|
|
||||||||
Minireview |
a Institute for Hormone and Fertility Research, University of Hamburg, D-22529 Hamburg, Germany
| ABSTRACT |
|---|
|
|
|---|
growth factors, lysolipids, ovary, signal transducers, signal transduction, testis
| HISTORICAL PERSPECTIVE |
|---|
|
|
|---|
| BIOCHEMISTRY OF LPA |
|---|
|
|
|---|
Although serum LPA is bound to albumin fractions, its biological effects are not critically dependent on the presence of albumin (for a detailed review see [2]). Serum contains 220 µM LPA and several other as yet unidentified lipids with LPA-like activity [2]. The major serum LPA band consists of a mixture of LPA species with unsaturated and saturated fatty acid chains 18:1 and 16:0 followed by the 18:2, 18:0, and 20:4 LPA [4]. The majority of LPA species in ascites from ovarian cancer patients [5] are 16:0 and 18:0, although polyunsaturated 18:2, 20:4, and 22:6 chains are also abundant [4]. LPA is more water soluble than long-chain phospholipids because of its free hydroxyl and phosphate moiety, (i.e., 1-oleoyl-sn-glycero-3-phosphate) [6]. Relatively little is known about the physical chemistry of LPA, such as phase behavior and its cation binding properties [2].
| PRODUCTION AND RELEASE OF LPA |
|---|
|
|
|---|
|
|
A so-called ovarian cancer activating factor, a lipid factor found in ascites from ovarian cancers in earlier studies, has now been identified [10] as a complex of several molecular species of LPA. Elevated serum LPA levels are found in more than 90% of ovarian cancer patients at stage I or stage II [5, 11, 12]. Rough estimates suggest that LPA levels in malignant effusions are in the high micromolar range, about 10-fold higher than normally found in serum [13, 14]. Taking this into account, Xu et al. [11] suggested that the level of LPA in plasma may represent a potential biomarker for ovarian cancer. Although the precise origin of LPA in plasma and ascites fluid remains to be elucidated, it seems likely that LPA is released by the tumor cells themselves. Metabolic pathways responsible for LPA synthesis are poorly delineated though some information is available on LPA production in ovarian cancer epithelium. However, more detailed information on the biosynthetic pathways has arisen from studies employing nonreproductive tissues. In thrombin-activated blood platelets, for example, phospholipase C, phospholipase A2 (PLA2), or phospholipase A1 may contribute to LPA production [8, 15]. Based on a literature survey, Gaits et al. [8] suggested a key role for the secretory nonpancreatic phospholipase A2 type II (sPLA2 II). This enzyme accumulates in inflammatory fluid and is produced by a number of cell types in response to cytokines. Although sPLA2 II is the first reported example of a phospholipase capable of generating LPA directly from phosphatidic acid [8, 15, 16], sPLA2 II hydrolyzes phospholipid substrates only in cell membranes that have lost their phospholipid asymmetry [15]. This loss may occur under physiological or pathological conditions and results in a translocation of phospholipids (such as phosphatidic acid) from the internal leaflet to the outer leaflet of the cell membrane.
LPA production by sPLA2 II in the LPA autoinduced ovarian cancer cell line OVCAR3 may not be seen in other ovarian carcinoma cells, which may require different stimuli [5]. For example, a significant stimulation of LPA release is observed following the treatment of ovarian and cervical cancer cells with phorbol ester [17]. Although in OVCAR3 cells LPA alone can induce LPA secretion [5, 18], cell lines such as HEY and OCC1 additionally require phorbol ester, whereas IOSE cells do not need LPA autoinduction at all [5]. An inhibitor of the cytosolic PLA2 (cPLA2) could block 75% of phorbol ester-induced LPA secretion in ovarian cancer cells, whereas autoinduction by LPA appears not to be dependent on cPLA2 [5]. Although the experiments performed with PLA2 inhibitors may predict a contribution by PLA2 to LPA secretion, direct evidence for the involvement of this enzyme has been elusive. There is no information about the expression of this enzyme in ovarian cancer tissues.
Using ovarian dispersates from healthy immature rats, Kol et al. [19] observed an induction of sPLA2 enzyme and increase in sPLA2 gene expression in response to interleukin 1 beta [19]. Although not mentioned by the authors, LPA secretion probably occurred under these experimental conditions.
