• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Biol. Author manuscript; available in PMC Nov 15, 2010.
Published in final edited form as:
PMCID: PMC2783421
NIHMSID: NIHMS146077

The Gα4 G protein subunit interacts with the MAP kinase ERK2 using a D-motif that regulates developmental morphogenesis in Dictyostelium

Abstract

G protein Gα subunits contribute to the specificity of different signal transduction pathways in Dictyostelium discoideum but Gα subunit-effector interactions have not been previously identified. The requirement of the Dictyostelium Gα4 subunit for MAP kinase (MAPK) activation and the identification of a putative MAPK docking site (D-motif) in this subunit suggested a possible interaction between the Gα4 subunit and MAPKs. In vivo association of the Gα4 subunit and ERK2 was demonstrated by pull-down and co-immunoprecipitation assays. Alteration of the D-motif reduced Gα4 subunit-ERK2 interactions but only slightly altered MAPK activation in response to folate. Expression of the Gα4 subunit with the altered D-motif in gα4 cells allowed for slug formation but not the morphogenesis associated with culmination. Expression of this mutant Gα4 subunit was sufficient to rescue chemotactic movement to folate. Alteration of the D-motif also reduced the aggregation defect associated with constitutively active Gα4 subunits. These results suggest Gα4 subunit-MAPK interactions are necessary for developmental morphogenesis but not for chemotaxis to folate.

Keywords: signal transduction, G protein, MAP kinase, Dictyostelium, development, chemotaxis

Introduction

G protein-coupled receptors activate many different cellular responses in eukaryotes using a wide variety of signal transduction pathways (Milligan and Kostenis, 2006; Neves et al., 2002; Simon et al., 1991). Most known pathways include the activation of heterotrimeric G proteins (Gαβγ) that disassociate into a Gα subunit and a Gβγ dimer capable of regulating the function of downstream effectors (Hamm, 1998). The Gα subunit provides specificity for receptor coupling and when activated the Gα releases the Gβγ dimer (Conklin et al., 1993). In some signaling pathways, the interaction of the Gα subunit with a specific effector (e.g., adenylyl cyclase, phospholipase C, etc.) has been established but in most pathways the interactions the Gα subunit with downstream effectors remains to be determined (Neer and Clapham, 1988; Neves et al., 2002). Identifying Gα subunit interactions with other signaling proteins will likely provide important insights with respect to pathway specificity and cellular responses mediated by G proteins.

The genome of the soil amoebae Dictyostelium discoideum encodes 12 different Gα subunits (gpa genes) and some of these subunits have been shown to function in pathways that regulate chemotactic responses and a variety of processes associated with the developmental life cycle of this organism (Brandon et al., 1997; Brzostowski et al., 2002; Hadwiger and Firtel, 1992; Hadwiger et al., 1996; Kumagai et al., 1991). Genetic analysis has indicated the Gα4 subunit is required for chemotaxis to folate and the promotion of spore cell development during multicellular development whereas the closely related Gα5 and Gα2 subunits are required for different or even opposing functions (Hadwiger and Firtel, 1992; Hadwiger et al., 1994; Hadwiger et al., 1996; Kumagai et al., 1991; Natarajan et al., 2000). The Gα2 subunit is required for cAMP chemotaxis and cell aggregation and the Gα5 subunit inhibits folate chemotaxis and promotes prestalk cell development (Hadwiger et al., 1996; Kumagai et al., 1991). The Gα4 subunit and these other Gα subunits are presumed to couple to the same Gβγ dimer because only single genes have been identified for the Gβ and Gγ subunits in the Dictyostelium genome and so only the receptor and the Gα subunit are likely to be the determinants of pathway specificity (Lilly et al., 1993; Zhang et al., 2001). A previous study of chimeric Gα subunits suggests the Gα4 subunit and other Gα subunits can contribute to the specificity of downstream responses, perhaps through interactions with signaling components other than the Gβγ dimer (Hadwiger, 2007). However, Gα subunit interactions with effectors in Dictyostelium have not been previously described.

