Autotaxin-Lysophosphatidic Acid: From Inflammation to Cancer Development
Lysophosphatidic acid (LPA) is a ubiquitous lysophospholipid and one of the main membrane-derived lipid signaling molecules. LPA acts as an autocrine/paracrine messenger through at least six G protein-coupled receptors (GPCRs), known as LPA1–6, to induce various cellular processes including wound healing, differentiation, proliferation, migration, and survival. LPA receptors and autotaxin (ATX), a secreted phosphodiesterase that produces this phospholipid, are overexpressed in many cancers and impact several features of the disease, including cancer-related inflammation, development, and progression. Many ongoing studies aim to understand ATX-LPA axis signaling in cancer and its potential as a therapeutic target. In this review, we discuss the evidence linking LPA signaling to cancer-related inflammation and its impact on cancer progression.
Lysophosphatidic acid (LPA) consists of an acyl chain at the sn-1 (or sn-2) position of a glycerol backbone and a phosphate head group. It is the smallest (molecular weight: 430–480 Da) and the simplest bioactive glycerophospholipid derived from membrane phospholipids [1, 2]. Nevertheless, it is involved in a wide range of activities, from phospholipid synthesis to a number of physiological responses as a lipid mediator . LPA activates at least six G-coupled protein receptors (LPA1–6) stimulating different signaling pathways through heterotrimeric G proteins such as Gi/0, G12/13, Gq/11, and Gs. The outcome of LPA signaling is dependent on cellular context and impacts on biological processes such as wound healing, differentiation, neurogenesis, and survival, to name a few . Due to its small structure, LPA is water soluble and concentrations > 5 μM have been reported in serum; concentrations < 1 μM have been found in other biofluids such as plasma, saliva, follicular fluid, cerebrospinal fluid, and malignant effusions [5–7]. It is known that ATX-LPA signaling increases during wound healing, and both are produced and detected in blister fluids, where they mediate platelet aggregation and skin reepithelization . During this process, ATX-LPA signaling induces production of proinflammatory cytokines. Therefore, aberrant activation of this axis promotes an inappropriate immune response that leads to a proinflammatory state in pathologies like cancer .
2. Lysophosphatidic Acid Synthesis and Metabolism
LPA is a membrane-derived lysophospholipid from phosphatidylcholine (PC), phosphatidylserine (PS), and phosphatidylethanolamine (PE) . Therefore, several species can be found, differing only in the length and saturation of the acyl or alkyl fatty acid chain [7, 10]. The most abundant plasma LPA species are 18:2 > 18:1 ≥ 18:0 > 16:0 > 20:4 with an acyl group [11, 12]. Although acyl-LPA 18:2 is the most numerous species, acyl-LPA 18:1 is the most frequently used in current research .
There are two major pathways for LPA production (Figure 1(a)). The main pathway is the cleavage of membrane phospholipids into lysophospholipids by the removal of a fatty acid chain by phospholipase A (PLA1 or PLA2). Subsequently, ATX cleaves the head group (choline, ethanolamine, or serine) on the lysophospholipids and turns them into LPA . ATX (also known as ENPP2) is a 125 kDa-secreted enzyme from the family of ectonucleotide pyrophosphatases/phosphodiesterases (reviewed by ) located on Chr8q24 . Among the seven members of this family, ATX is a unique enzyme that shows lysophospholipase D activity [17, 18]. This enzyme produces most of the extracellular LPA. Enpp2+/− mice and inhibitors targeting ATX decrease LPA plasma concentration by >50% [19–22]. ATX generates LPA from plasma membrane phospholipids and from circulating lysophosphatidylcholine (LPC) bound to albumin . ATX is essential for development since Enpp2−/− is lethal at embryonic day 9.5–10.5, with marked vascular and neural tube defects [20, 21]. ATX is also important in adipogenesis since it is upregulated during preadipocyte differentiation to adipocytes and secreted into circulation by the adipose tissue .
A second, less common, route of LPA production is the cleavage of phospholipids into phosphatidic acid (PA) by phospholipase D (PLD) at the cell surface. PA is then hydrolyzed in the outer leaflet of the plasma membrane by secreted PLA2 (sPLA2) releasing LPA to the microenvironment .
LPA turns over with a half-life of about 3 min in the circulation . Therefore, its main effects are autocrine and paracrine when bound to albumin . LPA turnover is regulated by ATX activity and LPA degradation by lipid phosphate phosphohydrolase type 1 (LPP1) which hydrolyze LPA into monoacylglycerol (MAG) in the outer leaflet of the cell membrane [26, 27] and LPA-acyltransferase (LPAAT), which transfer an acyl chain to LPA converting it into PA in the inner leaflet of the cell membrane . Recently, a negative feedback loop has been described for the ATX-LPA axis ; in this mechanism, LPA signaling through its receptor LPA1/3 induces downregulation of ATX mRNA. Similarly, low levels of circulating LPA increase ATX mRNA, particularly in the adipose tissue of female Balb/c mice .
