Leukotriene D₄ activates MAPK through a Ras-independent but PKCϵ-dependent pathway in intestinal epithelial cells

The powerful pro-inflammatory mediator LTD₄ has been implicated in the pathophysiology of several inflammatory disorders, particularly asthma and inflammatory bowel diseases(Samuelsson, 1983;Samuelsson, 2000;Horwitz et al., 1998). LTD₄ mediates its effects through the CysLT₁ receptor,which has been cloned and characterized as a seven transmembrane-spanning receptor (Lynch et al., 1999;Sarau et al., 1999). These findings confirm earlier observations made by our research group(Sjölander et al., 1990;Adolfsson et al., 1996) and by other investigators (Watanabe et al.,1990) showing that LTD₄ signaling occurs through heterotrimeric G-proteins. We have also found that the LTD₄-induced calcium signal is regulated by at least two different G-proteins in intestinal epithelial cells (Sjölander et al.,1990; Adolfsson et al.,1996; Grönroos et al.,1996). Moreover, we recently demonstrated that Gβγsubunits of heterotrimeric G-proteins function as `docking proteins' in the LTD₄-induced activation of PLC-γ1 in intestinal epithelial cells; we also demonstrated that activation of c-Src is essential for this interaction and signal (Thodeti et al.,2000).

Furthermore, ulcerative colitis is associated with an increased incidence of neoplastic transformation (Ekbom et al., 1990), and several studies have shown that colon cancer is under-represented in populations treated with non-steroidal anti-inflammatory drugs (Smalley and DuBois,1997). A possible link between inflammation and the occurrence of cancer has been suggested (Sheng et al.,1997). To determine whether LTD₄ is involved in the coupling between inflammatory bowel conditions and an increased risk of cancer, we have previously exposed non-transformed intestinal epithelial cells to LTD₄ for prolonged periods of time(Öhd et al., 2000). Such exposure caused an upregulation of the cancer-associated proteins COX-2 andβ-catenin, as well as the anti-apoptotic protein Bcl-2. Furthermore,LTD₄ also caused a PKC-dependent upregulation of active β1 integrins and an enhanced β1-integrin-dependent adhesion of intestinal epithelial cells (Massoumi and Sjölander, 2001). Taken together, these results suggest that LTD₄ signal a switch from cell death to cell survival.

MAPKs belong to a group of serine threonine kinases, and the MAPK family in mammalian cells includes extracellular signal-regulated kinase-1 and -2(Erk-1/2), the c-Jun NH2-terminal kinases (JNK) and p38 MAPK(Garrington and Johnson,1999). These MAPKs integrate multiple signals from various receptors and second messengers and are involved in the regulation of cellular proliferation and differentiation(Garrington and Johnson,1999). Once activated, a MAPK can translocate to the nucleus,where it presumably regulates the expression of different transcription factors (Garrington and Johnson,1999; Velarde et al.,1999). It has been shown that Erk-1/2 is activated by a variety of receptor tyrosine kinases and G-protein-coupled receptors. The mechanism underlying activation of Erk-1/2 seems to be highly receptor and cell specific(Daulhac et al., 1999), and,for many different types of receptors and cells, such activation is induced by a PKC- and/or a Ras-dependent signaling pathway(Hawes et al., 1995). It has been shown that different PKC isoforms stimulate Erk-1/2 along Ras-dependent or Ras-independent signaling pathways (Li et al., 1998), suggesting that PKC acts upstream of Ras and Raf-1(Miranti et al., 1999) or directly upstream of Raf-1 (Kolch et al.,1993; Cheng et al.,2001). Activation of the serine/threonine kinase Raf-1 is a complicated and not fully elucidated event that includes association with the active GTP-bound form of Ras at the membrane, Ras-dependent phosphorylation of Ser338 and most probably an additional phosphorylation of Tyr341(Mason et al., 1999). Different second messengers converge at Raf for subsequent downstream activation of MAPKs.

It has also been proposed that activation of Erk-1/2 by G-protein-coupled receptors occurs through transactivation of receptor tyrosine kinases(Daub et al., 1996;Rao et al., 1995). In Rat-1 fibroblasts and COS-7 cells, inhibition of EGF receptor function abrogates tyrosine phosphorylation of Shc and the subsequent activation of Erk-1/2 in response to LPA, endothelin-1 and thrombin, all of which bind to G-protein-coupled receptors (Daub et al.,1996; Daub et al.,1997). Such crosstalk between EGF and LTD4 receptors has been demonstrated in experiments showing that EGF-induced modulations of the cytoskeleton in fibroblasts are mediated by the CysLT₁ receptor(Peppelenbosch et al.,1995).

