Participation of reactive oxygen species in the lysophosphatidic acid-stimulated mitogen-activated protein kinase kinase activation pathway.

Recent evidence suggests that reactive oxygen species (ROS) may function as second messengers in intracellular signal transduction pathways. We explored the possibility that ROS were involved in lysophosphatidic acid (LPA)-induced mitogen-activated protein (MAP) kinase signaling pathway in HeLa cells. Antioxidant N-acetylcysteine inhibited the LPA-stimulated MAP kinase kinase activity. Direct exposure of HeLa cells to hydrogen peroxide resulted in a concentration- and time-dependent activation of MAP kinase kinase. Inhibition of catalase with aminotriazole enhanced the effect of LPA on induction of MAP kinase kinase. Further, LPA stimulated ROS production in HeLa cells. These findings suggest that ROS participate in the LPA-elicited MAP kinase signaling pathway.

hydrogen peroxide (H 2 O 2 ), and hydroxyl radical (OH ⅐ ), are potent microbicidal agents, but excess ROS can also cause oxidative damage to macromolecules of host cell (1). Previous studies have shown that elevated levels of ROS could trigger intracellular signaling transduction pathways that may mediate cellular protective responses (2)(3)(4). In addition to their roles in inflammatory and pathological processes, increasing evidence suggests that ROS may function as second messengers in cytokine (interleukin-1 and tumor necrosis factor ␣) and some growth factor signal transduction pathways that regulate transcription factors such as NF-B and AP-1 (5,6).
Lysophosphatidic acid (LPA) is released by activated platelets and is thought to be responsible for much of the activity in serum that promotes cell growth and adhesion (7,8). LPA elicits its biological responses through a putative receptor that is coupled to heterotrimeric G-proteins (9). Several proximal signaling events are known to be evoked by LPA, including phosphoinositide hydrolysis and Ca 2ϩ mobilization, release of arachidonic acid, inhibition of adenylate cyclase, and induction of protein tyrosine phosphorylation (10,11). It is likely that some of these signaling events cross-interact to induce synergistic responses.
LPA rapidly activates the mitogen-activated protein (MAP) kinase pathway (11)(12)(13)(14). MAP kinases are serine/threonineprotein kinases regulated by dual tyrosine and threonine phosphorylation. Three subfamilies of MAP kinases, MAPK, JNK, and HOG, have been cloned (for reviews, see Refs. [15][16][17]. The 42-kDa MAP kinase (p42 mapk ) also called extracellular signalregulated kinase 2 (ERK2)) and 44-kDa MAP kinase (p44 mapk , ERK1) are phosphorylated and activated by highly specific MAP kinase kinase 1 and MAP kinase kinase 2 (MKK1/2) (18). For simplicity, p42 mapk and p44 mapk will be referred to as MAP kinase in this report. MAP kinase has been shown to play a pivotal role in cell proliferation and differentiation (19). It has been shown that the LPA-induced MAP kinase activation is sensitive to pertussis toxin inhibition, indicating a critical role of a pertussis toxin-sensitive G i -protein (11,12,14). However, these data did not exclude the contribution of other signaling events to the LPA-induced MAP kinase activation. LPA triggers a biphasic arachidonic acid release in HeLa cells. 2 The first phase of the LPA-induced arachidonic acid release precedes the activation of MAP kinase kinase. 2 Arachidonic acid is known to give rise to ROS through its subsequent metabolism (20) and activation of NADPH oxidase (21), prompting us to investigate the possible involvement of ROS in the LPA-stimulated MAP kinase activation pathway. We present evidence here that demonstrates the involvement of ROS in the LPAinduced MAP kinase signaling pathway.
Cell Culture-HeLa cells were grown at 37°C in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum in an atmosphere of 7% CO 2 . Confluent cells were serum-starved for 18 -20 h in phenol red-free DMEM containing 25 mM Hepes and 0.1% BSA (fatty acid-and globulin-free). Serum-starved cells were washed twice with DMEM (phenol red-free, plus 25 mM Hepes) and incubated in DMEM (phenol red-free, plus 25 mM Hepes) for 90 min before use.