3. LPA Receptors
As previously mentioned, LPA signals through at least six G protein-coupled receptors LPA1–6 (Figure 1(b)): gene names are LPAR1-LPAR6 (human) and Lpar1-Lpar6 (mouse) [30, 31]. All LPA receptors are rhodopsin-like, with seven transmembrane domain receptors that range from 39 to 42 kDa and differ in their tissue distribution and downstream effectors . According to their homology, there are two LPA receptor families: the endothelial differentiation gene (EDG) family and the non-EDG family [32, 33]. In addition to homology, they differ in their activation by different LPA species (Figure 2). Although acyl-LPA 18:2 is the most abundant species, the EDG family is more potently stimulated by acyl-LPA (LPA1/2), and LPA3 preferentially bounds to 2-acyl-LPA. The non-EGD family member LPA5 is more potently stimulated by alkyl-LPA and LPA6 by 2-acyl-LPA, specifically . These differences show that a wide range of physiological effects is modulated through these receptors and LPA species in a context and cell type-dependent manner.
3.1. Endothelial Differentiation Gene Family
In 1996, LPA1 was the first receptor to be identified and it is the best studied to date. Hecht et al.  described a neuroblast cell line overexpressing the ventricular zone gene-1 receptor (Vgz-1), to which LPA binds specifically to induce cell rounding and activation of Gαi. Also known as EDG-2, Vgz-1 was later renamed LPA1. Right after its discovery, two other orphan receptors, LPA2 and LPA3, were identified based on their homology to LPA1 [36–38].
LPA1 is a 41 kDa protein of 364 amino acids located in Chr9q31.3 and consists of at least 5 exons [30, 31]. This receptor couples with and activates 3 types of G protein, Gαi/0, Gαq/11, and Gα12/13, which initiate downstream signaling through PI3K/AKT, Rho, MAPK, and PLC (Figure 1(b)). These pathways are involved in several cellular processes, including cell proliferation and survival, adhesion, migration, AC inhibition, and Ca2+ mobilization [31, 39]. It is widely expressed in most tissues such as brain, uterus, testis, lung, small intestine, heart, stomach, kidney, spleen, thymus, and skeletal muscle at different developmental stages with a variable expression, particularly in the central nervous system (CNS) [36, 39], where, during development, LPA1 is found in the ventricular zone, superficial marginal zone, and meninges. After birth, LPA1 expression is reduced in the aforementioned areas and continues in oligodendrocytes, particularly during myelination, as well as in astrocytes, where it elicits a wide range of processes (reviewed by ). Targeted deletion of Lpar1−/− showed a 50% of perinatal lethality related to an impaired suckling behavior probably due to defective olfaction. Surviving mice showed craniofacial malformations and reduced body size . Additionally, LPA1 has been closely related to the induction of neuropathic pain due to nerve injury via LPA1/RhoA/rock-mediated demyelination with a subsequent loss of the structural and functional integrity of neurons, as discussed elsewhere .
LPA2 receptor (EDG-4) has a ~50–60% homology to LPA1, with an estimated mass of 39 kDa and 348 amino acids . Located on Chr19p12, it consists of 3 exons in both humans and mice [30, 39]. LPA2 couples to the same G proteins as LPA1 (Figure 1(b)): Gαi/0, Gαq/11, and Gα12/13 [36, 39]; therefore, it can similarly activate downstream signaling but, unlike LPA1, can also promote migration through the focal adhesion molecule TRIP6 [43, 44]. LPA2 activation is associated with survival and migration. Compared with LPA1, its expression is more diffuse during development, more restricted in adults, and with high expression in leukocytes and testis in humans and in kidney, uterus, and testis in mice [36, 39, 45]. LPA2 knockout mice are mostly normal, suggesting a possible functional redundancy in relation to LPA1. A Lpar1−/− and Lpar2−/− model has also been evaluated . In this model, Lpar1−/− phenotype predominated with 50% perinatal lethality, cranial malformations, and reduced body size, but it also exhibited frontal hematomas .
LPA3 receptor (EDG-7) contains 3 exons, has 353 amino acids, and a 40 kDa-estimated mass [37, 38]. This receptor has 52% and 48% homology with LPA1 and LPA2, respectively, and is located on Chr1p22.3-p31.1 [30, 38, 39]. LPA3 couples to G proteins, Gαi/0 and Gα11/q (Figure 1(b)), and therefore mediates downstream activation of MAPK, PLC, and inactivation of AC . It has been reported that this receptor is more potently activated by 2-acyl-LPA with unsaturated fatty acids . In humans, LPA3 is expressed in heart, lung, pancreas, prostate, testis, ovaries, and brain . In mice, it is expressed in testis, kidney, lung, intestine, and moderately, small intestine . Functional deletion of LPA3 in female mice showed delayed and defective embryo implantation through the downregulation of cyclooxygenase-2 (COX-2) and reduced levels of prostaglandins, which are essential for this process . In deficient LPA1–3 male mice, an independent of testosterone signaling reduced sperm count and mating activity was found . This evidence suggests the role of LPA3 in reproductive functions.
3.2. Nonendothelial Differentiation Gene Family
In 2003, the first LPA receptors structurally distant from the EDG receptor family were described . The orphan GPCR P2Y9/GPR23 has only 20–24% homology with LPA1–3, but it specifically binds to LPA. Its signaling promotes an increase in intracellular Ca2+ concentration and adenylyl cyclase activity in “LPA receptor-null” cells exogenously expressing P2Y9 . Soon, LPA5 and LPA6 description followed [51–55].