Both basal and autoinduced LPA formation in ovarian cancer cells appears to require phospholipase D activity [5, 18]. Although there is copious production of LPA in ovarian cancer, phospholipase D-mediated phosphatidic acid formation in malignant ovary has not been described. However, enhanced phosphatidic acid formation in response to growth factor stimulation of ovarian luteal cells has been reported [20]. In 1986, Tokumura and his coworkers [21] described a phospholipase D responsible for the degradation of lyso derivatives of platelet activating factor to alkyl LPA. Later, the same group [9, 22] characterized a metal ion-dependent lysophospholipase D (LPLD) involved in the accumulation of LPA (both in rat plasma and in the follicular fluid collected from women undergoing treatment for in vitro fertilization [IVF]). The LPLD activity in the serum of IVF patients appears to be about 2.5-fold higher than that in unstimulated women [9].
A marked increase in serum LPLD [22] is correlated with the progress of pregnancy and immunoreactive PLA2 II increases in serum, at both preterm and term [23]. Because naturally occurring PLA2 inhibitors, annexins, decrease during the onset of labor both in amnion and chorion [24], a role for PLA2 is also possible. The expression of secretory phospholipases in gestational tissues and their contribution to LPA production is yet to be established. It is likely that several different phospholipases (and their activators or inhibitors) may work in concert.
In male reproductive tissues, there is evidence for the expression of PLA2 enzymes in various spermatogenic cells (for review see [25]). Partial purification revealed the existence of a 14- to 16-kDa sperm PLA2 showing some similarities with sPLA2 I and sPLA2 II. PLA2 activities have been reported in bull prostate, seminal vesicle, Cowper gland, and seminal plasma from both bulls and humans [26, 27]. There is compelling evidence that PLA2 plays a role in the release of fatty acids and lysophospholipids involved in sperm membrane fusion during acrosomal exocytosis [28, 29]. Recently, a new phosphatidic acid-preferring PLA1 enzyme has been purified from bovine testicular tissues [30]. This tetramer-forming phospholipase shares some properties with a 58-kDa brain PLA2, which shows an absolute preference for phosphatidic acid [31].
It is not known whether the tetrameric structure of these phospholipases is related to a functional regulatory mechanism. The family of PLA2 enzymes can be classified into several groups: high (85 kDa) and low (14 kDa) molecular mass cPLA2 enzymes and at least three so-called sPLA2 enzymes: sPLA2 I, which is produced by the pancreatic acinar cells, sPLA2 II, and sPLA2 V, which is present in several human tissues. Thus, five distinct mammalian sPLA2s and two main types of sPLA2 receptors have so far been identified (for review see [32]). The PLA2 enzymes have also been recognized as autocrine mediators [32].
Currently, none of the PLA1 or PLA2 enzymes can be considered as single direct effectors of LPA production; the activity of other phospholipases appears also to be relevant for the production of this biologically active phospholipid. In principle, several of the phospholipases present in various reproductive tissues should be able to generate LPA, although confirmatory evidence is still lacking. Information concerning LPA concentration in tissue and the localization of the physiological substrate remains unavailable. The most significant information on LPA production at the tissue level has been obtained from studies of ovarian cancer, where sPlA2 II appears to be a major contributor. However, under more physiological conditions, LPLD probably is the major player. Further studies are necessary to resolve this issue.
| LPA RECEPTORS |
|---|
|
|
|---|
Although no LPA receptor protein has been purified and biochemically characterized so far, the comparative sequence and genomic structure analysis has revealed that the receptor group edg (coding orphan receptors for endothelial differentiation genes) may include such an LPA receptor [2, 36]. One such cDNA encodes a GPCR, initially designated vzg-1 and renamed edg-2 based on its high homology to the human orphan GPCR edg-1 [37]. Overexpression of the human edg-2 protein in Chinese hamster ovary cells increased specific binding of [3H]LPA [37, 38]. A sequence-based search for other GPCRs specific for LPA resulted in the discovery of genes edg-1 to edg-5 [36]. These GPCRs belong to a large edg gene superfamily encoding receptors for LPA and other lysolipids [36, 37]. Chun et al. [37] proposed a provisional nomenclature for this receptor family (Fig. 3): the LpA1/A2/A3 group (edg-2, -4, and -7) includes high-affinity LPA receptors, and the LpB1/B2 group includes receptors for sphingosine-1-phosphate [37, 38]. Analysis of edg receptors by semiquantitative reverse transcription polymerase chain reaction (RT-PCR) and Western blotting using monoclonal antibodies shows prominent expression of the edg-4 (LpA2) receptor in ovarian cancer cells as compared with nonmalignant ovarian surface epithelium [39]. In contrast, levels of edg-2, -3, and -5 appear to be higher in nonmalignant epithelium [39]. The expression of edg-2 (LpA1) is especially high in cisplatin-resistant and slowly proliferating cell lines and is almost absent from cisplatin-sensitive and rapidly proliferating ovarian cancer cell lines [40]. Because the overexpression of edg-2 in A2780 ovarian cancer cells results in a reduced cell proliferation rate without altering the cisplatin sensitivity, the Furui et al. [40] suggested that edg-2 may be a negative regulator of ovarian epithelial cell growth and metastasis.
