Morphogen-induced decline in Gialpha2 triggers F9 teratocarcinoma stem cell progression via phospholipase C and mitogen-activated protein kinase.

The linkage between G and morphogen-induced promotion of F9 embryonic teratocarcinoma stem (F9 stem) cells to primitive endoderm was explored using probes of the mitogen-activated protein (MAP) kinase network. The morphogen-induced decline in G is shown to trigger activation of phospholipase C, thereby activating protein kinase C, MAP kinase, and cell progression to primitive endoderm. In the absence of retinoic acid, reduction-of-function mutants (G-deficient) display the effects of morphogen, i.e. activation of phospholipase C, protein kinase C, MAP kinase, and progression to primitive endoderm. Gain-of-function mutants (expressing the Q205L activating-mutation of G) displayed no activation of phospholipase C, protein kinase C, MAP kinase and no progression to primitive endoderm, even in the presence of retinoic acid. Selective inhibitors of protein kinase C, like the gain-of-function mutations, effectively block morphogen-induced progression to primitive endoderm. Morphogen triggers F9 stem cell progression by triggering G loss and thereby activation of downstream elements, including protein kinase C and MAP kinase.

Mouse F9 teratocarcinoma stem (F9 stem) cells provide a plentiful source of embryonic cells whose differentiation can be induced or influenced by exogenously added agents. F9 stem cells show little spontaneous differentiation in culture and can be induced to differentiate to primitive endoderm (PE) 1 by physiological concentrations of retinoic acid (RA) (1). The new cell type PE of F9 cells give rise to two different cell types, both of which are characteristic of the extraembryonic endoderm lineage in the mouse embryo; addition of dibutyryl cyclic AMP yields parietal endoderm (2), whereas aggregation yields cells characteristic of visceral endoderm (3). This differentiation process can be monitored by following the expression of several specific markers. The production of tissue plasminogen activator (tPA), Type IV collagen, laminin, and c-jun is induced with differentiation (4,5), while the stage-specific embryonic antigen disappears from the cell surface (6) and the level of c-myc is dramatically decreased (7). Because of these distinct and versatile responses to differentiation, F9 cells have been widely used as a model for studying the mechanism of embryonic differentiation. Despite numerous studies, however, the molecular mechanism of differentiation, particularly the nature of the intracellular cascade leading to differentiation, remains poorly understood.
The G-proteins are a family of membrane-associated guanine nucleotide-binding proteins that transduce signals from cell surface receptors to intracellular effectors. Members of this family are heterotrimers composed of ␣, ␤, and ␥ subunits (8). The ␣ subunit confers receptor and effector specificity on the heterotrimer. When the G-protein is activated by interaction with receptor, the ␣ subunit exchanges bound GDP for GTP. The intrinsic GTPase activity of the ␣ subunit restores it to the basal state in which GDP is bound. G ␤␥ subunits, too, modulate the activities of effectors units, including some K ϩ channels, PLC␤ (forms 2 and 3), pheromone action in budding yeast, and some adenylyl cyclase (for a recent review, see Ref. 9). This form of signal transduction is basic to the mechanisms that cells use in responding to hormones, neurotransmitters, and growth factors.
RA induces F9 stem cell progression to PE and a sharp decline in G i␣2 (10). A decrease in G i␣2 by expression of antisense G i␣2 RNA (reduction-of-function) mutant induces PE-like phenotype in the absence of RA, while an increase in G i␣2 by expression of a constitutively active (gain-of-function) mutant G i␣2 Q205L is sufficient to block differentiation (11). Pertussis toxin-sensitive G-proteins, like G i␣2 , have been implicated as potential mediators both of inhibitory phospholipase C (PLC) signaling and of mitogenic responses in several cell lines (12)(13)(14)(15)(16)(17)(18)(19). Recently, suppression of G i␣2 in cultured cells and adipocytes from transgenic mice has been shown to enhance PLC signaling, providing a possible link between G i␣2 and upstream regulators of the mitogen-activated protein kinase (MAPK) network (20). We tested this linkage, exploring the PLC pathway.