MAP Kinase Kinase Assay-Total activity of MAP kinase kinase 1 and MAP kinase kinase 2 (MKK1/2) was determined by the following procedure, modified from that described previously (23). After stimulation, cells were chilled on ice and washed with cold phosphate-buffered saline (PBS). Each 6-cm plate of cells was lysed in 0.6 ml of cold lysis buffer (10 mM Tris acetate, 100 mM NaCl, 1 mM dithiothreitol, 40 mM ␤-glycerophosphate, 1.5 mM EGTA, 0.5 mM EDTA, 25 mM NaF, 1 mM sodium pyrophosphate, 0.5 mM sodium orthovanade, 20 mM 4-nitrophenyl phosphate, 1% Triton X-100, 1 mM benzamidine, 2 g/ml aprotinin, 2 g/ml leupeptin, 1 g/ml pepstatin, and 0.1 mg/ml phenylmethylsulfonyl fluoride, pH 7.5 at 25°C). Cell lysates were centrifuged (4°C) at 16,000 ϫ g for 15 min. Protein concentrations of cell lysate supernatants were determined. Equal portions (2 g of protein) of cell lysate * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
supernatants were mixed with KR kinase assay mixture (20 mM Hepes, pH 7.5, 10 mM MgCl 2 , 1 mM dithiothreitol, 10 mM 4-nitrophenyl phosphate, 40 M [␥-32 P]ATP (5000 cpm/pmol), and 2.5 g of K52R). The reaction mixtures were incubated at 30°C for 12 min or as indicated in the figure legends. The kinase reaction was stopped by addition of SDS sample buffer and heat denaturation. Phosphorylation of K52R was analyzed by autoradiography and phosphoimaging after electrophoresis on 10% SDS-polyacrylamide gels.
Detection of Intracellular H 2 O 2 -Intracellular levels of H 2 O 2 were analyzed by fluorescence-activated cell sorting (FACS) using dihydrorhodamine 123 as a probe (6,24). Experiments were performed under dim light. Confluent, serum-deprived HeLa cells were incubated in DMEM (phenol red-free, plus 25 mM Hepes) containing 10 M dihydrorhodamine 123 for 20 min. Cells were treated with aminotriazole (50 mM), LPA (40 M), or BSA, then chilled on ice and washed with cold PBS. Washed cells were detached from culture plates by trypsin digestion. The activity of trypsin was quenched with 0.05% BSA (fatty acidand globulin-free) in PBS. Following PBS wash, cells were fixed in 1% paraformaldehyde. The fluorescence intensities of rhodamine 123 of 10,000 cells from each sample were analyzed by flow cytometry using a FACScan flow cytometer equipped with an air-cooled argon laser (Becton Dickinson).

RESULTS AND DISCUSSION
In all cell types examined, p42 mapk is specifically phosphorylated and activated by dual specificity MKK1 and MKK2 (MKK1/2). Total activity of MKK1/2 in the cells represents a valid measurement for the activation state of the MAP kinase pathway in the rapid activation phase (14,23). As illustrated in Fig. 1, LPA and EGF markedly stimulated the MKK1/2 activity in HeLa cells. To test for the possible involvement of ROS in MKK1/2 activation induced by LPA, we examined the effect of antioxidant N-acetylcysteine on MKK1/2 activation. N-Acetylcysteine directly scavenges ROS and also increases the intracellular levels of reduced glutathione (GSH). GSH is a hydroxyl radical scavenger and a substrate of glutathione peroxidase which degrades H 2 O 2 . N-Acetylcysteine has been used extensively to study the role of ROS in signaling pathways (6,(25)(26)(27). An inhibition by N-acetylcysteine can be taken as an indication of the involvement of ROS. The LPA-stimulated MKK1/2 activity was inhibited by 82 Ϯ 4% (average of two experiments Ϯ range) in cells pretreated with N-acetylcysteine (30 mM), suggesting that ROS are involved in the LPA-induced MKK1/2 activation (Fig. 1). A similar result was obtained in Rat-1 cells. A lesser, but statistically significant, attenuation (38 Ϯ 8% in two experiments) by N-acetylcysteine of the EGF-stimulated MKK1/2 activity was also observed (Fig. 1), but was not investigated further in the current study.
To verify that the inhibitory effect of N-acetylcysteine is attributable to its ability to scavenge ROS, we examined the effects of two other ROS scavengers, dimethyl sulfoxide and ascorbic acid, on the LPA-stimulated MKK1/2 activity. Di-methyl sulfoxide is an effective hydroxyl radical scavenger (28). Ascorbic acid blocks free radical chain reaction, but may also directly remove hydroxyl radical (29). HeLa cells were pretreated with ascorbic acid (100 M, 60 min) or dimethyl sulfoxide (4%, 20 min) and stimulated with LPA (10 M, 5 min). The LPA-stimulated MKK1/2 activity was inhibited by 88 Ϯ 7% by dimethyl sulfoxide and 38 Ϯ 1% by ascorbic acid.