|
Contos and Chun [41] looked at the regulation and genetics of the edg-4 receptor, the prominently expressed subtype in ovarian cancer cells. These authors analyzed several splice variants of the edg-4 receptor and determined that one such variant arises from a frame shift mutation near the 3' end of the coding region [41]. This G-deletion variant is derived from an ovarian tumor. In addition to increasing our understanding of the mechanisms underlying LPA signaling and lysophospholipid receptor gene evolution, these results have important implications regarding the genomic targeting and oncogenic potential of LpA2. A newly identified member of the LPA receptor family, edg-7 (LpA3) [42], has properties distinct from the known cloned LPA receptors and is prominently expressed in prostate [43].
To address the possibility of a isoform-specific signaling mechanism, receptor subtype-specific antagonists are necessary. In addition to the pharmacological approaches, the molecular biological approach should bring more light in this issue. RT-PCR, RNase protection assays, and molecular cloning strategies can be used to identify and characterize various LPA receptor isoforms in reproductive tissues under physiological conditions, which will be an important area for future investigation. Little is known about the expression of various receptor isoforms in nonmalignant reproductive tissues. Furthermore, the potential physiological modulation of LPA receptor expression has not been reported thus far.
| BIOLOGICAL RESPONSES |
|---|
|
|
|---|
In addition to its mitogenic effects, LPA also produces dramatic effects on the actin cytoskeleton [44]. In particular, the cells of neuronal origin appear to show clearly the morphoregulatory effects of LPA (for review see [44]). LPA may also be one of the factors regulating tissue remodeling processes taking place during the growth, differentiation, and regression of the corpus luteum [50]. Luteal cells are known to be mainly regulated by LH (during the ovarian cycle) or chorionic gonadotropins (during pregnancy). The action of the gonadotropins appear to be under modulatory regulation of LPA in the ovary. Several important functions of the ovary, such as steroid production, can only be modulated by LPA at particular stages of the cycle (unpublished results).
Other parts of the female reproductive tract may also be affected by this locally produced bioactive lipid. Kunikata et al. [51] reported that the addition of LPA significantly enhanced ovum transport in the mouse oviduct. The outgrowth of embryos on decidual cells similarly increases in response to LPA [52].
In the male reproductive system, a local role for LPA may exist. Garbi et al. [53] demonstrated the presence of phospholipase D1 in bovine sperm, and it is known that exogenous phospholipase D is capable of generating LPA [9, 54]. At the time of fertilization, the spermatozoon undergoes a process of regulated exocytosis, the acrosomal reaction. Many agonists may initiate this reaction, and others have been reported as modulators or cofactors [25]. Some of these factors are present in oviductal fluid. Earlier studies suggested that phospholipases and unidentified lysolipids are involved in sperm membrane fusion during acrosomal exocytosis [25]. Recent studies have revealed that LPA can activate the acrosomal reaction [53], suggesting that LPA, which is both locally produced and biologically active, could be the unidentified lysolipid.
Most reproductive tissues can be considered possible LPA targets. The biological effects of LPA may differ among the various organs and may depend on the particular physiological status of the target tissue. In the ovary, the effects of LPA may vary depending upon the various stages of the cycle. LPA may also have different functions within a given tissue. Although the available evidence is preliminary, the published data suggest that LPA may play an important role in various processes of the oviduct such as ovum transport and facilitation of the acrosomal reaction.