Downstream in many G protein-mediated signaling pathways is the activation of MAP kinases (MAPKs) that can phosphorylate both cytoplasmic and nuclear targets to regulate cell growth and differentiation (Caunt et al., 2006; Chen and Thorner, 2007; Goldsmith and Dhanasekaran, 2007). Dictyostelium has only two MAPKs, ERK1 and ERK2 that belong to a subclass of MAPKs known as extracellular signal-regulated kinases (Goldberg et al., 2006). ERK1 regulates cell aggregate size and the timing of gene expression during the development and ERK2 is necessary for cAMP-mediated cellular aggregation and prespore development (Gaskins et al., 1996; Gaskins et al., 1994; Segall et al., 1995; Sobko et al., 2002). In cAMP-stimulated cells, ERK2 is required for cAMP accumulation through the inhibition of the phosphodiesterase RegA (Maeda et al., 2004; Segall et al., 1995). MAPKs, including the Dictyostelium ERK1 and ERK2, contain a common docking (CD) site that allows them to associate with their activators, MAPK kinases (MAPKKs) and substrates (Tanoue et al., 2000). The CD sites contribute to interactions with MAPK docking sites (D motifs) on other proteins to tether and facilitate interactions (Grewal et al., 2006; Remenyi et al., 2005). In some G protein signaling pathways, MAPK activation occurs via the release of Gβγ dimers that transduce signals through Ras proteins and kinase cascades that include MAPKKs and MAPKK kinases (MAPKKKs) (Belcheva and Coscia, 2002; Chen and Thorner, 2007). This type of signal transduction pathway has been extensively studied in the yeast (Saccharomyces cerevisiae) mating response pathway and recently the Gα subunit, Gpa1, of this pathway has been shown to contain a D-motif that allows for direct interaction between the Gpa1 subunit and the MAPK Fus3 (Metodiev et al., 2002). These interactions are thought to regulate a small change in the cytoplasmic/nuclear distribution of Fus3 and promote adaptation to mating pheromone signaling (Blackwell et al., 2003).

In this study we characterized the role of the Gα4 subunit in MAPK function in Dictyostelium. Gα4 subunit-ERK2 interactions were analyzed using a pull-down assay in Dictyostelium. The primary structure of the Gα4 subunit was analyzed for D motifs in regions that might be exposed to interactions with other proteins. Gα4 mutants with an altered D-motif were tested for the ability to interact and activate MAPKs and the ability to rescue developmental morphology and chemotaxis of gα4 mutants. The results of this study support a role for Gα4 subunit-MAPK interactions in developmental processes such as cell differentiation.

Materials and methods

Strains and media

All Dictyostelium strains used in the study were derived from the wild-type strain KAx3. The gα4, Gα4HC, and JH8 strains were previously described (Hadwiger and Firtel, 1992). The erk2 strain was created in JH8 (pyr5-6) cells using a gene disruption construct previously described and generously provided by J. Segall (Albert Einstein College of Medicine, Bronx, NY) and the Dictyostelium Stock Center. The erk2 gene disruption was verified by genomic DNA blot analysis and the erk2 mutant displayed the phenotypes previously described for erk2 cells in other strain backgrounds (Segall et al., 1995). Cells were grown in axenic HL5 medium or on bacterial lawns of Klebsiella aerogenes (Watts and Ashworth, 1970). DNA vectors were electroporated into cells as previously described (Hadwiger, 2007). Transformed cells were selected and maintained in medium containing 3–10 μg/ml of the drug G418 or 3 μg/ml of the drug blasticidin S and drug selection was removed several hours prior to analysis. Folate solutions were prepared by neutralizing folic acid with NaHCO3. The modeling of the Gα4 subunit structure was conducted using the Swiss-Model program and displayed using the PyMOL Molecular Graphics System 2008 (DeLano Scientific) (Arnold et al., 2006).

Recombinant DNA constructs

The wild-type Gα4 gene (also designated as gpaD) in the vector pBluescriptIISK+ (Stratagene) was previously described (Hadwiger, 2007). Alterations to the Gα4 subunit D-motif, Gα4R107E,R108E subunit (designated Gα4d−), were created from the wild-type Gα4 cDNA in the vector pBluescriptIISK+ using the Gene Tailor PCR mutagenesis system (Invitrogen) and overlapping oligonucleotides: (sense strand: 5′-CAAAGAGCAGCAAATGTACTcgaggaAACTATTGGTAATGAACC) and (antisense strand: 5′-GTACATTTGCTGCTCTTTGTTTATTTTC) (lower case nucleotides differ from those of the template DNA). In addition to substituting two arginine codons with two glutamate codons, one silent mutation was introduced to create XhoI site thus allowing the mutant gene to be detected by restriction enzyme digestion analysis. A gene encoding a constitutively-active Gα4Q200L subunit (designated Gα4*) was also created from the wild-type Gα4 gene by PCR mutagenesis using overlapping oligonucleotides (sense strand: 5′-GATTAGATTAAAGATTGTAGAcGTCGGTGGTCtAAGATCTCAAAGAAGAAAATGG) and (antisense strand: 5′-CAGAATTTACATTTGATAAGATTAGATTAAAGATTGTAG). A silent mutation, creating an AatII site for restriction enzyme analysis verification, was also part of the PCR mutagenesis. This mutation conferring constitutive activity was also created in the gene encoding the Gα4d− subunit resulting in a gene encoding a Gα4d−,* subunit. All mutations created in this study were verified by sequence analysis and each mutant Gα4 gene was then inserted into the Dictyostelium expression vector pDXA-GFP2 (Ddp2-based plasmid), replacing the HindIII/XbaI fragment that contained the GFP2 reading frame downstream of the act15 promoter (Levi et al., 2000). The Ddp2-based vectors integrate into the genome unless the cell also contains the pREP vector. Each pact15/Gα4 gene construct was also transferred as SalI/XbaI fragments into the Ddp1-based pTX-GFP vector (extrachromosomal vector) replacing the pact15/GFP2 gene (Levi et al., 2000).