In light of our previous study of intestinal epithelial cells, showing that LTD₄ can reduce apoptosis and upregulate distinct proteins such as COX, we conducted the present investigation to determine if and how LTD₄ affects proliferation in these cells.

Phosphospecific antibodies to MAPK Erk-1/2 (p42/44) and MEK inhibitor PD98059 were purchased from New England BioLabs, Inc. (Beverly, MA); protein tyrosine kinase inhibitor genistein and antibodies to anti-MAPK were from Calbiochem (San Diego, CA). Isoform-specific PKC antibodies were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-EGF receptor antibodies and EGF were from Upstate Biotechnology Inc. (Lake Placid, NY); the activated form of EGF receptor antibody and the Ras antibody (RO2120) were from Transduction Laboratories (Lexington, KY). LTD₄ was purchased from Cayman Chemical Company (Ann Arbour, MI). ECL Western blot detection reagents and hyperfilm were from Amersham International (Buckinghamshire, UK). Ionomycin, MAPTAM, EGF-receptor inhibitor PD153035, PI-3Kinase inhibitors wortmannin, LY294002 and farnesyltransferase inhibitor FTI-277 were from Calbiochem (San Diego, CA). Pertussis toxin (PTX) was obtained from Speywood Pharma Ltd. (Maidenhead, UK). Peroxidase-linked goat anti-rabbit and mouse IgG originated from Dako A/S (Copenhagen, Denmark). Src family kinase inhibitor PP1 was from Alexis (San Diego, CA). All other chemicals were of analytical grade and obtained from Sigma Chemical Co. (St. Louis, MO).

Human embryonic intestinal epithelial cells [Intestine 407(Henle and Dienhardt, 1957)],which exhibit typical epithelial morphology and growth, were cultured as a monolayer to approximately 80% confluence for 5 days. Cell cultures were kept at 37°C in a humidified atmosphere of 5% CO₂ and 95% air in Eagle's basal medium supplemented with 15% new-born calf serum, 55 IU/ml penicillin and 55 μg/ml streptomycin. The cells were regularly tested to ensure the absence of mycoplasma contamination.

The cells were cultured on Nunclon (Nalge Nunc International, Denmark)96-well plates (5-10×10³ cells per well) for three days. They were subsequently pre-incubated in the absence or presence of the MEK inhibitor PD98059 (50 μM for 30 minutes), the PKC inhibitor GF109203X (30μM for 30 minutes) or the farnesyltransferase inhibitor FTI-277 [20 μM for 48 hours (Lerner et al.,1995)]. The cells were then allowed to grow in fresh media for another two days in the absence or presence of 80 nM LTD₄ or 100 ng/ml EGF and the above inhibitors. The control cells were allowed to grow in the absence of LTD₄, EGF and any inhibitor for the same period of time as the treated cells. The MTS assay (Promega, Madison, WI) was carried out according to the protocol provided by the manufacturer. Briefly, the cells were incubated in 20 μl of MTS/PMS solution for 2 hours, after which soluble formazan (reduced MTS tetrazolium) was measured at 490 nm in a BMG plate reader (Offenburg, Germany).

The cells were cultured for three days in 35×10 mm Petri dishes. The cells were subsequently pre-incubated with or without pertussis toxin, PTX(500 ng/ml for 2 hours) or FTI-277 (20 μM for 48 hours). Then the cells were allowed to grow in fresh media for 2 days in the absence or presence of 80 nM LTD₄ or 100 ng/ml EGF and the above inhibitors. The effects of 80 nM LTD₄ or 100 ng/ml EGF were also tested in cells transfected with N17 Ras or the empty vector. To determine the number of viable cells, all cell counts were performed in the presence of 0.2% trypan blue.

Cells were transfected for 24 or 48 hours with a full-length human HA-tagged N17 Ras construct (Odajima et al., 2000), GFP-tagged RD-PKCϵ, RD-PKCδ(Zeidman et al., 1999) or a W437 kinase-dead PKCϵ construct (K^-PKCϵ), which was generously provided by Arthur Mercurio (Beth Israel Deaconess Medical Center,Boston, MA, USA) or HA-tagged kinase inactive c-Raf construct(K^-Raf-1), generously provided by Larry Karnitz (Mayo Clinic,Rochester, MN, USA). Control cells were transfected with empty pEGFP-N1 vector from Clontech. Transient transfections of the cells were achieved using 3.5μl of Lipofectamine (Gibco) and 1.8 μg of plasmid DNA/ml and were performed in serum-free medium, essentially according to the protocol provided by the supplier.