EXPERIMENTAL PROCEDURES
Cell Growth and Transfection-The wild-type and transfected F9 stem cells were cultured on 0.1% gelatin-coated Falcon Petri dishes in Dulbecco's modified Eagle's medium supplemented with 15% of fetal calf serum (HyClone Laboratories). The retrovirus infected reductionof-function mutants pLNCX-ASG i␣2 F9 cells (F33) and gain-of-function mutants pCW1G i␣2 Q205L F9 cells (FQ) have been previously described (11). Clones were selected in 500 g/ml and maintained in 100 g/ml (active form) of G418 sulfate (Life Technologies, Inc.).
[ 3 H]Inositol 1,4,5-Triphosphate (IP 3 ) Mass Assay-Cells (10 6 ) were cultured for 4 days in the presence and absence of RA (100 nM). PLC agonist was added for 10 s and the assay terminated by addition of perchloric acid. After neutralization, the mass of IP 3 was determined by a competitive protein binding assay of [ 3 H]IP 3 using rabbit cerebellar membrane as the IP 3 receptor (21). Binding assays contained cell extracts, 2.5 nM [ 3 H]IP 3 , and 100 l of binding protein in 50 mM Tris-HCl, 1 mM EDTA, pH 8.3. Assays were incubated 10 min at 4°C followed by centrifugation at 10,000 ϫ g for 5 min. Pellets were counted for radioactivity determination.
Determination of PE-Production of tPA is the hallmark for PE. Stem cells induced to PE produce and secrete tPA as well as assume a characteristic morphology, (i.e. extended spindle shape with defined foci of growth). To induce PE, RA was added for 4 days at 100 nM (22). For tPA determinations, the culture medium of cells was assayed using the amidolytic assay (23). One unit of tPA is arbitrarily defined as that amount of tPA that results in a reaction rate of 10 Ϫ7 ⌬]A 405 min Ϫ2 (change in the optical absorbance at 405 nm divided by the square of the time, min). For phase-contrast microscopy, the cells were cultured for 4 days and then fixed with 3% (w/v) paraformaldehyde. Fixed cells were photographed for phase-contrast microscopy with a Zeiss Axiphot.

HPLC Analysis of Metabolically Labeled [ 3 H]Inositol
Phosphates-Confluent cells (10 7 ) were labeled with myo-[ 3 H]inositol (90 Ci/ml) at day 3 of culture for 24 h. The incubation was terminated at day 4 by addition of perchloric acid, and inositol phosphates were extracted with fluorotrichloromethane and tri-n-octylamine. The extent of metabolic labeling was similar for all clones tested. Equivalent amounts of label, [ 3 H]inositol phosphates, were subjected to analysis by anion-exchange HPLC using a Whatman Partisil-10 SAX column and a linear gradient of formic acid, pH 3.5 (24). Inositol phosphates were identified by reference to the standards of inositol 1-phosphate, inositol 1,4-bisphosphate, inositol 1,3,4-trisphosphate, inositol 1,4,5-trisphosphate, and inositol 1,3,4,5-tetraphosphate under the same separation conditions. Radioactivity was quantitated by liquid scintillation counting.
Immunoblotting-Aliquots of crude membrane fractions (100 g/ lane) from each subclone were subjected to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE), the separated proteins transferred to nitrocellulose, and the blots stained with antibodies specific to G i␣2 , G q␣ , G i␣3 , and G ␤1 (10,11). The immune complexes were made visible by staining with a second antibody goat anti-rabbit IgG coupled to calf alkaline phosphatase.