If ROS are the signaling molecules that mediate the LPAinduced MKK1/2 activation, then an increase in intracellular concentrations of ROS would be expected to mimic the effect of LPA on MKK1/2 activation. H 2 O 2 is the product of superoxide dismutases and several oxidases in the cells. Thus, cells that produce superoxide would also generate H 2 1. Inhibition by N-acetylcysteine on MKK1/2  target, whereas the external added H 2 O 2 diffuses indiscriminately. Second, HeLa cells may contain relatively high catalase activity, and, thus, high concentrations of H 2 O 2 are required to offset the catalase activity. In fact, inhibition of catalase by preincubation of HeLa cells with catalase inhibitor aminotriazole (26) prior to the addition of H 2 O 2 resulted in a marked shift of the H 2 O 2 dose-response curve to the left ( Fig. 2A). However, even in the presence of aminotriazole, greater than 0.5 mM H 2 O 2 was still required to activate MKK1/2 to a similar extent as that induced by 20 M LPA. Finally, it is likely that one or more signaling events besides production of ROS are critical for the induction of MKK1/2 activation by LPA, and ROS may function as only one of the parallel signaling intermediates. Thus, although H 2 O 2 alone at low concentrations (Ͻ0.5 mM) has a marginal effect on MKK1/2 activation, H 2 O 2 and the derived radicals may have a greater effect in the presence of other LPA-induced signaling intermediates because of synergism.
To further confirm the involvement of ROS, we treated HeLa cells with or without the catalase inhibitor aminotriazole prior to LPA stimulation. If ROS participate in the LPA-stimulated MKK1/2 activation pathway, inhibition of catalase would potentially augment the response to LPA. A 4.5-fold increase in H 2 O 2 -induced MKK1/2 activity was observed when HeLa cells were pretreated with aminotriazole (50 mM, 60 min), demonstrating the effectiveness of the catalase inhibitor (Fig. 3, see also Fig. 2A). In cells pretreated with aminotriazole, the LPAstimulated MKK1/2 activity was 1.9-fold that of cells without aminotriazole pretreatment (36.3-and 18.8-fold above basal, respectively) (Fig. 3). Thus, decreasing the catalase activity effectively enhances the cellular response to LPA, indicating the involvement of ROS.
For ROS to fulfill the role of signaling intermediates for LPA, LPA must be able to induce the production of ROS. As described above, LPA rapidly liberates arachidonic acid in HeLa cells. Arachidonic acid is known to generate ROS. Tumor necrosis factor ␣ and interleukin-1, both of which are known to utilize ROS as signaling intermediates, also stimulate the release of arachidonic acid (31). However, other routes of ROS generation are not excluded. We measured the relative concentrations of H 2 O 2 in HeLa cells using dihydrorhodamine 123 and fluorescence-activated cell sorting (FACS) (6,24). Dihydrorhodamine 123 is oxidized to membrane-impermeable, fluorescent rhodamine 123 in the presence of H 2 O 2 and possibly ROS derived from it (24). To minimize the loss of H 2 O 2 , aminotriazole was also added to the media. Incubation of HeLa cells with aminotriazole resulted in a time-dependent increase in fluorescence intensity (Fig. 4 and data not shown). A small increase of the fluorescence intensity induced by LPA was detectable at the earliest time (5 min) examined, but more consistent data were obtained if cells were stimulated for 10 min. In two duplicated experiments, cells treated for 10 min with aminotriazole (50 mM) plus BSA had an average 25% increase in mean fluorescence intensity of rhodamine 123 compared with BSA-treated cells (Fig. 4). An additional 22% increase in mean fluorescence intensity was detected in cells treated for 10 min with LPA (30 M) plus aminotriazole (50 mM) (Fig. 4). Thus, LPA is capable of generating ROS in HeLa cells.
In summary, data presented in this study show that ROS are involved in the LPA-induced MAP kinase kinase activation and LPA can stimulate the production of ROS in HeLa cells. Additional experiments using a p42 mapk immune complex kinase assay (14) showed that N-acetylcysteine partially inhibited the LPA-stimulated p42 mapk activation, H 2 O 2 stimulated p42 mapk activity in HeLa and NIH 3T3 cells, and aminotriazole enhanced the effect of LPA on p42 mapk activation (data not shown). Previous studies have shown that ROS are involved in the tumor necrosis factor ␣-stimulated NF-B activity (5) and the basic fibroblast growth factor-induced c-fos expression (6). Other data that we have obtained showed that N-acetylcysteine also inhibited the LPA-stimulated NF-B and AP-1 DNA binding activities in HeLa cells. 3 Thus, ROS appear to function as signaling intermediates of LPA and mediate a branch of the LPA signaling pathways. Our findings lend support to the emerging concept that ROS can function as physiological signaling intermediates. It will be interesting to examine whether ROS also participate in the signaling pathways of other phospholipids, such as platelet-activating factor (32) and sphingosine 1-phosphate (33). Clearly, further investigation of the mechanisms by which LPA increases the intracellular levels of ROS and ROS relay the cellular regulatory signals is warranted.