Cells were serum-starved for 2 hours, pre-incubated with inhibitors for the indicated periods of time and stimulations were terminated by adding ice-cold lysis buffer (50 mM Tris [pH 7.5], 1 mM EDTA, 1 mM EGTA, 1 mM Na₃VO₄, 1% Triton X-100, 50 mM NaF, 5 mM sodium pyrophosphate, 10 mM sodium glycerophosphate, 4 μg/ml Leupeptin and 30μg/ml phenylmethanesulfonyl-fluoride, PMSF). Thereafter, the cells were kept on ice for 30 minutes in the lysis buffer, and the remaining cell debris was scraped loose into the buffer. The lysates were homogenized 10 times on ice in a glass tissue grinder (Dounce) and then centrifuged at 10,000 g for 15 minutes. The protein content of each supernatant was measured and compensated for prior to electrophoresis.

Cell stimulations were terminated by adding ice-cold buffer A containing 20 mM NaHepes (pH 8), 2 mM MgCl₂, 1 mM EDTA, 2 mM Na₃VO₄, 4 μg/ml leupeptin and 30 μg/ml PMSF. Thereafter, the cells were scraped loose into the cold buffer, homogenized 10 times on ice in a glass tissue grinder (Dounce) and then centrifuged at 200 g for 10 minutes. The protein content of the supernatant was measured and compensated for, and the supernatant was subsequently centrifuged at 1000 g for 5 minutes. The supernatant of the 1000 g fraction was further centrifuged at 200,000 g for 30 minutes. The resulting membrane-rich pellet was suspended in 150 μl of buffer A.

The cDNA clone encoding the GST fusion protein of the Raf minimal binding domain (RBD) of Ras in pGEX vector was transformed into Escherichia coli and cultured at 30°C(Hallberg et al., 1994). Expression of the GST fusion proteins was induced with 1 mM isopropyl-l-thio-D-galactopyranoside, and the E. coli were subsequently collected by centrifugation at 3500 g for 15 minutes followed by sonication in phosphate-buffered saline. Triton X-100 was added to the lysate (final concentration 1%), and particulate matter was removed by centrifuging at 5000 g for 15 minutes. The cleared lysate was incubated with glutathione-agarose beads (Sigma) for 1 hour at 4°C, and the beads were subsequently washed three times with ice-cold PBS. Lysates of unstimulated or stimulated Int 407 cells were prepared in 1.0 ml of the lysis buffer supplemented with 10 mM MgCl₂. GST fusion protein (5-10μg) or GST alone was pre-bound to agarose beads and incubated with 1.0 ml of one of the cell lysates (1 mg/ml total cell protein) for 2 hours at 4°C. Thereafter, the beads were washed once with ice-cold lysis buffer supplemented with 0.5 M NaCl and twice with buffer A.

Cell lysates, membrane fractions or precipitated proteins were solubilized by boiling at 100°C for 5 minutes in a sample buffer [62 mM Tris (pH 6.8),1.0% SDS, 10% glycerol, 15 mg/ml dithiothreitol, and 0.05% bromophenol blue]. The solubilized proteins were subjected to electrophoresis on 10-12%homogeneous polyacrylamide gels in the presence of SDS.

The separated proteins were electrophoretically transferred to a PVDF membrane. All membranes were blocked for 1 hour with 5% non-fat dried milk at room temperature and then incubated with a primary antibody for 1 hour at room temperature or overnight at 4°C. A 1:500 dilution was used for the anti-Ras (RO2120) antibody, whereas 1:1000 dilutions were used for all other antibodies. Subsequently all membranes were washed extensively and incubated with a horseradish-peroxidase-linked goat anti-rabbit, anti-sheep or anti-mouse antibody (1:5000) for 1 hour at room temperature. Thereafter, the membrane was again washed extensively, incubated with ECL western blot detection reagents and finally exposed to hyperfilm-ECL to visualize immunoreactive proteins. The phospho-MAPK blots were stripped and reprobed to detect total MAPK.