LPA4 (P2Y9/GPR23) is encoded by 1 exon containing 370 amino acids with a 42 kDa mass [30, 50, 56]. Located on ChrXq21.1, it was the first to be described that couples to four G proteins: Gαi/0, Gα11/q, Gα12/13, and Gαs (Figure 1(b)) . LPA4 signaling promotes Rho-mediated neurite retraction and stresses fiber formation, Ca2+ mobilization, and regulation of cAMP concentration . In humans, LPA4 expression is high in ovaries, moderated in thymus and pancreas, and low in brain, heart, small intestine, testis, prostate, colon, and spleen [13, 50]. In mice, it is expressed in heart, ovaries, thymus, skin, and developing brain [57, 58]. Lpar4 −/− mice showed no apparent abnormality, but there was a 30% lethality, probably due to blood vessel defects during embryogenesis [58, 59].
LPA5 (GPR92) is a 41 kDa protein consisting of 372 amino acids coded in an intronless open reading frame [51, 52]. This receptor is located on Chr12p13.31 and has a 35% homology with LPA4 [51, 52]. LPA5 couples to G proteins, Gα11/q and Gα12/13 (Figure 1(b)), by which Ca2+ mobilization, inositol phosphate production, neurite retraction, and stress fiber formation are mediated [51, 52]. It has been reported that LPA5 preferentially binds to alkyl-LPA (16:0), rather than acyl-LPA (18:1) . LPA5 is found in heart, placenta, spleen, brain, lung, and gut in humans . It is also highly expressed in the lymphocyte compartment of the gastrointestinal tract and platelets [51, 60]. In mice, it is found in the brain, heart, kidney, liver, lung, muscle, skin, spleen, stomach, small intestine, testis, and thymus . Lpar5−/− mice have no apparent phenotypic defects but show a reduced pain sensitivity, faster recovery from inflammation, and reduction in social exploration [61, 62]. They also exhibit nocturnal hyperactivity and anxiety compared to Lpar5+/+ mice . Null mice were also protected from developing neuropathic pain by a mechanism different from LPA1 .
LPA6 (P2Y5) is the most recently identified LPA receptor and the last accepted by the IUPHAR Nomenclature Committee in 2010 [31, 53, 54]. It is a 344-amino acid protein with an estimated mass of 39 kDa . Regarding homology with LPA4 , it is the closest receptor and is located on Chr13q14 [30, 55]. LPA6 couples to Gαi/0 and Gα12/13 (Figure 1(b)), by which a decrease in cAMP, Rho-dependent morphological changes, Ca2+ mobilization, and MAPK activation are mediated [53, 54]. It has also been reported that LPA6 is preferentially activated by 2-acyl-LPA, rather than 1-acyl-LPA . This receptor has been found in rats’ brain, heart, lung, kidney, pancreas, liver, stomach, and small and large intestine . In humans, it has been related to hair growth since a mutation of LPAR6 was found in patients with hypotrichosis simplex, an alopecia-causing disorder .
3.3. EDG and Non-EDG Receptor Effects in Cancer
Extensive evidence demonstrate that the receptors from the EDG family promote tumor progression in a wide variety of cancers by enhancing proliferation, survival, migration, and invasion . Conversely, evidence shows that members from the non-EDG family have the opposite effect.
Reconstitution of Lpar4 in mouse embryonic fibroblasts derived from Lpar4−/− mice reduces cell motility due to an LPA-induced decrease in Rac activation . Also, LPA4 expression in colon cancer cells (DLD1 and HTC116) suppresses cell migration and invasion compared to null-LPA4 cells [58, 63]. Similarly, in rat sarcoma cells, overexpression of Lpar5 significantly reduced motility and suppressed MMP2 activation. On the other hand, Lpar5 knockdown induced the opposite effect . In B16F10 mice melanoma cells, LPA5 reduced migration through a cAMP/PKA-dependent pathway and induced chemorepulsion instead of attraction via LPA . Additionally, in colon cancer cells, lines DLD1, and HCT116, LPA6 expression significantly reduced cell growth and motility .
In rat lung adenocarcinoma, loss of LPA3 due to methylation of the promoter enhances tumor progression by increasing invasion, suggesting a protective role of LPA3 in this neoplasia . By contrast, in human fibrosarcoma, LPA4 was shown to increase cAMP levels and subsequently activate Rac1 to induce invadopodia, a process directly correlated with invasion and metastasis . Additionally, in rat lung carcinoma, LPA5 is highly expressed due to unmethylation of the promoter, and cells expressing only LPA5 showed enhanced proliferation, migration, and invasion . Moreover, hepatocellular carcinoma (HCC) cells overexpressing LPA6 sustain an increase in tumor growth, migration, and invasion. Moreover, LPA6 expression was associated with a worse clinical outcome in these patients .
In brief, LPA receptors can have homologous and antagonistic effects depending on the tumor. Therefore, they should be studied in a cancer-specific context to better evaluate their role in tumor development and progression, as well as their potential therapeutic value.