Calcium/Calmodulin-dependent Protein Kinase II (CaM Kinase II) Assay-CaM kinase II activity was assayed by incorporation of [␥-32 P]ATP (at 30°C for 2 min) into the peptide substrate syntide-2 in the presence and absence of Ca 2ϩ /CaM (25). Reactions were terminated by spotting the reaction mixture on P-81 phosphocellulose papers, and quantified in a scintillation counter after washing in 1% (v/v) phosphoric acid. CaM kinase II activity is defined as activity sensitive to the CaM kinase-specific inhibitor KN-62.
In Situ Kinase Assay in Renatured SDS Gels-20 g of protein from cell lysates were subjected to SDS-PAGE on a 10% acrylamide gel with myelin basic protein (MBP, 0.1 mg/ml) copolymerized in the running gel (26). After electrophoresis, the proteins in the gel were successively denaturated, renaturated, and subjected to in situ phosphorylation by [␥-32 P]ATP at 30°C for 1 h. The reaction was terminated and washed extensively by immersing the gel in 5% trichloroacetic acid and 10 mM sodium pyrophosphate.
Quantitation of Diacylglycerol (DAG) Mass-Cells (10 6 ) were grown to confluence for 4 days, and lipids were extracted into chloroform. DAG was measured using a protein kinase assay (27), followed by thin layer chromatography to separate radioactivity incorporated into phosphatidic acid generated from the reaction. The DAG mass was calculated from an authentic sn-1,2-DAG standard curve.
Northern Blot-Total cellular RNA was prepared using RNA-STAT 60 reagent (Tel-Test, Friendswood, TX) from cells grown for 4 days with or without treatment by RA. A 40-g aliquot of total RNA was subjected to separation on 1.2% formaldehyde-containing agarose gel, transferred by capillary to Nytran membrane, cross-linked by ultraviolet radiation (Stratalinker), and then probed by hybridization with a 32 P-labeled mouse PKC␣ cDNA probe. The cDNA probe was labeled by random Determination of MAPK Activities and Proteins-Cells (5 ϫ 10 7 ) grown for 4 days were harvested and lysed. To provide an independent measure of MAPK activation, cells were serum-starved in medium containing 0.1% bovine serum albumin for 18 h and then stimulated for 5 min with 50 ng/ml mouse epidermal growth factor (EGF, Calbiochem). Soluble extracts (1 ml) were fractionated by chromatography on a Mono Q HR 5/5 FPLC column (Pharmacia Biotech Inc.), developed with a 28-ml, linear 0 -350 mM NaCl gradient, while collecting 1-ml fractions (30). MAPK activity was quantified by measurement of the phosphorylation of EGF receptor peptide (662-681) (15 min at 30°C), which contains the consensus MAPK phosphorylation site, PXTP. Column fractions were precipitated with tricholoracetic acid/sodium deoxycholate and subjected to SDS-PAGE on 10% acrylamide gels. The separated proteins were transferred to a nitrocellulose membrane, probed (1:1000) with an anti-MAP2 kinase monoclonal antibody (Life Technologies, Inc.) and subsequently with an alkaline phosphatase-conjugated second antibody (1:10,000), made visible by staining. For determination of MAPK expression, whole-cell extracts were immunoprecipitated with anti-MAP2 kinase monoclonal antibody followed by incubation with protein G-agarose (Life Technologies, Inc.) and subjected to SDS-PAGE. The transferred proteins were immunoblotted with anti-MAP2 kinase monoclonal antibody and then stained with a peroxidase-conjugated second antibody, made visible by ECL (DuPont).
MEK Activity Assay-Cells were grown, treated, harvested, and lysed as described above. The lysates were applied to a Mono S HR 5/5 FPLC column and eluted with a 28-ml, linear 0 -350 mM NaCl gradient, with 1-ml fractions. Column fractions were assayed for the ability to phosphorylate recombinant, kinase-deficient MAPK at 30°C for 30 min (30). Reaction samples were spotted onto P-81 phosphocellulose filters, which were then washed in 1% phosphoric acid. Radioactivity was determined by liquid scintillation counting.