MEK-1, Raf-1 and B-Raf kinase were assayed using commercial kits from Upstate Biotechnology. The cells were first pre-incubated in the absence or presence of the MEK inhibitor PD98059 (50 μM for 30 minutes), the Src tyrosine kinase family inhibitor PP1 (10 μM for 15 minutes), PTX (500 ng/ml for 2 hours) or the PKC inhibitor GF109203X (30 μM for 30 minutes). Alternatively, the cells were depleted of PKC by incubation with 1 μM TPA(12-O-Tetradecanoylphorbol 13 acetate) for 24 hours. Thereafter, the cells were incubated in the absence or presence of 80 nM LTD₄ for 3 minutes and lysed. The cell lysates (1 mg aliquots) were then used for immunoprecipitation (1 hour at 4°C) with either 2 μg of an anti-MEK-1,2 μg of an anti-Raf-1 or 4 μg of an anti-B-Raf antibody. Thereafter, 30μl of a 3 mg/ml solution of protein G-agarose beads was added, and the mixture was allowed to stand for an additional hour at 4°C. The immunoprecipitates were washed twice with lysis buffer, after which kinase activities were measured with a coupled-enzyme assay. In short, the immune complexes were incubated for 30 minutes at 30°C with 10 μl of cold Mg-ATP buffer (Upstate Biotechnology) and the specified substrates (0.4 μg of inactive MEK-1 and 1 μg of inactive Erk-2 for Raf assays and MEK assays). This mixture (4 μl) was incubated for 10 minutes at 30°C with 10 μl of 2 mg/ml MBP substrate and 10 μl of a 1:10 dilution of[γ-³²P]ATP (1 mCi/100 μl) diluted with the cold Mg-ATP buffer. The reaction mixture (25 μl) was spotted onto the center of a P81 phosphocellulose paper square and washed thoroughly several times with 0.75%phosphoric acid and then once with acetone and thereafter subjected to liquid scintillation counting.

The cells were seeded onto glass coverslips and grown for 5 days, during the last 24 hours they were contransfected with N17 Ras and EGFP (empty vector). Thereafter, the cells were serum-starved for 2 hours and stimulated with 80 nM LTD₄ for 3 minutes or 100 ng/ml EGF for 5 minutes at 37°C. The stimulations were terminated by fixation of the cells for 10 minutes at room temperature in a 3.7% paraformaldehyde/PBS solution, after which the cells were permeabilised in a 0.5% Triton X-100/PBS solution for 5 minutes. The coverslips were subsequently washed twice in PBS and incubated at room temperature in a 3% BSA/PBS solution for 15 minutes. The cells were stained for 1 hour with a phospho-specific antibody against Erk-1/2. Thereafter, the coverslips were washed six times in PBS and incubated with a 1:200 dilution (in blocking buffer) of Alexa Fluor 568 goat anti-rabbit secondary antibody. The coverslips were finally washed six times in PBS and mounted in fluorescent mounting medium (DAKO A/S). Samples were examined and photographed in a Nikon Eclipse 800 microscope, using a 60× objective. Images were recorded with a scientific-grade, charge-coupled device (CCD)camera (Hamamatsu, Japan) and subsequently analysed with HazeBuster deconvolution software (VayTek, Inc., Fairland, CT, USA).

LTD₄-induced proliferation in Int 407 intestinal epithelial cells (Fig. 1), as determined by the MTT assay (Cory et al.,1991) and cell counting. The former is based on the following:tetrazolium salts are reduced to formazan compounds by dehydrogenase enzymes in metabolically active cells. The LTD₄-mediated increase in cell proliferation was effectively abolished by PTX (500 ng/ml;Fig. 1B), the PKC inhibitor GF109203X (30 μM; Fig. 1A),the MEK inhibitor PD98059 (50 μM; Fig. 1A), and the Ras farnesyltransferase inhibitor FTI-277 (20 μM;Fig. 1). Transfecting the cells with N17 Ras also blocked this proliferative response(Fig. 1B). As a comparison we show that the EGF-induced proliferative response (striped bars) was also effectively blocked by Ras inhibition (Fig. 1), supporting an active role for Ras in the proliferative response in Int 407 cells. However, EGF induced proliferation was unaffected by PTX treatment (Fig. 1B). The control cells (white bars) were allowed to grow in the absence of LTD₄, EGF and any inhibitor for the same period of time as the treated cells.

We observed a significant activation of Erk-1/2 in cells stimulated with LTD₄ (Fig. 2),although this was not as pronounced as the response induced by EGF(Fig. 3C). A concentration of 0.8 nM LTD₄ was sufficient to induce activation of Erk-1/2, and the response at that level was half of that noted at 80 nM LTD₄(Fig. 2A). We refrained from using higher and non-physiological concentrations. The specific LTD₄ receptor CysLT₁ antagonist ZM198,615 (ICI-198,615,50 μM for 15 minutes) abolished Erk-1/2 activation induced by 80 nM LTD₄, indicating that the effect is mediated by the CysLT₁ receptor (Fig. 2A). The response to LTD₄ was rapid, reached a peak after 3 minutes and returned to basal level after about 30 minutes(Fig. 2B).