4. Autotaxin-LPA Axis in Cancer-Related Inflammation
Since the 19th century, an association between inflammation and cancer was proposed . Inflammatory components are often present in most types of cancer, such as white blood cells, tumor-associated macrophages, and proinflammatory ILs [70, 71]. In several cases, inflammation can predispose individuals to certain types of cancer, including cervical, gastric, colon, hepatic, breast, lung, ovarian, prostate, and thyroid cancer [72–81]. There is also evidence that the use of nonsteroidal anti-inflammatory drugs can reduce the risk of developing colon and breast cancer and reduce the related mortality, as discussed elsewhere [82, 83].
In general, two mechanisms have been proposed to link inflammation and cancer. In the intrinsic pathway, genetic events promoting development initiate the expression of inflammation-related circuits leading to an inflammatory microenvironment. Conversely, in the extrinsic pathway, inflammatory conditions facilitate cancer development. In both cases, a cancer-related inflammation (CRI) is induced and it is proposed as a tumor-enabling characteristic and the seventh hallmark of cancer . CRI enables unlimited replicative potential, independence of growth factors, resistance to growth inhibition, escape of cell death, enhanced angiogenesis, tumor extravasation, and metastasis . Therefore, understanding key components of inflammation is important for better therapeutics in cancer and other diseases.
The ATX-LPA axis is involved in wound healing response, where it induces platelet aggregation, lymphocyte homing, cytokine production, keratinocyte migration, proliferation, and differentiation under physiological conditions . When acute inflammation becomes chronic in unpaired homeostasis, ATX-LPA signaling induces an augmented cytokine production and lymphocyte infiltration, aggravating the inflammation in conditions such as asthma, pulmonary fibrosis, and rheumatoid arthritis, to name a few . In a cancer context, it also promotes cell survival, proliferation, migration, invasion, and angiogenesis, enhancing its progression in a state similar to a “wound that never heals” [84, 87].
ATX-LPA axis has been studied in airway inflammation where protein kinase C δ (PKCδ) mediates LPA-induced NFκB transcription and IL-8 secretion in human bronchial epithelial cells (HBEpCs) ; LPA activation of PKCδ/NFκB and IL-8 production were inhibited by rottlerin (a nonspecific PKCδ inhibitor) and by an overexpression of dominant-negative PKCδ. In vivo LPA administration in mice leads to increased levels of a murine homolog of IL-8 and of neutrophils in the bronchoalveolar fluid . Moreover, LPA signaling induces EGFR transactivation via Lyn kinase, from Src kinase family, to promote matrix metalloprotease (MMP) secretion as well as IL-8 . Additionally, activation of the signal transducers and activators of the transcription 3 (STAT3) in alveolar epithelial cells during host defense promotes inflammation and spontaneous lung cancer . Through these signaling cascades, a chronic inflammation is pursued and could lead to malignant transformation. In lung cancer, inhibition of ATX-LPA axis reduced cell migration, invasion, and vascularization in a 3-D lung cancer xenograft model . There is evidence that ATX is highly expressed in poorer differentiated lung carcinomas, particularly in tumor-adjacent B lymphocytes  and that LPA5 may play a key role in the progression of these carcinomas , while LPA3 could have a protective role . Furthermore, LPA and other phospholipid levels are upregulated as a side effect of chemo- and radiotherapy, inducing a prometastatic microenvironment in lung cancer . Interestingly, LPA did not induce proliferation nor survival in these cells, but rather an increase in motility, adhesion to bone marrow stroma, and enhanced secretion of ATP, another potent chemokinetic factor, from stroma cells . Together, evidence suggests a significant role of ATX-LPA axis in inflammation and lung cancer through the increase of proinflammatory cytokines.
In breast cancer (BCa), the ATX-LPA axis induces inflammation and tumor formation in the mammary gland through LPA1–3 and high ATX expression, which is produced in the adjacent mammary adipose tissue rather than actual cancer cells [94, 95]. Individual overexpression of each of the EDG family receptors, but especially of LPA2, induced a high frequency of late-onset, estrogen receptor (ER) positive, and invasive and metastatic mammary cancer . Moreover, bone metastases are frequent in BCa; ATX expression in these tumors can control the progression of osteolytic bone metastases in vivo through the procoagulant activity of BCa cells that induce platelet-derived LPA .
ATX-LPA axis is a strong inducer of inflammatory mediators like IL-8, IL-6, TNF-α, and growth factors such as the vascular endothelial growth factor (VEGF) and the granulocyte colony-stimulating factor (G-CSF) . Some molecules (IL-8 and VEGF) were detected earlier than tumorigenesis in vivo . Inhibition of ATX induced a twofold reduction in at least 20 of these inflammatory mediators in the tumor-adjacent mammary adipose tissue-reducing inflammation and tumorigenesis . Additionally, expression of LPA1–3 increased phosphorylation of STAT3, STAT5, NFκB and ATF2, and master inflammatory transcription factors, in mouse mammary carcinomas . Furthermore, cytokines produced in the microenvironment (i.e., IL-6) can activate STAT3 through its receptors inducing an inflammatory loop . Adipose tissue adjacent to breast tumors stimulates autotaxin (ATX) secretion, which increases tumor growth and metastasis . Interestingly, radiotherapy in adipose tissue of rats and humans increased mRNA expression of ATX, multiple inflammatory mediators, and LPA1–2. Such effect could promote LPA signaling and further inflammatory signaling, which in turn could potentially protect cancer cells from subsequent radiation therapy . ATX inhibition reduced the leukocyte infiltration and tumor growth in vivo . All these evidence suggest that chronic inflammation contributes to tumor development in BCa. Controlling inflammation and cancer progression could be achieved by targeting the ATX-LPA axis.