Protein Kinase Inhibitor Studies-Cells were grown for 4 days in the presence and absence of RA (100 nM) to induce PE. Selective protein kinase inhibitors were examined for their ability to inhibit target kinases in F9 cells over a wide range of concentrations. The K i was established and agreed well with published values derived from studies in other mammalian cells. For studies of the effects of the inhibitors on progression, inhibitors were added simultaneously with the morphogen RA and the effects on progression as well as MAPK activity measured as indicated in the legend to Table I. Progression was established by measurement of the PE-marker protein tPA, as well as visually by the cell morphology.

RESULTS AND DISCUSSION
We explored the relationship between G i␣2 and F9 stem cell progression to PE using stem cells, loss-of-function, and gainof-function mutants. G i␣2 is deficient in the F33 loss-of-function clones (Fig. 1), mimicking the loss of G i␣2 in F9 stem cells induced to PE by RA (9). The gain-of-function clones express the Q205L mutant of G i␣2 , which is constitutively activated (FQ). Owing to expression of the Q205L mutant form, immunoreactive G i␣2 levels are increased in immunoblots of FQ FIG. 1. Expression of G i␣2 is elevated in FQ clones and reduced in F33 clones, while expression of G q␣ , G i␣3 and G ␤1 are similar among F9 stem cells, F33, and FQ clones. Proteins (100 g/lane) were separated SDS-PAGE on 10% gels, transferred to nitrocellulose membranes (Schleicher & Schuell), and probed with primary antibodies to (1:200) G i␣2 , G q␣ , G i␣3 , and G ␤1 (10,11). Blots were incubated with an alkaline phosphatase-conjugated second antibody (1:1000), and developed with nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate p-toluidine. The vector stably transfected cells showed expression similar to that F9 stem cells (data not shown).
clones, whereas immunoreactive G i␣2 is essentially absent in the F33 clones. The levels of G q␣ , G i␣3 , and G ␤1 , in contrast, were quite similar in F33 and F9 clones (Fig. 1).
In order to determine to what extent PLC activity was altered, IP 3 mass accumulation was measured. IP 3 levels were found to increase severalfold as F9 stem cells (STEM) progressed to PE in response to RA (Fig. 2A). Both F9 stem cells and PE displayed elevated IP 3 levels in response to stimulation of PLC by 5-hydroxytyramine (5HT; Fig. 2B), as well as to bradykinin and ␣-adrenergic agonists (not shown).
We tested the linkage further using loss-of-function mutants, stable transfectants expressing RNA antisense to G i␣2 and lacking G i␣2 (F33). In the absence of morphogen, loss-of-function mutants displayed elevated IP 3 levels and progression to PE. Gain-of-function mutants, stably expressing the Q205L activating mutant G i␣2 (FQ), displayed no activation of PLC, no tPA production, and a stem-like phenotype in the absence (Fig.  2) or presence (not shown) of RA. Analysis by HPLC of the water-soluble inositol phosphates from cells metabolically labeled with [ 3 H]inositol confirmed the mass assays of IP 3 levels for the different clones (Fig. 2C). Levels of IP 3 (arrows) were elevated markedly in cells induced to PE by RA and in the F33 G i␣2 -deficient clones that progress to PE in the absence of RA. Quantification of the label recovered as IP 3 from HPLC of  (n ϭ 4). By Student's test, significance of difference is p Ͻ 0.05 (*) or Ͻ 0.01 (**) with respect to wild type. C and D, progression of F9 stem cells to PE either by RA or by deficiency of G i␣2 (F33) activates PLC and increases IP 3 levels, as measured by metabolic labeling (C) and quantified by liquid scintillation (D). The elution of IP 3 in indicated by the arrows. Metabolic labeling of the different clones was equivalent quantitatively on a "per cell" basis, and equivalent amounts of label were subjected to analysis by HPLC. E and F, deficiency in G i␣2 (F33), like the action of RA on F9 stem cells, induced progression of stem cell to PE phenotype, as analyzed by the PEmarker, tPA (E) or by phase-contrast microscopy (F). One unit of tPA is arbitrarily defined as that amount of tPA that results in a reaction rate of 10 Ϫ7 ⌬A 405 min Ϫ2 . For phase-contrast microscopy, the cells were fixed with 3% (w/v) paraformaldehyde and photographed with a Zeiss Axiphot. The micrographs shown are representative of five independent experiments for each clone or condition. Stem cells stably transfected with vectors alone, showed activities similar to that F9 stem cells throughout (data not shown).