As mentioned above, G-protein-coupled receptors are known to activate Erk-1/2 through different second messenger pathways or by transactivation of EGF receptors. To identify the signals involved in LTD₄-induced activation of Erk-1/2 and to compare them with those involved in LTD₄-effected cell proliferation(Fig. 1), we initially used compounds that inhibit various signaling molecules. We found that pre-incubation with 500 ng/ml PTX for 2 hours, 30 μM GF109203X for 30 minutes, 10 μM PP1 for 15 minutes (a Src family kinase inhibitor) and 50μM PD98059 for 30 minutes blocked LTD₄-induced activation of Erk-1/2 (Fig. 2C). Neither the protein kinase A inhibitor Rp-cAMPS (50 μM for 30 minutes) nor the PI 3-kinase inhibitors wortmannin (100 nM for 10 minutes) and LY294002 (50 μM for 30 minutes) had any inhibitory effect on the LTD₄-induced activation of Erk-1/2 (Fig. 2D). These results suggest that a heterotrimeric G-protein, PKC,and MEK are involved in LTD₄-mediated activation of Erk-1/2. Inhibition of LTD₄-induced Erk-1/2 activation by the PP1 can readily be explained by the involvement of a Src-like kinase(s), although it could also imply crosstalk between the LTD₄ receptor and the EGF receptor.

To determine whether crosstalk occurs between LTD₄ and EGF receptors, we first investigated possible signals involved in EGF-induced Erk-1/2 activation. As expected, EGF-induced activation of Erk-1/2 was inhibited by genistein (a protein tyrosine kinase inhibitor; 50 μg/ml for 30 minutes), PD153035 (an EGF-receptor inhibitor; 2 μM for 30 minutes),PD98059 (50 μM for 30 minutes) and to a lesser extent by PP1 (10 μM for 15 minutes), but not by 500 ng/ml PTX (for 2 hours), GF109203X (30 μM for 30 minutes) or PKC depletion by 1 μM TPA for 24 hours(Fig. 3A,C). The lack of effect of PTX indicates that EGF-mediated Erk-1/2 activation in intestinal epithelial cells does not occur via the LTD₄ receptor, which has been suggested for the effects of EGF on the cytoskeleton in fibroblasts(Peppelenbosch et al., 1995). We also used an antibody that recognizes the phosphorylated form of the EGF receptor to determine whether LTD₄ participates in activation of the EGF receptor. As shown in Fig. 3B, stimulation with LTD₄ did not lead to any detectable phosphorylation of the EGF receptor, whereas treatment with EGF caused a 40-fold increase in EGF receptor phosphorylation. Furthermore, the LTD₄-induced activation of Erk-1/2 was not affected by the EGF-receptor inhibitor PD153035 (Fig. 3C). Nor did pretreatment with the FGF-1-receptor inhibitor SU5402 or the more general receptor-tyrosine kinase inhibitor SU4984 (blocking FGF-,PDGF- and insulin-receptors) affect the LTD₄-induced activation of Erk-1/2 (data not shown). These results suggest that EGF-induced activation of Erk-1/2 is not mediated through the CysLT₁ receptor and that LTD₄ is not involved in stimulation of the EGF receptor.