In ovarian cancer (OC), ATX is highly expressed and secreted by cancer cells . Therefore, LPA is present at high concentrations in the ascites fluid of OC patients compared to benign and healthy controls and has been proposed as a potential biomarker [100–102]. LPA acts as a growth factor and prevents apoptosis in OC cells by signaling through redox-dependent activation of ERK, AKT, and NFκB signaling pathways. Inhibiting ROS production blocked LPA/NFκB signaling and cell proliferation . Additionally, LPA has been shown to upregulate the expression of human telomerase reverse transcriptase (hTERT) and telomerase activity in OC cell lines, through a PI3K and HIF-1α-dependent mechanism, enabling replicative immortality . On the other hand, OC cell lines, SKOV-3, and OVCAR3 that expressed increased LPA1–3 receptors showed more invasiveness compared to knockdowns. Moreover, via LPA2–3, OC cells promote production of IL-6, IL-8, and VEGF in vitro  and induced urokinase plasminogen activator (uPA) secretion in a MAPK- (p38) and PI3K-dependent mechanism that required Src kinase for optimal MAPK phosphorylation, enhancing OC invasion .
Liver cirrhosis, a terminal stage of chronic inflammatory and fibrotic liver diseases, and chronic hepatitis C are distinct risk factors for hepatocellular carcinoma (HCC) [107, 108]. Increased serum ATX activity and plasma LPA levels have been found in patients with chronic hepatitis C in association with a histological stage of liver fibrosis . Furthermore, in HCC, ATX is expressed in 89% of tumor tissues, especially in those with cirrhosis or hepatitis C, compared to 20% in normal hepatocytes . Additionally, in HCC cell lines, TNF-α/NFκB pathway, known to contribute to inflammation-associated cancer, was shown to upregulate ATX expression and LPA production. The latter resulted in an increased cellular invasion . Similarly, LPA modulates tumor microenvironment by inducing transdifferentiation of peritumoral fibroblasts to a CAF-like myofibroblastic phenotype which enhances proliferation, migration, and invasion in HCC . Additionally, LPA6 mediates tumor growth and tumorigenicity by upregulating Pim-3 protooncogene through a STAT3-dependent mechanism . Recently, human cirrhosis regulatory gene modules were identified through a transcriptome meta-analysis . This analysis provides an overview of a molecular dysregulation common to a wide range of liver disease etiologies in which the ATX-LPA axis is a central regulator . This study marks a great breakthrough in the area and provides a promising target for HCC chemoprevention through this axis; mainly due to the compounds of ongoing clinical trials on idiopathic pulmonary fibrosis and systemic sclerosis (Table 1). If approved, they could be tested as preventive therapy in cirrhosis patients and as adjuvant therapy in HCC [107, 111].
In human colorectal cancer (CC), expression of LPA1 and LPA2 is increased compared to normal mucosa. Conversely, LPA3 has a low expression in malignant tissues . Evidence suggests a probable role of LPA1/2 receptors in CC. Furthermore, LPA-stimulated proliferation through the MAPK pathway, as well as migration through Rho kinase, and chemoresistance through the PI3K/AKT pathway . Inflammation is an established risk for developing CC. Interestingly, in a colitis-associated mice cancer model, Lpar2−/− showed a decrease in tumor incidence and in progression to colon adenocarcinomas by reducing proliferation and proinflammatory factors such as monocyte chemoattractant protein-1 (MCP-1) and macrophage migration inhibitory factor (MIF) . The latter affected the infiltration of macrophages to the tumor microenvironment . Moreover, although LPA increased tumor incidence in ApcMin/+ mice predisposed to adenomas, in Lpar2−/− ApcMin/+, tumor incidence was reduced by 50% [114, 115]. In addition, the expression levels of KLF5, cyclin D1, c-Myc, and HIF-1α were lower compared to ApcMin/+ mice, while β-catenin was primarily cytoplasmic in Lpar2−/− ApcMin/+ mice compared to its nuclear localization in ApcMin/+ mice . This evidence suggests an important role of ATX-LPA axis in tumorigenesis derived from colon chronic inflammation.
Along with cancers previously described, ATX-LPA axis and its signaling pathways have been studied in several other carcinomas such as melanoma, where LPA signaling suppresses antigen receptor signaling, cell activation, and proliferation in CD8 T cells that express LPA5, inhibiting immune response  and promoting tumorigenesis. In pancreatic cancer, LPA1 and LPA3 promote proliferation, invasion through MMP2 secretion, and activation of focal adhesion kinase (FAK) and Paxillin, as well as drug resistance [117, 118]. In glioblastoma multiforme (GBM), an increased ATX-LPA axis has been described to promote cell proliferation and migration through LPA1 . GBM is also characterized by high levels of inflammatory mediators and activation of AKT and NFκB signaling pathways, although the link between ATX-LPA and inflammation remains to be studied . In thyroid cancer, ATX is highly expressed in papillary thyroid carcinomas compared with benign neoplasm . ATX-LPA axis induces at least 16 inflammatory mediators, including IL1-β, IL6, IL8, G-CSF, and TNF-α in vivo; at the same time, these mediators induce ATX expression and increase LPA levels. Blocking the ATX-LPA axis induced a reduction of inflammatory mediators, tumor volume, and angiogenesis . In renal cell carcinoma, ATX-LPA axis is associated to chemoresistance through LPA1. Coadministration of Ki16425, an LPA1/3 antagonist, with sunitinib, a tyrosine kinase inhibitor, prolonged the responsiveness of renal cell carcinoma to sunitinib in xenograft models .
So far, the evidence shows that ATX-LPA signaling in cancer is more complex than previously thought. In addition to promoting proliferation, aggressiveness, and metastasis, it induces an enabling inflammatory setting (Figure 3) and contributes to the differentiation of CAFs , leukocyte infiltration [92, 116], angiogenesis , and stem cell maintenance ; all of them are important components of tumor microenvironment (Figure 4). Thus, the ATX-LPA axis represents a crucial target to reduce CRI and cancer progression.
5. Targeting Autotaxin-LPA Axis for Cancer Therapy
LPA signaling is regulated by ATX activity, LPA receptors, and LPA degradation by LPP1 and LPAAT [125, 126]. In numerous cancers, ATX protein is overexpressed, leading to increased LPA levels in the tumor microenvironment and peripheral blood [99, 101, 127]. Cancer cells have a higher LPA receptor content on their cell surface compared to normal and benign cells and a downregulated expression of LPPs . Therefore, targeting LPA signaling through these components is currently under study and constantly reviewed [4, 127, 129–132]. In this section, we summarize some of the drugs studied regarding ATX inhibition and LPA receptor antagonism (Table 1).
ATX-LPA axis has been shown to induce chemoresistance by upregulating antioxidant genes, multidrug-resistant transporters (ABCC1, ABCG2, ABCC2, and ABCC3), aldehyde dehydrogenase 1 (ALDH1), and stem cell maintenance [99, 136]. Additionally, ATX is among the top 40 most upregulated genes in metastatic cancer . Therefore, inhibition of the axis has shown great results as adjuvant therapy to enhance both chemo- and radiotherapy in vitro and in vivo, as well as tumor growth reduction. Additionally, as we described, CRI is an enabling setting for tumor development. We suggest that a strategy to be considered regarding the ATX-LPA axis in CRI should be a multitarget approach, where both proinflammatory cytokines and ATX-LPA are taken into consideration for better outcomes.
Currently, drugs of ongoing clinical trials are for noncancer diseases; nevertheless, once approved, they could be tested in various cancers. Meanwhile, improvement of physiological and pathological knowledge regarding signal transduction by this axis will lead to the development of more specific therapeutic drugs to better target this signaling cascade.
The ATX-LPA signaling pathway is physiologically relevant during development and adulthood. Dysregulation of this axis is linked to several pathologies, including inflammation-related conditions such as rheumatoid arthritis, fibrosis, neuropathic pain, and cancer. In cancer, it has a major involvement in key components of the microenvironment, including leukocyte infiltration, angiogenesis, and decreased immune response. Interestingly, this axis has been shown to mediate cancer-related inflammation through diverse signaling pathways, crosstalk, and positive loops. Therefore, it enhances a proinflammatory microenvironment and, at the same time, ATX-LPA signaling augments. Breaking the inflammatory cycle and blocking LPA signaling and production should provide an innovative treatment for cancer by decreasing CRI, tumor growth, metastasis, and resistance to cancer treatments. Recent evidence in cirrhosis patients point to this axis as a key regulator in HCC tumorigenesis, providing a very interesting potential target for cancer prevention.
As we wait for ATX-LPA inhibitors to move from preclinical into clinical trials, further investigation is needed regarding this complex signaling pathway to achieve more efficient therapeutics in cancer and other ATX-LPA axis-related pathologies.
|GPCR:||G protein-coupled receptor|
|sPLA2:||Secreted phospholipase A2|
|LPP1:||Lipid phosphate phosphohydrolase type 1|
|LPAAT:||Lysophosphatidic acid acyltransferase|
|cPLA2:||Cytosolic phospholipase A2|
|EDG family:||Endothelial differentiation gene family|
|Vgz-1:||Ventricular zone gene-1|
|CNS:||Central nervous system|
|PKC:||Protein kinase C|
|HBEpCs:||Human bronchial epithelial cells|
|Stat:||Signal transducers and activators of the transcription|
|TNF-α:||Tumor necrosis factor α|
|VEGF:||Vascular endothelial growth factor|
|G-CSF:||Granulocyte colony-stimulating factor|
|NFκB:||Nuclear factor kappa-light-chain-enhancer of activated B cells|
|ATF2:||Activating transcription factor 2|
|ROS:||Reactive oxygen species|
|hTERT:||Human telomerase reverse transcriptase|
|uPA:||Urokinase plasminogen activator|
|MCP-1:||Monocyte chemoattractant protein-1|
|MIF:||Macrophage migration inhibitory factor|
|KLF5:||Krüpple-like factor 5|
|FAK:||Focal adhesion kinase|
|EGFR:||Epidermal growth factor receptor|
|ALDH1:||Aldehyde dehydrogenase 1.|
Conflicts of Interest
The authors declare that there is no conflict of interest regarding the publication of this paper.
This work was supported by UNAM-PAPIIT IA200718. Silvia Anahi Valdés-Rives is a doctoral student from Programa de Doctorado en Ciencias Biomédicas, Universidad Nacional Autónoma de México (UNAM), and received Fellowship 582548 from CONACYT.
D. L. Baker, D. M. Desiderio, D. D. Miller, B. Tolley, and G. J. Tigyi, “Direct quantitative analysis of lysophosphatidic acid molecular species by stable isotope dilution electrospray ionization liquid chromatography–mass spectrometry,” Analytical Biochemistry, vol. 292, no. 2, pp. 287–295, 2001.View at: Publisher Site | Google Scholar
A. Tokumura, E. Majima, Y. Kariya et al., “Identification of human plasma lysophospholipase D, a lysophosphatidic acid-producing enzyme, as autotaxin, a multifunctional phosphodiesterase,” Journal of Biological Chemistry, vol. 277, no. 42, pp. 39436–39442, 2002.View at: Publisher Site | Google Scholar
G. Ferry, E. Tellier, A. Try et al., “Autotaxin is released from adipocytes, catalyzes lysophosphatidic acid synthesis, and activates preadipocyte proliferation. Up-regulated expression with adipocyte differentiation and obesity,” Journal of Biological Chemistry, vol. 278, no. 20, pp. 18162–18169, 2003.View at: Publisher Site | Google Scholar
H. M. H. G. Albers, A. Dong, L. A. van Meeteren et al., “Boronic acid-based inhibitor of autotaxin reveals rapid turnover of LPA in the circulation,” Proceedings of the National Academy of Sciences of the United States of America, vol. 107, no. 16, pp. 7257–7262, 2010.View at: Publisher Site | Google Scholar
J. H. Hecht, J. A. Weiner, S. R. Post, and J. Chun, “Ventricular zone gene-1 (vzg-1) encodes a lysophosphatidic acid receptor expressed in neurogenic regions of the developing cerebral cortex,” The Journal of Cell Biology, vol. 135, no. 4, pp. 1071–1083, 1996.View at: Publisher Site | Google Scholar
J. J. A. Contos, I. Ishii, N. Fukushima et al., “Characterization of lpa2 (Edg4) and lpa1/lpa2 (Edg2/Edg4) lysophosphatidic acid receptor knockout mice: signaling deficits without obvious phenotypic abnormality attributable to lpa2,” Molecular and Cellular Biology, vol. 22, no. 19, pp. 6921–6929, 2002.View at: Publisher Site | Google Scholar
I. Ishii, J. J. Contos, N. Fukushima, and J. Chun, “Functional comparisons of the lysophosphatidic acid receptors, LPA1/VZG-1/EDG-2, LPA2/EDG-4, and LPA3/EDG-7 in neuronal cell lines using a retrovirus expression system,” Molecular Pharmacology, vol. 58, no. 5, pp. 895–902, 2000.View at: Publisher Site | Google Scholar
K. Kotarsky, Å. Boketoft, J. Bristulf et al., “Lysophosphatidic acid binds to and activates GPR92, a G protein-coupled receptor highly expressed in gastrointestinal lymphocytes,” Journal of Pharmacology and Experimental Therapeutics, vol. 318, no. 2, pp. 619–628, 2006.View at: Publisher Site | Google Scholar
M. Lee, S. Choi, G. Halldén, S. J. Yo, D. Schichnes, and G. W. Aponte, “P2Y5 is a Gαi, Gα12/13 G protein-coupled receptor activated by lysophosphatidic acid that reduces intestinal cell adhesion,” American Journal of Physiology - Gastrointestinal and Liver Physiology, vol. 297, no. 4, pp. G641–G654, 2009.View at: Publisher Site | Google Scholar
K. Harper, D. Arsenault, S. Boulay-Jean, A. Lauzier, F. Lucien, and C. M. Dubois, “Autotaxin promotes cancer invasion via the lysophosphatidic acid receptor 4: participation of the cyclic AMP/EPAC/Rac1 signaling pathway in invadopodia formation,” Cancer Research, vol. 70, no. 11, pp. 4634–4643, 2010.View at: Publisher Site | Google Scholar
H. Barash, E. R. Gross, Y. Edrei et al., “Accelerated carcinogenesis following liver regeneration is associated with chronic inflammation-induced double-strand DNA breaks,” Proceedings of the National Academy of Sciences of the United States of America, vol. 107, no. 5, pp. 2207–2212, 2010.View at: Publisher Site | Google Scholar
A. M. Al Murri, J. M. S. Bartlett, P. A. Canney, J. C. Doughty, C. Wilson, and D. C. McMillan, “Evaluation of an inflammation-based prognostic score (GPS) in patients with metastatic breast cancer,” British Journal of Cancer, vol. 94, no. 2, pp. 227–230, 2006.View at: Publisher Site | Google Scholar
D. N. Brindley, M. G. K. Benesch, and M. M. Murph, “Autotaxin–an enzymatic augmenter of malignant progression linked to inflammation,” in Melanoma - Current Clinical Management and Future Therapeutics, M. Murph, Ed., p. 12, InTech, Rijeka, 2015.View at: Google Scholar
Y. Zhao, D. He, B. Saatian et al., “Regulation of lysophosphatidic acid-induced epidermal growth factor receptor transactivation and interleukin-8 secretion in human bronchial epithelial cells by protein kinase Cδ, Lyn kinase, and matrix metalloproteinases,” The Journal of Biological Chemistry, vol. 281, no. 28, pp. 19501–19511, 2006.View at: Publisher Site | Google Scholar
Y. Li, H. Du, Y. Qin, J. Roberts, O. W. Cummings, and C. Yan, “Activation of the signal transducers and activators of the transcription 3 pathway in alveolar epithelial cells induces inflammation and adenocarcinomas in mouse lung,” Cancer Research, vol. 67, no. 18, pp. 8494–8503, 2007.View at: Publisher Site | Google Scholar
S.-Y. Zhang, W. Shi, P. Cheng, and M. J. Zaworotko, “A mixed-crystal lanthanide zeolite-like metal–organic framework as a fluorescent indicator for lysophosphatidic acid, a cancer biomarker,” Journal of the American Chemical Society, vol. 137, no. 38, pp. 12203–12206, 2015.View at: Publisher Site | Google Scholar
V. C. Estrella, A. M. Eder, S. Liu et al., “Lysophosphatidic acid induction of urokinase plasminogen activator secretion requires activation of the p38MAPK pathway,” International Journal of Oncology, vol. 31, no. 2, pp. 441–449, 2007.View at: Google Scholar
A. Mazzocca, F. Dituri, L. Lupo, M. Quaranta, S. Antonaci, and G. Giannelli, “Tumor-secreted lysophostatidic acid accelerates hepatocellular carcinoma progression by promoting differentiation of peritumoral fibroblasts in myofibroblasts,” Hepatology, vol. 54, no. 3, pp. 920–930, 2011.View at: Publisher Site | Google Scholar
Y. Liao, G. Mu, L. Zhang, W. Zhou, J. Zhang, and H. Yu, “Lysophosphatidic acid stimulates activation of focal adhesion kinase and paxillin and promotes cell motility, via LPA1–3, in human pancreatic cancer,” Digestive Diseases and Sciences, vol. 58, no. 12, pp. 3524–3533, 2013.View at: Publisher Site | Google Scholar
Y. Kishi, S. Okudaira, M. Tanaka et al., “Autotaxin is overexpressed in glioblastoma multiforme and contributes to cell motility of glioblastoma by converting lysophosphatidylcholine to lysophosphatidic acid,” The Journal of Biological Chemistry, vol. 281, no. 25, pp. 17492–17500, 2006.View at: Publisher Site | Google Scholar
N. Samadi, R. Bekele, D. Capatos, G. Venkatraman, M. Sariahmetoglu, and D. N. Brindley, “Regulation of lysophosphatidate signaling by autotaxin and lipid phosphate phosphatases with respect to tumor progression, angiogenesis, metastasis and chemo-resistance,” Biochimie, vol. 93, no. 1, pp. 61–70, 2011.View at: Publisher Site | Google Scholar
J. L. Tanyi, Y. Hasegawa, R. Lapushin et al., “Role of decreased levels of lipid phosphate phosphatase-1 in accumulation of lysophosphatidic acid in ovarian cancer,” Clinical Cancer Research, vol. 9, no. 10, Part 1, pp. 3534–3545, 2003.View at: Google Scholar
G. Venkatraman, M. G. K. Benesch, X. Tang, J. Dewald, T. P. W. McMullen, and D. N. Brindley, “Lysophosphatidate signaling stabilizes Nrf2 and increases the expression of genes involved in drug resistance and oxidative stress responses: implications for cancer treatment,” The FASEB Journal, vol. 29, no. 3, pp. 772–785, 2015.View at: Publisher Site | Google Scholar
N. Desroy, C. Housseman, X. Bock et al., “Discovery of 2-[[2-Ethyl-6-[4-[2-(3-hydroxyazetidin-1-yl)-2-oxoethyl]piperazin-1-yl]-8-methylimidazo[1,2-a]pyridin-3-yl]methylamino]-4-(4-fluorophenyl)thiazole-5-carbonitrile (GLPG1690), a first-in-class Autotaxin inhibitor undergoing clinical evaluation,” Journal of Medicinal Chemistry, vol. 60, no. 9, pp. 3580–3590, 2017.View at: Publisher Site | Google Scholar
Study to Assess Safety, Tolerability, Pharmacokinetic and Pharmacodynamic Properties of GLPG1690, 2017, https://clinicaltrials.gov/ct2/show/NCT02738801?term=GLPG1690&rank=2.
Safety and Efficacy of a Lysophosphatidic Acid Receptor Antagonist in Idiopathic Pulmonary Fibrosis, https://clinicaltrials.gov/ct2/show/record/NCT01766817.
Proof of Biological Activity of SAR100842 in Systemic Sclerosis, 2016, https://clinicaltrials.gov/ct2/show/record/NCT01651143?term=SAR100842&rank=1.
N. Euer, M. Schwirzke, V. Evtimova et al., “Identification of genes associated with metastasis of mammary carcinoma in metastatic versus non-metastatic cell lines,” Anticancer Research, vol. 22, no. 2A, pp. 733–740, 2002.View at: Google Scholar