| LPA-INDUCED SIGNAL TRANSDUCTION |
|---|
|
|
|---|
In studies performed with cell lines expressing a single LpA1 receptor, this receptor activated distinct signaling pathways, i.e., the phospholipase C/MAP kinase pathway, arachidonic acid release, and the inhibition of adenylyl cyclase [56]. Heterologous expression of LpA2 in the same cell system produces LPA-dependent cell rounding in addition to the other effects [55, 56]. The stimulation of the LPA-induced proliferation of ovarian cancer cells is correlated with an increase in [Ca2+]i and the activation of p125 FAK and MAP kinases [57]. Whether these pathways are responsible for the mitogenic effects of LPA remains to be determined. Ovarian cancer cells show high expression levels of LpA2 receptors (edg-4) but low expression of LpA1 receptors [56]. This difference may indicate that the LpA2-induced (rather than LpA1-induced) activation of the MAP kinase pathway contributes to the transmission of LPA-induced proliferative signals in ovarian carcinoma cells. A similar activation of the MAP kinases Erk1/2 is also observed in prostate cancer cells in response to LPA [58]. However, because the type of LPA receptor expressed in this prostate cancer cell line is unknown, these results are difficult to compare with those observed in the ovarian cells. Results obtained by others [43] have indicated that prostate tissues express a novel LPA receptor form, edg-7. It is not known whether the observed mitogenic effect of LPA in prostate cancer cells is a consequence of an activation of the edg-7 receptor.
LPA receptors can couple to several classes of heterotrimeric G proteins, i.e., G12/13, Gi/o, Gq, and probably other G proteins except GS [44, 56]. In the case of PC-3 prostate cancer cells [58], pretreatment with pertussis toxin inhibits the mitogenic response by 70%80% indicating coupling to Gi/o. The expression of the G beta-gamma sequestrant peptides results in an inhibition of LPA-mediated Erk1/2 activation [58]. LPA-induced mitogenic stimulation in the prostate requires the participation of epidermal growth factor-mediated responses and c-Src protein kinases. In tissues outside the reproductive system, crosstalk to growth factor receptor kinase, a novel c-Src-effector kinase, is involved in the LPA-induced signal transduction cascade [2].
Morphological responses to LPA are often manifested through its dramatic effects on the actin-based cytoskeleton [2, 36, 44]. In nonreproductive tissues such as motile fibroblasts, a depletion of LPA triggers a loss of detyrosinated microtubules, whereas the addition of LPA to astrocytes inhibits the cAMP-dependent stellation process [59, 60]. In cultured luteal cells, LPA influences the actin cytoskeleton by modulating the morphological changes induced by LH [50]. These morphoregulatory effects of LPA could be mimicked by CNF1, a bacterial toxin known to activate small G-proteins from the Rho family. However, botulinum-C3-exotransferase, which acts mainly through the inhibition of RhoA, mimics the effects of LH. Therefore, we suggested that the morphomodulatory effects of LPA in the ovary are mediated primarily through the activation of RhoGTPases [50]. In astrocytes [60] and melanoma cells [61], the overexpression of RhoA protein leads to an inhibition of the stellation process induced by either cAMP or C3-exotransferase. Although these mechanisms appear to be physiologically far removed from the developing corpus luteum, the underlying biochemical pathways may not be too different.
A pathway involving Rho kinase has also been recently implicated in LPA- and oxytocin-induced uterine contractions [62]. Stimulation of cultured myometrial cells with LPA plus oxytocin resulted in an increase in stress fiber formation, a process that contributes to the initiation and maintenance of smooth muscle contraction [62]. Although the molecular mechanism of the LPA-induced stress fiber formation differs from that involved in stellation, both processes converge at the level of RhoGTPases. Further downstream of the RhoGTPases, LPA activates a series of kinases, including various protein kinase C (PKC) isoforms. LPA-induced activation of sperm PKC
has been implicated in the sperm acrosomal reaction [53], and the activation of protein kinase C
by LPA has been reported in ovarian theca lutein cells [63].
The substrates for the PKC-induced phosphorylation in sperm have not been identified. PKC
in the sperm coimmunoprecipitates with phospholipase D1 in cells stimulated with either LPA or phorbol ester [53]. Although the function of phospholipase D1 in the PKC signal transduction is as yet unknown, it is tempting to speculate that intracellular production of phosphatidic acid may contribute to LPA receptor-mediated effects on the sperm.