We have previously shown that LTD₄ induces translocations (i.e. activation) of α, δ and ϵ PKC isoforms but not of PKCβII, PKCμ or PKCζ (no other novel PKC isoforms is expressed)in intestinal epithelial cells (Thodeti et al., 2001). Furthermore when these cells were subjected to prolonged (24 hours) treatment with TPA they exhibited total downregulation of PKCα and PKCϵ but only partial downregulation of PKCδ(Thodeti et al., 2001). In the leukemia cell line THP-1, LTD₄ has been suggested to activate Erk-1/2 through a PKCα-dependent pathway(Hoshino et al., 1998). However, LTD₄-induced activation of Erk-1/2 in intestinal epithelial cells was only abolished by a high concentration (30 μM) of the PKC inhibitor GF109203X (Fig. 2C). A lower concentration of GF109203X (2-10 μM) impaired the TPA- but not the LTD₄-induced activation of Erk-1/2 (data not shown), suggesting that a novel PKC isoform(s) is involved in producing the effects of LTD₄ on Erk-1/2. In order to identify the PKC isoform involved in Erk-1/2 activation in intestinal cells, we first investigated the effect of PTX, which blocks the LTD₄-mediated activation of Erk-1/2, on the LTD₄-induced translocation of PKCα, δand ϵ. LTD₄ induced a rapid PTX-dependent translocation of PKCα, δ and ϵ to a membrane fraction and a subsequent reduction of these isoforms in a cytosolic fraction from these cells(Fig. 4A). We then investigated a possible involvement of calcium-dependent PKCs. Cells were preincubated with the calcium chelator MAPTAM (10 μM for 1 hour) before stimulation with LTD₄. Such a chelation of cytosolic free calcium did not reduce the LTD₄-induced activation Erk-1/2(Fig. 4B). Furthermore,addition of the calcium ionophore ionomycin (1 μM for 5 minutes) did not stimulate Erk-1/2 nor did it affect the LTD₄-induced activation of ERK-1/2 (Fig. 4B). These data make participation of the calcium-dependent PKCα isoform unlikely. We also noted that the PKCδ inhibitor rottlerin (10 or 30 μM was added for 30 minutes) had no effect on the LTD₄-induced activation of Erk-1/2 (Fig. 4C). These data argue against an involvement of PKCδ in the LTD₄-induced activation of Erk-1/2. To obtain more direct evidence for a role of PKCϵin the LTD₄-induced Erk-1/2 activation, we performed the following three experiments. Firstly, we studied the time course of TPA-induced downregulation of the different PKC isoforms, and this revealed that PKCϵwas downregulated much earlier (4 hours) than PKCα and PKCδ(Fig. 4D). Parallel experiments showed that such a pretreatment with TPA for only 4 hours abolished the LTD₄-induced activation of Erk-1/2(Fig. 4D), suggesting that PKCϵ is involved in such activation. Secondly, we examined the effect of LTD₄ on Erk-1/2 activation in cells transfected with either the regulatory domain of PKCϵ (RD-PKCϵ) or the regulatory domain of PKCδ (RD-PKCδ). The isolated regulatory domains have been suggested to work as isoform-specific dominant-negative inhibitors of PKC(Jaken, 1996), and inhibition of specific isoforms with these domains has been successfully utilized in several studies (Cai et al.,1997; Kiley et al.,1999; Massoumi and Sjölander, 2001). We noted that expression of RD-PKCϵblocked the LTD₄-induced activation of Erk-1/2, whereas expression of RD-PKCδ did not (Fig. 4E). These results were obtained even though the expression level of the GFP-tagged RD-PKCϵ is less than that of the GFP-tagged RD-PKCδ, which was revealed by reprobing the western blot with an anti-GFP antibody (Fig. 4E). Thirdly, we transfected cells with kinase-dead PKCϵ(K^-PKCϵ) or the corresponding empty vector and examined their effects on LTD₄-induced Erk-1/2 activation. We noted that expression of K^-PKCϵ totally inhibited the LTD₄-induced activation of Erk-1/2, whereas transfection of the empty vector had no effect (Fig. 4F). These results clearly show that PKCϵ is the isoform involved in the LTD₄-induced activation of Erk-1/2 in intestinal cells.

Employing in vitro kinase assays, we found that LTD₄ induced rapid activation of Raf-1 (Fig. 5A), but not B-Raf (Fig. 5B), in intestinal epithelial cells. The stimulation of Raf-1 peaked approximately 2 minutes after addition of the leulotriene(Fig. 5A). G-protein- and PKC-dependent activation of Erk has also been found to be mediated by MEKK1 rather than Raf-1 (Vuong et al.,2000). To investigate whether the LTD₄-induced activation of Erk-1/2 is mediated by Raf-1, we transfected cells with either a HA-tagged kinase-dead Raf-1 expressing vector [K^-Raf-1(Sutor et al., 1999)] or an empty vector and examined their effect on LTD₄-induced Erk-1/2 activation in these cells. The results clearly show that expression of K^-Raf-1 inhibited the LTD₄-induced activation of Erk-1/2, whereas the empty vector had no such effect(Fig. 5C). In subsequent experiments, LTD₄-mediated activation of Raf-1 was demonstrated to be abolished by pre-incubation with PTX (500 ng/ml for 2 hours) or GF109203X(30 μM for 30 minutes) or by TPA-induced downregulation of PKC(Fig. 5D). These data indicate that Raf-1 is located downstream of PKC in LTD₄-induced activation of Erk-1/2 (Fig. 5D). On the basis of these results, we performed immunoprecipitations to examine the possibility of an LTD₄-mediated association between PKCϵ and Raf-1, but we found no such association (data not shown). It is quite likely that Raf-1 is activated by PKCϵ, even if there is no physical connection between the two proteins; alternatively PKCϵ could stimulate Raf-1 indirectly via activation of Ras. It has previously been demonstrated that PKC can play a role in the activation of Ras in lymphocytes(Downward et al., 1990). To test the ability of PD98059 to inhibit the LTD₄-induced activation of MEK, we employed an in vitro kinase assay(Fig. 5E). We found that pre-incubation with PD98059 (50 μM for 30 minutes), PTX (500 ng/ml for 2 hours) or PP1 (10 μM for 15 minutes) inhibited the LTD₄-induced activation of MEK.