samples loaded with equivalent total counts is shown (Fig. 2D). In stem or FQ cells, in contrast, the IP 3 levels were lower (Fig.  2C). Excepting the changes in IP 3 levels, overall profiles of inositol metabolites (Fig. 2C) and labeling of inositol-containing lipids (not shown) were similar. Expression of the PEspecific marker, tPA, and phase-contrast microscopy define morphogen-induced progression to PE (Fig. 2, E and F, respectively). Stem cells stably transfected with vectors alone showed activities similar to that ES cells throughout.
Calcium mobilization in response to elevated IP 3 levels was probed by in vitro assays of calcium/calmodulin-dependent protein kinase II (CaM kinase II, Fig. 3). Progression to PE by RA elevated CaM kinase II activity in either the absence (Basal) or the presence of activator calcium and calmodulin (ϩCaM). The in situ assay of CaM kinase II (26, 31) revealed a sharp increase in the ␤-form (mass of 60 kDa) compared to ␣-form (50 kDa) of kinase activity (inset , Fig. 3). G i␣2 deficiency alone (F33), like RA, increased CaM kinase II activity, well above that of PE

FIG. 4. RA-induced progression to PE, like G i␣2 deficiency (F33), activates PLC and increases intracellular DAG levels.
Cells (10 6 ) were grown to confluence for 4 days. Cells were treated with and without 5HT (10 Ϫ5 M, 10 s at 37°C), and then extracted into chloroform. DAG was measured using a protein kinase assay (27), followed by thin-layer chromatography to separate radioactivity incorporated into phosphatidic acid generated from the reaction. The DAG mass was calculated from an authentic sn-1,2-DAG standard curve. The data are mean values Ϯ S.E. from three separate experiments. By Student's test significance of difference is p Ͻ 0.01 (**) with respect to wild type. Stem cells stably transfected with vectors alone, showed activities similar to that F9 stem cells (data not shown). with the ␤-isoform activity predominant. Expression of a constitutively active G i␣2 (FQ), in contrast, not only blocks progression, but also reduces CaM kinase II activity to levels below those of stem cells. Ionophore (A23187 or ionomycin)induced activation of CaM kinase II activity revealed that less than half of the CaM kinase II was activated in stem cells, whereas Ͼ70% was activated by RA treatment (PE) or loss of G i␣2 (F33, data not shown).
PKC plays a prominent role in cell signaling, growth and proliferation (32,33). The levels of DAG, a metabolite of the PLC pathway and intracellular activator of PKC, were explored in cells progressing to PE by RA or by loss of G i␣2 (F33). DAG content of stem cells nearly doubled when induced to PE by RA (Fig. 4). Likewise, G i␣2 deficiency alone, in the absence of RA, elevated DAG content nearly 2.5-fold. Expression of Q205L mutant G i␣2 displayed DAG levels similar to that of stem cells. 5HT elevated DAG levels in stem, PE and G i␣2 -deficient cells, whereas in cells expressing the Q205L mutant G i␣2 the response to 5HT was essentially abolished.