Although these results add another piece to the LPA signal transduction puzzle, a clear picture is yet to emerge. We are still in the process of learning how bioactive lipids transmit signals within a single cell, and we are only beginning to understand the heterogeneity of the LPA receptors, the variability of G-protein coupling, and the multitude of downstream effector proteins (for review see [44, 56]. Although the vast majority of LPA effects appear to be receptor mediated, the relative importance of non-receptor-mediated mechanisms for the action of LPA remains to be elucidated. Because LPA can signal without recourse to its receptor, LPA may use both classical and nonclassical pathways to mediate the signal.
| LPA DEGRADATION |
|---|
|
|
|---|
Lipids such as LPA cannot move rapidly in an uncontrolled manner within the plasma membrane unless they are helped by an enzyme such as flipase, scramblase, or translocase. Therefore, a possible colocalization (intra- or extracellular) of both the lipid substrate and these enzymes must be considered [64]. The LPA-metabolizing enzymes are potentially important regulators of LPA signaling. In particular, the ecto-lipid-phosphatase (acting extracellularly) has been implicated as a possible negative regulator of LPA-induced mitogenic signals [67]. The LPA-degrading enzymes (LPA phospholipases, LPA phosphatases) are probably also actively involved in modulation of LPA signaling. These enzymes may influence the duration of the lipid signal and may also generate new potential signaling molecules, such as phoshatidic acid, diacylglycerol, or monoacylglycerol.
| CONCLUSION AND FUTURE PERSPECTIVES |
|---|
|
|
|---|
The mode of LPA action is a function of its local concentration, the differentiation status of the target cell, and its receptor status. Each of these aspects must be studied in detail because LPA may play more than one role within a particular tissue. Elucidation of these roles will significantly advances our understanding of the biological and pathological effects of this fascinating lipid.
Of special interest is the molecular mechanism responsible for the functional transition of a potential survival factor into an aggressive carcinogen. This identification of this mechanism will be both challenging and rewarding, will extend our knowledge of the biology of LPA, and will lead to identification of new and clinically useful pathological markers and development of possible therapeutic applications. Studies designed to determine the synthesis and degradation pathways and the mechanism of LPA action have already begun in many laboratories, and those being planned currently will soon throw more light on the physiological and pathological actions of this fascinating small phospholipid in reproductive tissues.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Correspondence: L.T. Budnik, Institute for Hormone and Fertility Research, University of Hamburg, Grandweg 64, D-22529 Hamburg, Germany. FAX: 4940 56190864; budnik{at}ihf.de ![]()
Accepted: October 3, 2001.
Received: July 26, 2001.
| REFERENCES |
|---|
|
|
|---|
is redistributed to the nucleus in lysohosphatidic acid stimulated theca lutein cells. Biol Reprod 2001; 64(suppl):255 This article has been cited by other articles:
![]() |
S.-U. Chen, H. Lee, D.-Y. Chang, C.-H. Chou, C.-Y. Chang, K.-H. Chao, C.-W. Lin, and Y.-S. Yang Lysophosphatidic Acid Mediates Interleukin-8 Expression in Human Endometrial Stromal Cells through Its Receptor and Nuclear Factor-{kappa}B-Dependent Pathway: A Possible Role in Angiogenesis of Endometrium and Placenta Endocrinology, November 1, 2008; 149(11): 5888 - 5896. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Ye Lysophospholipid signaling in the function and pathology of the reproductive system Hum. Reprod. Update, September 1, 2008; 14(5): 519 - 536. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Ye, M. K. Skinner, G. Kennedy, and J. Chun Age-Dependent Loss of Sperm Production in Mice via Impaired Lysophosphatidic Acid Signaling Biol Reprod, August 1, 2008; 79(2): 328 - 336. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Zhao, D. He, J. Zhao, L. Wang, A. R. Leff, E. Wm. Spannhake, S. Georas, and V. Natarajan Lysophosphatidic Acid Induces Interleukin-13 (IL-13) Receptor {alpha}2 Expression and Inhibits IL-13 Signaling in Primary Human Bronchial Epithelial Cells J. Biol. Chem., April 6, 2007; 282(14): 10172 - 10179. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. T. Budnik and B. Brunswig-Spickenheier Differential effects of lysolipids on steroid synthesis in cells expressing endogenous LPA2 receptor J. Lipid Res., May 1, 2005; 46(5): 930 - 941. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. T. Budnik, B. Brunswig-Spickenheier, and A. K. Mukhopadhyay Lysophosphatidic Acid Signals through Mitogen-Activated Protein Kinase-Extracellular Signal Regulated Kinase in Ovarian Theca Cells Expressing the LPA1/edg2-Receptor: Involvement of a Nonclassical Pathway? Mol. Endocrinol., August 1, 2003; 17(8): 1593 - 1606. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |