouse model. Indeed, it is known that pro-inflammatory cytokines, such as TNFα, upregulate SphK1.^86,87 Interestingly, we have shown that the S1P levels in blood are significantly elevated in animals with inflammation, and FTY720 suppresses cancer progression more efficiently in these animals. Although targeting S1P pathways in TME holds promise for cancer patients, further investigation is needed to develop new therapies with these strategies.

Conclusion

S1P plays an important role in cancer biology, especially in the complex interactions between cancer cells and the TME. With the increased focus on the TME in cancer treatment in cancer drug discovery, S1P provides exciting opportunities for research. Because S1P is present at a high concentration in the tumor IF and plays a key role in inflammation and cancer biology, including interactions between cancer cells and the TME, S1P and its associated pathways offer investigators a broad range of opportunities to develop novel therapeutic strategies for targeted treatments in cancer patients.