In the present study we clearly show that LTD₄ causes a rapid and transient activation of Ras (Fig. 6A). To explore a possible role for active Ras in the LTD₄-induced activation of Erk-1/2, we either incubated the cells with the Ras farnesyltransferase inhibitor FTI-277 (20 μM for 48 hours) or transfected them with HA-tagged N17 Ras. The latter is an Asn-17 mutant of Ha-Ras, which blocks multiple downstream signals such as activation of Raf-1 and phosphorylation of MAPK (Odajima et al., 2000). In cells pre-incubated with FTI-277, we detected almost no increase in GTP-Ras in the pull-down assay of cells stimulated with LTD₄ for 1 minute (Fig. 6B); however the LTD₄-induced activation of Erk-1/2 was not affected at all. Under identical conditions we noted reductions of EGF-induced Ras activation and simultaneous and similar reductions in EGF-induced Erk-1/2 activation. To gain further support for LTD₄-induced Ras-independent activation of Erk-1/2, we cotransfected cells with N17 Ras and empty EGFP vector, and following stimulation with LTD₄ or EGF the cells were immunostained with a phospho-Erk antibody (Fig. 7). In unstimulated cells, the staining with the phospho-Erk antibody was weak in both N17-Ras-transfected and non-transfected cells. In contrast, cells stimulated with LTD₄ stained brightly with the phospho-Erk antibody regardless, of whether the cells were transfected with N17 Ras or not(Fig. 7), thus indicating that LTD₄-induced Erk-1/2 activation is Ras-independent. As a positive control, cells transfected with or without N17 Ras were stimulated with EGF(Fig. 7). Cells stimulated with EGF stained brightly with the phospho-Erk antibody, provided that they were not transfected with N17 Ras (Fig. 7). These latter control data are in agreement with the previously reported Rasdependency of EGF-induced Erk-1/2 activation(Daub et al., 1996). Furthermore, parallel experiments revealed that neither preincubation with PTX or GF109203X nor TPA downregulation of PKC impaired the LTD₄-induced activation of Ras (data not shown). These findings clearly indicate that LTD₄-mediated activation of Ras is separate from the Erk-1/2 signaling pathway.

We found that in intestinal epithelial cells the LTD₄-induced proliferative response is regulated by dual intracellular signaling pathways:one that is initiated by a pertussis toxin-sensitive G-protein and requires activation of Erk-1/2 and one that is pertussis toxin-insensitive but dependent on Ras activation (Fig. 8). The initiation and existence of two parallel pathways is in good agreement with previous reports regarding the Ca²⁺ signaling properties of the LTD₄ receptor(Sjölander et al., 1990;Adolfsson et al., 1996;Grönroos et al., 1996;Thodeti et al., 2000). Our initial observation that Erk-1/2 was involved in the LTD₄-induced proliferative response suggests that a G-protein-Erk-1/2 signaling pathway is activated in these epithelial cells. This pathway may involve receptor crosstalk, because it is well known that Erk-1/2 and the other MAPKs integrate signals from different receptors and second messengers(Garrington and Johnson, 1999;Robinson and Cobb, 1997), and communication between the EGF and the LTD₄ receptor has been reported (Peppelenbosch et al.,1995). However, in our study, EGF-induced Erk-1/2 activation was not affected by PTX, PP1, GFX or PKC depletion, and LTD₄ did not trigger activation of the EGF receptor, findings that argue against a possible transactivation between these two receptors. Furthermore, pre-incubation with the EGF-receptor inhibitor PD153035 did not affect the LTD₄-induced activation of Erk-1/2. Consequently, our results suggest that a distinct LTD₄—G-protein—Erk-1/2 signaling pathway mediates proliferation of intestinal epithelial cells. In agreement with this, other investigators (Le Gall et al.,2000) have proposed that the MAP kinase pathway plays a role in protecting cells against apoptosis, possibly by counteracting cell death induced by loss of matrix contact, cytoskeletal integrity or extracellular mitogenic factors. Therefore, we extended our characterization of the signaling elements that are involved in LTD₄-mediated regulation of Erk-1/2 activity and thereby also participate in the proliferative response of intestinal epithelial cells.

Our finding that LTD₄ causes a transient, time- and concentration-dependent activation of Erk-1/2 in intestinal epithelial cells is compatible with the effect of PD98059, a specific MEK inhibitor, on the LTD₄-induced proliferative response in these cells. Also in accordance with effects on LTD₄-mediated cell proliferation, we found that activation of Erk-1/2 by LTD₄ involves stimulation of a PTX-sensitive G-protein, a Src-like protein and MEK. Other types of G_i-protein-coupled receptors in other kinds of cells have been shown to initiate the MAPK signaling cascade through release of Gβγsubunits and activation of a Src-like kinase(s)(van Biesen et al., 1996). We found both similarities and discrepancies between the signaling pathways leading to activation of Erk-1/2. These observations agree with the conclusion drawn by Luttrell and coworkers that the mechanisms involved in activation of MAPK are heterogeneous and appear to depend not only on the nature of the G-protein-coupled receptor but also on the cell type(Luttrell et al., 1996). In addition to the signals discussed above, our findings that PKC inhibitors and downregulation of PKC isoforms impair the LTD₄-induced activation of Erk-1/2 in intestinal epithelial cells suggest that PKC plays an important role in this signaling cascade.