A critical linkage between PKC and both G i␣2 and stem cell To measure growth factor-stimulated MEK activities, the cells were serum-starved for 18 h and incubated with mouse EGF (50 ng/ml) for 5 min. Cell extracts (1 ml) were fractionated by chromatography on a Mono S FPLC column, developed with a 28-ml, linear 0 -350 mM NaCl gradient, while collecting 1-ml fractions. The enzyme activity was quantified by measurement of the phosphorylation of inactive rMAPK. Endogenous MAPK elutes in fractions 1-5, whereas MEK is contained in fractions 8 -13. Fractions 1-7, which include endogenous MAPK activity, are not displayed; only the region of the chromatogram relevant to MEK is shown (30). progression was revealed by measurement of PKC activity. In the absence of an exogenous activator (PMA), PKC activity in PE was more than 10-fold greater than in F9 stem cells not induced by RA (Fig. 5). Remarkably, total PKC activity measured in the presence of PMA showed the same striking difference following progression, i.e. PE display a 14-fold greater amount of PKC activity than stem cells. The G i␣2 -deficient clones (F33) that progress to PE in the absence of RA mimicked the cells induced to PE by RA, displaying elevated basal and total PKC activity. Expression of the constitutively active Q205L mutant G i␣2 (FQ), in stark contrast, resulted in an stem cell-like phenotype refractory to RA with levels of PKC similar to those of F9 stem cells prior to progression. Immunoblots of DEAE-cellulose-purified whole-cell extracts stained with antibodies to PKC␣ are consistent with the activity measurements (Fig. 6A). Northern analysis of total cellular RNA identified PKC␣ transcripts of 3.5 and 8.1 kilobases in PE and F33 clones. PKC␣ transcripts were not detected in Northern blots of RNA from either F9 stem cells or FQ clones (Fig. 6B, top), with relative equivalent sample loading (Fig. 6C, bottom). PKC isoforms other than PKC␣ do not appear to be major contributors to the changes in activity, as PKC␤ protein is undetected (data not shown), and the level of PKC␤ and PKC␥ transcripts is reduced, rather than increased, in progression to PE (34). Thus, stimulation by the morphogen RA leads not only to an increase in DAG generation and thereby PKC activation, but also to a sharp increase in expression of the PKC␣ mRNA.
MAPKs are members of a group of serine/threonine kinases that are activated in response to growth factors, mitogenic stimuli and neurotransmitters, that control cell proliferation and differentiation (35). Tyrosine kinase-encoded growth factor receptors and G i -coupled receptors (e.g. thrombin receptor) are capable of activating MAPKs in response to stimulation (36). Phorbol esters, like PMA, also activate MAPK in many cells. Immunoblotting of immunoprecipitated whole-cell extracts with antibodies to p42 reveals an enhanced expression of MAPK in PE and the F33 clones (Fig. 7A). In an effort to probe the downstream events triggered by G i␣2 decline and PKC activation, MAPK activity was assessed in cell extracts first subjected to anion-exchange chromatography FPLC. MAPK activity resolves as two peaks, the major p42 MAPK eluting in fractions 12-16 and a minor p44 MAPK in fractions 17-19. In F9 stem cells, MAPK activity was relatively low. Cells induced to PE by RA displayed a striking, 5-fold increase in MAPK activity (Fig. 7B). The G i␣2 -deficient F33 clones, like the PE cells, displayed a striking increase in MAPK activity in the absence of RA. The FQ clones expressing Q205L G i␣2 , in contrast, were shown to have profiles of MAPK not unlike the untreated ES cells (Fig. 7C). Treatment of the cells with EGF provides a positive control, demonstrating that RA-induced progression to PE is accompanied by a substantial, though lesser activation of MAPK, approximately 60% of that obtained by stimulation with EGF (Fig. 7D).
For PE and F33 clones, immunoblotting of MAPK shows a modest increase in protein and an elution profile in which the p42 isoform is most prominent (Fig. 7E), which, together with the MAPK protein expression of the whole-cell extracts, suggests that activation of MAPK is the basis for increased activity both in F9 stem cells promoted to PE by RA and in G i␣2deficient cells that progress to PE in the absence of RA. Activation of the upstream element regulating MAPK activity, MEK, revealed a similar pattern in which progression to PE induced by RA activated MEK to a significant but lesser level that EGF (Fig. 8).
To probe events downstream of RA-induced G i␣2 loss mediating progression of stem cells to PE, the effects of protein kinase inhibitors on F9 stem cell progression to PE were explored (Table I). F9 stem cells were treated with RA in the absence and presence of selective inhibitors of PKC, protein kinase A, or CaM kinase, after establishing the dose-response curve for inhibition in F9 cells to be equivalent (not shown) to those in the literature performed with other cell lines (37)(38)(39)(40)(41). PKC-selective inhibitors calphostin C (37) (0.1 M), bisindoylmaleimide (38) (0.1 M), and H7 (39) (25 M) effectively blocked the ability of RA to activate MAPK and to promote F9 stem cell progression to PE. In agreement with earlier observations that cyclic AMP analogues and cholera toxin treatment do not influence progression to PE, inhibitors of protein kinase A, such as KT5720 (40) and HA1004 (39) at concentrations 20-fold in excess of their IC 50 values were shown to be without effect on either parameter. KN62 (41), a potent inhibitor of CaM kinase, blocked CaM kinase activity (data not shown), but neither MAPK activation nor progression to PE. CaM kinase II activity increases dramatically in stem cells progressing to PE and the RA-induced change may well be important to both calcium FIG. 9. Oligodeoxynucleotides antisense, but not sense, to MAPK abolish RA-induced progression of F9 stem cells to PE. F9 cells were incubated with 1.5 mM phosphorothioate oligodeoxynucleotides sense (S) and antisense (AS) to MAPK mRNA, using the sequences and conditions reported elsewhere (42). Inset, the level of MAPK expression was determined from whole-cell extracts subjected to immune precipitation and immunoblotting. The ability of RA to induce progression was measured by quantification of the production of tPA, the hallmark of the PE phenotype.  M) to induce PE. A series of dilution of protein kinase inhibitors were added at the same time with RA when the cells were split in an attempt to block F9 stem cells progression to PE. The inhibition of RA-induced PE was judged by two criteria, i.e. no extended spindle shape, and the same level of tPA secretion as that of the stem cells. The inhibitor concentrations listed were those in a wide range of inhibitor dilution tested. Data are from two independent experiments, each with identical effects on progression to PE.

Addition
Protein kinase inhibitors MAPK Activation signaling and longer term features of gene expression associated with the PE-phenotype. Clearly, RA triggers a decline in G i␣2 , thereby activating PLC, PKC, MAPK, and progression of F9 stem cells to PE. To probe the linkage between MAPK and RA-induced progression to PE, we employed phosphorothioate oligodeoxynucleotides antisense to MAPK p42 (42). Oligodeoxynucleotides antisense, but not sense, to MAPK nearly abolished immunoreactive MAPK in whole-cell blots of F9 cells (Fig. 9,  top). Measurement of MAPK activity by the in situ assay confirmed the loss of MAPK (not shown). Elimination of MAPK abolished the ability of RA to induce progression of F9 stem cells to PE (Fig. 9, bottom), establishing an obligate role of MAPK in the cascade from G i␣2 to stem cell progression.
RA-induced differentiation of a variety of tumor cell types is accompanied by increased expression of PKC, especially the ␣-isoforms (34,43). The role of G i␣2 in progression of these tumor cells in response to RA is not known. Loss-of-function mutants mimic the loss of G i␣2 and promote progression in the absence of RA (present study). Gain-of-function mutants with constitutively active G i␣2 show no activation of PLC, PKC, MAPK, and progression in response to RA (present study). Like the recent discovery of a critical role of G o in Caenorhabditis elegans behavior (44), the current study highlights a new dimension of G-protein function in development, one in which a G-protein (G i␣2 ) triggers stem cell progression via activation (derepression) of PLC, and thereby activation of PKC and the MAPK regulatory network.