Hoshino and colleagues (Hoshino et al.,1998) have demonstrated that LTD₄ activates Erk-1/2 via a PTX-insensitive but PKCα- and Raf-1-dependent pathway in the monocytic leukemia cell line THP-1. However, we found that LTD₄ activates Erk-1/2 via a PTX-sensitive G-protein/PKC-dependent pathway in Int 407 intestinal epithelial cells. We also obtained evidence that it is the ϵisoform of PKC that is involved in LTD₄-mediated stimulation of Erk-1/2. First, a high concentration of GF109203X was required to impair LTD₄-induced activation of Erk-1/2, and this compound is known to be a more potent inhibitor of classical PKC than of novel PKC isoforms(Chen et al., 1999), implying involvement of a novel isoform, such as PKCϵ. Second, we have earlier shown that LTD₄ not only resulted in translocation/activation of PKC α but also of the δ and ϵ isoforms(Thodeti et al., 2001). Third,exposing the cells to TPA for 4 hours, a sufficient amount of time for such a TPA treatment to abolish the LTD₄-induced activation of Erk-1/2,caused a significant downregulation only of PKCϵ. Fourth, by transfecting and using the regulatory domains of PKCϵ and PKCδ as isoform-specific dominant-negative inhibitors of PKC(Jaken, 1996), we conclude that PKCϵ, but not PKCδ, is involved in the LTD₄-induced activation of Erk-1/2. Finally, the LTD₄-induced activation of Erk-1/2 was totally inhibited in cells transfected with K-PKCϵ.

The exact location at which PKC takes part in the activation of Erk-1/2 most probably depends on the stimulus and the cell type examined. It has been shown that the signaling pathway triggered by activation of the T-cell receptor on T-lymphocytes involves a PKC upstream of Ras(Downward et al., 1990). Nonetheless, we found that LTD₄-induced activation of Ras was insensitive to both PKC inhibitors and downregulation of PKC. In addition,despite our demonstration that LTD₄ can activate Ras, this activation does not seem to be involved in the LTD₄-effected stimulation of Erk-1/2. This conclusion was formed on the basis of the observation that inhibition of Ras, by either the Ras inhibitor FTI-277 or transfection with N17 Ras, had no effect on the LTD₄-induced activation of Erk-1/2. Our results instead agree with data showing that activation of Erk-1/2 by the M1 receptor involves PKC at a point that is upstream of Raf-1 activation (Marais et al., 1998).

We performed in vitro assays for Raf activities and observed that LTD₄ caused activation of Raf-1 with a peak around 2 minutes, which is in line with the LTD₄-induced activation of Erk-1/2. LTD₄-mediated activation of Raf-1 is sensitive to PTX, PKC inhibitors and downregulation of PKC by TPA. In accordance with our data indicating involvement of PKCϵ in LTD₄-induced activation of Erk-1/2, several investigators have shown that, once activated, PKCϵ can stimulate Raf-1 via phosphorylation of serine(Kolch et al., 1993;Khalil and Morgan, 1993;Cai et al., 1997). In support of our results, Velarde (Velarde et al.,1999) studied vascular smooth muscle cells and demonstrated that activation of PKCϵ is required for bradykinin-induced stimulation of Erk-1/2.

In conclusion, our results demonstrate that LTD₄ can promote proliferation of intestinal epithelial cells by a traditional Ras-dependent pathway but, more interestingly, even in the absence of Ras activity a G-protein/PKCϵ/Raf-1/MEK signaling pathway can induce proliferation via activation of Erk-1/2 in these cells. Activation of such signaling pathways and the subsequent increase in proliferation indicate that this inflammatory mediator can contribute to growth of intestinal cells during pathological inflammatory conditions.

The authors are grateful to R. Metcalf, Zeneca Pharmaceuticals,Macclesfield, UK, for the ZM198,615, Arthur Mercurio, Beth Israel Deaconess Medical Center, Boston, MA, USA for the W437 kinasedead PKCϵ construct and Larry M. Karnitz, Oncology Research, Mayo Clinic, Rochester, MN, USA for kinase-inactive c-Raf. We are also indebted to Patricia Ödman for linguistic revision of the manuscript. This work was supported by the grants to A.S. from the Swedish Medical Research Council, the Magnus Bergvall's Foundation, the Greta and Johan Kocks Foundation, the Åke Wiberg Foundation, the Crafoord Foundation, the Foundations at Malmö University Hospital, and the Österlund Foundation.