To create Myc-tagged wild-type and mutant Gα4 subunits the various Gα4 genes were modified with a 5′-BamHI and a 3′-XbaI restriction site using PCR and the oligonucleotides: (sense strand 5′-gccggcggatccATGAGATTCAAGTGTTTTGGATCAG) and (antisense strand 5′-cggcgctctagaTTAGAAGTGTTCTAATGCTTGAGATAAAATTGTTTGTCTAAC). Each amplified Gα4 gene was inserted at BamHI and XbaI sites into a pBluescriptIISK+ vector containing the HindIII-Myc-BamHI linker created from the oligonucleotides: (sense strand 5′-AGCTTATGGAACAAAAATTATTATCAGAAGAAGATTTAG) and (antisense strand 5′-GATCCTAAATCTTCTTCTGATAATAATTTTTGTTCCATA). This linker adds the amino acid sequence MEQKLLSEEDLGS to the amino terminus of each Gα4 gene (underlined residues represent the Myc epitope). Each gene was then transferred into the Dictyostelium expression vector pDXA-GFP2 as described above. A His6-tagged ERK2 was constructed by PCR amplification of an ERK2 (also designated as erkB) cDNA kindly provided by J. Segall using the oligonucleotides: (sense strand 5′-cgcaagcttggatccctcgagacacaATGTCATCTGAAGATATAGATAAACATG) and (antisense strand 5′ – GCGGTCGACTCTAGATTATGTTGATAAAGTTGGAGCAGTTGTACT). The amplified ERK2 gene was inserted into the TOPO vector (Invitrogen) and then transferred into XbaI and XhoI sites of the Dictyostelium expression vector pDXA-HC containing His6-tag (Manstein et al., 1995).

MAPK activation assay

Cells were grown to mid-log phase (approximately 2–3 × 106 cells/ml), washed twice in phosphate buffer (12mM NaH2PO4 adjusted to pH 6.1 with KOH), and suspended in phosphate buffer (5 × 107 cells/ml). The cell suspension was shaken for 1 hr and then stimulated with 50 μM folate. Cells were harvested by mixing with SDS-PAGE loading buffer and boiled. Samples (8 × 107 cells/lane) were subjected to SDS-PAGE and immunoblot analysis using a rabbit α-phospho-p44/p42 MAPK antibody and secondary goat HRP-conjugated α-rabbit IgG antibody for chemiluminescence detection (Cell Signaling Technology).

Development and chemotaxis assays

Cells were grown to mid-log phase (approximately 2–3 × 106 cells/ml), washed twice in phosphate buffer (12mM NaH2PO4 adjusted to pH 6.1 with KOH), and suspended in phosphate buffer (1 × 108 cells/ml), before spotting on nonnutrient plates (phosphate buffer, 1.5% agar) for development or chemotaxis as described (Hadwiger et al., 1996). Cell development was analyzed using a dissecting microscope or fluorescence microscopy. Chemotaxis assays were performed by spotting droplets of cell suspension (107 to 108 cells/ml) on nonnutrient plates followed by the spotting of 1μl of folate solutions (10−2 to 10−4 M) approximately 2–3 mm away from the cell droplet. Cell movement was monitored with a dissecting microscope. All strains were treated identically for each experiment.

Pull-down assays

Cells grown to 3×106 cells/ml were collected and washed in phosphate as described above. Pellets were suspended at 5×108 cell/ml in lysis buffer (300 mM NaCl, 50 mM sodium phosphate) containing a protease inhibitor cocktail (Sigma). Cells were lysed by freezing at −80°C for 30 min, thawed on ice, vortexed, and centrifuged at 14,000 rpm for 5 sec to eliminate insoluble materials. Cell extracts were incubated with Talon Co2+ resin (Clontech) for 20 min in ice with gentle shaking. Extract-treated resins were washed three times on spin columns with wash buffer (300 mM NaCl, 50 mM sodium phosphate, 10 mM imidazol and 10% glycerol) and finally eluted using an elution buffer (300 mM NaCl, 50 mM sodium phosphate, 150 mM imidazol). The eluates were subjected to SDS-PAGE and western blotting. Blots were incubated with the primary mouse α-Myc antibody and then a secondary goat HRP-conjugated α-mouse IgG before bioluminescence detection.

Immunoprecipitation assays

Cells were prepared and lysed as described above for the pull-down assays. Cell extracts were treated with Protein A/G-agarose (25% v/v) (Santa Cruz Biotechnology) for 30 min on ice for preclearing treatment according to the manufacturer’s instructions. The collected supernatants were incubated with mouse anti-Myc antibody for 1 hr and then with Protein A/G-Agarose for 1 hr on ice. The agarose was precipitated and washed 2 times with 1 ml of lysis buffer. The resulting immune complexes were subjected to SDS PAGE and western blotting. Blots were incubated with the primary rabbit α-His antibody and then a secondary goat HRP-conjugated α-rabbit IgG before bioluminescence detection.

Results

Gα4 function is required for MAPK activation

While several Gα subunit genes have been identified in Dictyostelium genome none of the encoded subunits have been shown to specifically interact with downstream signaling proteins that would define their role in pathway specific responses. Several reports have indicated a requirement for the Gα4 subunit for the activation of ERK2 in the response to folate whereas the Gα2 subunit is not required for ERK2 activation in response to cAMP suggesting the Gα4 subunit might directly regulated the activity of this MAPK (Maeda et al., 1996; Maeda and Firtel, 1997). Verification of the requirement of the Gα4 subunit in MAPK activation was examined through the detection of the phosphorylated MAPK in wild-type and gα4 cells (Fig. 1). Only a low level of MAPK activation was observed (band with approximate molecular weight of 42,000) after folate stimulation of gα4 cells compared to the level observed in wild-type cells. A slighty higher level of phosphorylated MAPK was detected in erk2 cells and this detected band might be residual ERK2 produced from the leakiness of the erk2 allele as suggested by others (Kosaka and Pears, 1997). This detected band might also correspond to phosphorylated ERK1 because it also contains the highly conserved TEY segment that is phosphorylated by MAPKKs and recognized by the α-phospho-p42/p44 antibody. Therefore, we chose to designate the detected band as pMAPKs (phosphorylated ERK1/ERK2) because the possible presence of phosphorylated ERK1. In addition, MAPK activation was also analyzed in Gα4HC cells that over-express the Gα4 gene due in a high copy number of Gα4 genomic sequence. The over-expression of the Gα4 subunit did not increase the level of phosphorylated MAPKs suggesting Gα4 subunit expression is not a limiting factor.

Figure 1
Folate stimulated MAPK activation in wild-type and mutant cells. Wild-type cells (WT), gα4 cells (gα4), and Gα4 subunit over-expression (high copy gene number) cells (Gα4HC), and erk2 cells ( ...

D-motif in the Gα4 subunit

The requirement for Gα4 subunit in the folate stimulation of ERK2 might be due to a physical interaction between the Gα4 subunit and ERK2 as a Gα subunit-MAPK interaction has been previously reported in the pheromone response pathway in yeast (Metodiev et al., 2002). To assess whether or not the Gα4 subunit might interact with ERK2, a search for potential D-motifs [K/R1-2 - spacer(1-6) - I/L-X-I/L] in the Gα4 subunit revealed a putative D-motif spanning residues 107–118 (Fig. 2A). The Gα4 subunit did not contain a D-motif near the amino terminus where the yeast Gpa1 D-motif exists and also where several other Dictyostelium Gα subunits have putative D-motifs. Only one other Dictyostelium Gα subunit, Gα5, had a potential D-motif in the same region as the Gα4 subunit and this motif had an unusually long spacer region compared to other known D-motifs. To assess whether the Gα4 D-motif at positions 107–118 was on the surface of the Gα4 subunit, a modeling program was used to predict a 3-dimensional model of the Gα4 subunit based on the crystallographic studies of another Gα subunit (Fig. 2B). The D-motif is predicted to be part of a loop on the surface of the Gα4 subunit located away from the predicted binding surfaces for receptors or the Gβγ dimer suggesting this D-motif is accessible to other proteins such as MAP kinases. Another potential D-motif was found at positions 25–37 in the Gα4 subunit but this region is not likely to function as a MAPK docking site because the region overlaps with the highly conserved G1 region that contributes to the pocket for guanine nucleotide binding (Conklin and Bourne, 1993). The D-motifs found in Dictyostelium Gα subunits can potentially interact with either ERK1 and ERK2 because both MAPKs contain conserved regions, such as the CD-motif, that have been implicated in D-motif interactions (Fig 2C) (Tanoue and Nishida, 2003). However other regions of MAPKs are also known to contribute to D-motif binding.

Figure 2
(A) Alignments of D-motifs in Gα subunits. Primary sequence of the Dictyostelium Gα4 (DdGα4), Gα5 (DdGα5), Gα2 (DdGα2), yeast Gpa1 (ScGpaI), and human Gα16 (HsGα16) Gα subunits ...

Analysis of Gα4-ERK2 interaction in vivo

Possible Gα4 subunit-ERK2 interactions in Dictyostelium were investigated using a pull-down assay that examined the presence of Myc-tagged Gα4 subunit in the extraction of a His6-tagged ERK2 protein. A vector expressing a Myc-tagged Gα4 subunit (Myc-Gα4) was transformed into an erk2 strain already containing an integrated expression vector for a His6-tagged ERK2 (His6-ERK2). In these cells, the His6-ERK2 was sufficient to allow for developmental aggregation indicating the His6-tag does not destroy ERK2 function (data not shown). The His6-ERK2 kinase was pulled down from cell lysates using Co2+ resin and washed and the eluate from the resin was subjected to immunoblot analysis for the presence of the Myc-Gα4 subunit. The level of the Myc-Gα4 subunit detected in eluates was approximately 2-fold greater than that found in 1/10 of the total cell lysate (~22% yield) indicating that the Gα4 subunit and ERK2 are capable of interacting in vivo (Fig. 2D). To determine if the Myc-Gα4 subunit in the pull-down eluates was dependent on the presence of His6-ERK2, the Myc-Gα4 subunit was also expressed in erk2 cells that expressed ERK2 without the His6-tag. Using the same Co2+ resin pull-down conditions, no Myc-Gα4 subunit was detected in the eluate indicating the Myc-Gα4 association with the pull-down assay is dependent on the His6-ERK2.

The Gα4 subunit-ERK2 interaction was also demonstrated by the detection of the His6-ERK2 in immunoprecipitates of myc-Gα4 subunit. In erk2 cells transformed with the His6-ERK2 and myc-Gα4 subunit expression vectors, the Myc-Gα4 subunits were immunoprecipitated using α-Myc antibodies and the resulting immune complexes contained the His6-ERK2 (Fig. 2D). The His6-ERK2 was not detected in cells expressing the untagged Gα4 subunit indicating that the presence of the His6-ERK2 in the immunoprecipitations was dependent on the Myc-Gα4 subunit. The ability to detect ERK2 in Gα4 subunit-immunoprecipitates provides further support for the association of these proteins in vivo.

Alteration of the D-motif reduces Gα4 subunit association with ERK2

A study of the D-motif in the yeast Gpa1 subunit has indicated that the replacement of the positively charged residues with negatively charged residues reduces the Gpa1 subunit interactions with the MAP kinase Fus3 and reduces the ability of the Gpa1 subunit to promote cellular adaptation to mating pheromone (Metodiev et al., 2002). Therefore a similar alteration was created in the Gα4 subunit D-motif through the replacement of the two arginine residues with glutamate residues (Gα4R107E,R108E referred to as Gα4d−) and the Myc-tagged version of this Gα4d− subunit was assessed for associations with ERK2 using the previously described His6-ERK2 pull-down assay. The level of Gα4d− subunit in the eluates was nearly equal that found in 1/10th of the total cell lysate (~9% yield) indicating a large reduction (~ 2.5-fold) in the association with His6-ERK2 compared to the wild-type Gα4 subunit (Fig. 3). This result suggests the D-motif alteration impairs but does not eliminate the interaction between the Gα4 subunit and ERK2.

Figure 3
Association of mutant Gα4 subunits with the ERK2 kinase in Dictyostelium. erk2 cells expressing either the His6-ERK2 or ERK2 were transformed with a vector expressing either the Myc-Gα4 subunit, Myc-Gα4d− subunit, ...

In yeast the Gpa1 subunit-Fus3 interaction is enhanced by the activated state of the Gpa1 subunit and so a constitutively active Gα4Q200L subunit (referred to as Gα4*) was also tested in the His6-ERK2 pull-down assay. The level of Myc-Gα4* subunit in the eluates were similar to that found in 1/10th of the total cell lysate (~9% yield) indicating a reduced association with the His6-ERK2 kinase compared to that of the Myc-Gα4 subunit (Fig. 3). This 3-fold difference in yield suggests the activation of the Gα4 subunit might reduce interactions with ERK2. To determine if the association of the constitutively active Gα4* is dependent on the D-motif, a mutant Gα4R107E, R108E, Q200L subunit (referred to as Gα4d−,*) containing both the docking site alteration and constitutively active mutation was tested in the His6-ERK2 pull-down assay. The levels of the Myc-tagged Gα4d−,* subunit associated with the eluates were nearly 3-fold less than that found in 1/10th of the total cell lysate (~4% yield) indicating a dependence on the D-motif. In all assays, the association of the mutant Myc-Gα4 subunits with the eluates was dependent on the His6-tagged ERK2.

Alteration of the D-motif in the Gα4 subunit affects MAPK activation

The Gα4 subunit-ERK2 interaction might be important for the activation of ERK2 because of the requirement for Gα4 subunit function in this process. The stimulation of gα4 cells expressing the Gα4d− subunit with folate resulted in MAPK activation but the level of pMAPKs was slightly reduced compared to that observed for gα4 cells expressing the wild-type Gα4 subunit (Fig. 4). This slight reduction in MAPK activation suggests the Gα4 D-motif only provides a minor contribution to the activated state of MAPKs, possibly through the interaction with ERK2. The expression of the Gα4* subunit or the Gα4d−,* subunit in gα4 cells also allowed for MAPK activation but the activation was significantly reduced in both strains compared to cells expressing the wild-type Gα4 subunit. The duration of MAPK activation was also noticeably shorter in cells expressing the Gα4* or Gα4d−,* subunit as the level of activated MAPK dropped considerably before the 2 min time point in these cells implying a possible reduction of MAPKK activity or increase in phosphatase activity.

Figure 4
Folate stimulated MAPK activation in Gα4 D-motif mutants. gα4 cells expressing the Gα4 subunit vector (Gα4), Gα4d− vector (Gα4d−), Gα4* vector (Gα4*), Gα4d ...

Alteration of the D-motif in the Gα4 subunit affects development

To assess the function of the Gα4 D-motif during development, gα4 cells expressing the Gα4d− subunit or other Gα4 subunits were examined during development on nonnutrient agar. Starved gα4 cells aggregate and form anterior tips composed of prestalk cells but these tips extend from the mound leaving behind most prespore cells (Fig. 5A). As one tips moves away from the mound, subsequent tips can develop and repeat the process. Prespore cells are typical scattered throughout the mound but some do move with the anterior prestalk cells (Hadwiger and Firtel, 1992; Hadwiger et al., 1994; Srinivasan et al., 1999). The expression of the wild-type Gα4 subunit in gα4 cells allows for wild-type development with slug migration and mature fruiting body formation (i.e., spore mass supported by a stalk) indicating a complete rescue of development. The gα4 cells expressing the Gα4d− subunit aggregated into mounds but this process often involved ring-shaped intermediates as the loose aggregates organized into more compact mounds. Similar aggregate reorganization is also observed in wild-type cells but this process was much more apparent in the aggregates expressing the Gα4d− subunit. The Gα4d− subunit expressing mounds developed into finger-like structures or slugs but these structures did not undergo culmination to form fruiting bodies. Rather, these structures developed a bulge near the anterior end of the slug where spores eventually formed. However the amount of spores observed by visual inspection in these structures was typically much lower than the spore production in a fruiting body of wild-type cells (Fig. 5C). The anterior tips of the Gα4d− expressing structures remained intact but often drooped over as the bulge of spores formed. Wild-type cells expressing the Gα4d− subunit displayed normal development with fruiting body formation indicating that Gα4d− subunit does not affect the development of cells expressing a wild-type Gα4 subunit (Fig. 5B). Interestingly, gα4 aggregates without any Gα4 subunit expression vector can achieve structures similar to the Gα4d− aggregates when development occurs after cells have grown on bacterial lawns or in shaking bacterial cultures (data not shown). The basis of this conditional developmental morphology of gα4 cells is not known but the development of gα4 cells expressing the Gα4d− subunit does not appear to be affected by whether cells are grown in axenic medium or on bacteria (data not shown).

Figure 5
Developmental phenotypes of Gα4 D-motif mutants. Cells were grown and plated for development as described in the Materials and Methods section. (A) Development of wild-type cells (WT), gα4 cells without an expression vector (none) ...

Expression of the constitutively active Gα4* subunit in gα4 cells resulted in impaired aggregation and so development did not proceed beyond loosely-formed small aggregates (Fig. 5A). This phenotype was similar to the aggregation defect described for Gα4* subunit expression from the Gα4 promoter and this aggregation deficiency was previously attributed to the inhibition of cAMP chemotaxis (Srinivasan et al., 1999). Similar aggregation defects have also been observed for other constitutively active Gα subunits, including Gα2* and Gα5* subunits. The expression of the Gα4d−,* subunit in gα4 cells did not inhibit aggregation as did the Gα4* subunit. Rather, the Gα4d−,* subunit expressing cells formed aggregates that produced extended tip structures similar to that of gα4 cells. The ability of these cells to aggregate suggests that the D-motif alteration removes the inhibition of aggregation associated with the constitutively active Gα4* subunit.

Development of cells expressing the Gα4d− subunit can be rescued by increased ERK2 or Gα4d− subunit expression

Protein-protein interactions that are compromised due to changes in protein structure can often be enhanced through increasing the concentration of one or both proteins (Hadwiger et al., 1989a; Hadwiger et al., 1989b). Consistent with this idea, increased selection (> 10 μg/ml G418) for the Gα4d− subunit expression vector in gα4 cells allowed for fruiting body development (data not shown). An increase in ERK2 expression might also stabilize Gα4d− subunit-MAPK interactions and so to test this idea gα4 cells expressing the Gα4d− subunit without G418 selection were co-transformed with an ERK2 expression vector and a blasticidin S resistance vector. Transformants were selected using only blasticidin S to avoid increasing Gα4d− subunit expression. Multiple clones were found that could complete the fruiting body process suggesting that ERK2 over-expression can rescue developmental morphogenesis (Fig. 5B). The rescue of developmental morphogenesis with increased Gα4d− subunit or ERK2 expression implies that the developmental defects associated with the Gα4d− subunit are likely due to impaired Gα4 subunit-MAPK interactions.

Defects in Gα4d− stalk development are not cell autonomous

The inability of gα4 cells expressing the Gα4d− subunit to undergo fruiting body development suggests possible defects in both spore and stalk cell development. However, a previous analysis of gα4 cells has indicated that Gα4 function is necessary primarily for spore development and that the lack of stalk development is due to the lack of extracellular signaling (Hadwiger and Firtel, 1992; Hadwiger and Srinivasan, 1999). To determine if gα4 cells expressing the Gα4d− subunit can form stalks, these cells were mixed with Gα4HC cells expressing GFP to form chimeric aggregates. Previous reports have shown that Gα4HC cells are capable of forming spores but not stalk cells in chimeras with wild-type or gα4 cells (Hadwiger and Srinivasan, 1999). The chimeric aggregates of Gα4d− subunit expressing cells and Gα4HC cells were capable of forming fruiting bodies but none of the Gα4HC cells were found in the stalk regions indicating that the stalks are be composed of Gα4d− subunit expressing cells (Fig 5D). This result suggests the absence of stalk formation in aggregates of gα4 cells expressing the Gα4d− subunit is due to the lack of an intercellular signal that can be provided by cells with a wild-type Gα4 subunit.

Stalk formation was also observed in chimeras of gα4 cells expressing the Gα4d− subunit mixed with wild-type cells. Either the Gα4d− expressing cells or the wild-type cells were labeled using a GFP expression vector and the stalks of the resulting chimeric fruiting bodies were examined for fluorescent cells. Both wild-type and Gα4d− expressing cells were found in the stalks and capable of developing into vacuolated stalk cells in which the GFP was excluded from the center of the cell (Fig. 5E). The stalks of these chimeras typically contained more wild-type cells than Gα4d− expressing cells even though the chimeric aggregates contained equal amounts of both strains implying that Gα4d− expressing cells might have a reduced preference for stalk cell development.

Chemotaxis to folate

Previous studies have shown the Gα4 subunit is essential for chemotactic responses to folate and related pterin compounds allowing cells to search out nearby bacterial food sources (Hadwiger et al., 1994). MAPK function has also been implicated in folate chemotaxis as erk2 cells have been reported to have reduced chemotactic responses to folate (Wang et al., 1998). To determine if the Gα4 D-motif is important for chemotactic responses, gα4 cells expressing the Gα4 or Gα4d− subunit were analyzed for the ability to chemotax to folate in an above agar assay. The expression of the Gα4d− subunit in gα4 cells restored folate chemotaxis similar to that observed for cells expressing the Gα4 subunit (Fig. 6A–C). The analysis of individual cell movement using time-lapse photography also suggested Gα4d− expressing cells move similar to those cells expressing the Gα4 subunit (Fig. 6D). The chemotaxis index of cells expressing the Gα4d− was also as high as that observed for cells expressing the Gα4 subunit but less than that for wild-type cells (Fig. 6E).

Figure 6
Chemotaxis of Gα4 D-motif mutants. Wild-type cells (WT), gα4 cells without an expression vector (none) and gα4 cells with the Gα4 (Gα4), Gα4d− (Gα4d−), Gα4* ...

The chemotaxis of cells expressing the Gα4* subunit to folate was significantly less than that observed for cells expressing the wild-type Gα4 subunit (Fig. 6A–C). However, expression of the Gα4* subunit increased the overall movement of cells compared to gα4 cells and the direction of cell movement appeared to be random as indicated by the chemotaxis index (Fig. 6E). This inability to direct cell movement up the folate gradient suggests the Gα4* subunit might affect the reception of the chemotactic signal or interfere with the ability of the cell to establish directed movement up the chemotacic gradient. However, the nearly equal movement of cells regardless of the proximity to the source of folate implies that cell movement does not depend on the folate stimulus. A similar chemotaxis phenotype was observed for gα4 cells expressing the Gα4d−,* subunit suggesting the D-motif alteration does not change the cell movement phenotype associated with the Gα4* subunit (Fig. 6A&B).

Discussion

Stimulated G protein-coupled receptors are known to activate MAPKs through transduction pathways that involve Gβγ dimers and Ras proteins and this mechanism also exists in Dictyostelium as the Gβ subunit and rasD protein can be necessary for MAPK activation (Belcheva and Coscia, 2002; Chen and Thorner, 2007; Knetsch et al., 1996; Maeda and Firtel, 1997). In response to folate, the Gα4 subunit presumably triggers MAPK activation through the release of the Gβγ dimer as suggested by the requirement of Gβ for this activation. Our results indicate the Gα4 subunit can also have a minor effect MAPK activation through direct interactions with a MAPK using a D-motif. The contribution of the Gα4 subunit D-motif to MAPK activation is difficult to assess because the Gα4d− alteration only reduced but did not completely eliminate Gα4 subunit-ERK2 interactions. The recruitment of a MAPK to the Gα4 subunit D-motif could localize the MAPK to a region with enhanced MAPKK activity or reduced phosphatase activity. While the relationship of Gα4 subunit-MAPK binding to MAPK activation is not completely resolved, it is clear that alterations in the Gα4 subunit D-motif can have important developmental consequences with respect to cell differentiation and morphogenesis.

The location of the D-motif in the Gα4 subunit is unusual compared to most other Dictyostelium, yeast, and human Gα subunits analyzed for D-motifs but this feature might be associated with the ability of the Gα4 subunit to facilitate MAPK activation whereas other Gα subunits, such as the Dictyostelium Gα2 subunit or the yeast Gpa1, are not required for MAPK activation. The amino terminal location of D-motifs found in other Gα subunits might prevent access to MAPK docking unless the Gα subunit is activated, as previously demonstrated for the yeast Gpa1 subunit-Fus3 interaction, and this restriction might limit the possible functions associated with the interaction (Metodiev et al., 2002). The availability of the D-motif, even when the Gα4 subunit is associated with the receptor and Gβγ dimer, could potentially localize or position a MAPK prior to the stimulation of the pathway. In support of this idea, cells expressing the constitutively active Gα4* subunit had reduced ERK2 interactions and MAPK activation compared to cells expressing the wild-type Gα4 subunit. The Gα4 subunit, like many Gα subunits, has a cysteine residue near the amino terminus that is likely to be palmitoylated as a mechanism to keep the Gα4 subunit anchored in the membrane and so an interactive MAPK would likely be tethered near the membrane (Wedegaertner, 1998). The Gα4 D-motif is predicted to be distal to the membrane surface whereas the amino terminal D-motifs of other Gα subunits are located proximal to the membrane surface. Such differences in D-motif location could impact MAPK orientation with respect to other regulatory proteins (e.g., MAPKKs and phosphatases) or MAPK substrates at the membrane surface. Alternatively, all of the Gα subunits with D-motifs might serve a common function such as restricting the movement of MAPK into the nucleus.

The gα4 cells expressing the Gα4d− subunit, like gα4 cells, have reduced spore production and deficiencies in signals for stalk formation suggesting that Gα4 subunit-MAPK interactions are important for these developmental processes. ERK2 is a likely candidate for these interactions because previous chimera development and gene expression studies indicate ERK2 is necessary for prespore development (Gaskins et al., 1996). As for the production of extracellular signals for stalk formation, a recent report has inferred Gα4 function in the production of extracellular signals such as GABA and SDF-2 necessary for spore development and so perhaps these or other signals also coordinate stalk formation during the final stages of spore development (Anjard et al., 2009). The developmental defects observed for Gα4d− expressing aggregates are not likely due to insufficient of Gα4d− subunit expression because the developmental defects occur with cells that are completely rescued in folate chemotaxis. Rather, defects in development might reflect specific Gα4 functions that depend on MAPK interactions.

One major developmental distinction between gα4 cells expressing the Gα4d− subunit compared to gα4 cells is that a greater proportion of the mound moves with the anterior tip, allowing the formation of a slug rather than just a tip extension. The mechanism for the movement of cells in the central and posterior regions of aggregates might share some similarities to the chemotactic movement of cells to folate since neither movement requires Gα4 subunit-MAPK interactions. The lack of a Gα4-MAPK interaction for cell movement was not expected because an earlier report indicated erk2 cells have reduced folate chemotaxis (Wang et al., 1998). We have confirmed this defect in recently created erk2 strains but found only a slight reduction in folate chemotaxis using the same assay described in the Materials and methods section (Hadwiger, unpublished data). We have also examined erk1 cells and found these cells also chemotax to folate suggesting neither of the MAPKs is essential for this response. Therefore, the ability of gα4 cells expressing the Gα4d− subunit to chemotax to folate or to follow the anterior tip in multicellular development is consistent with MAPKs playing only a minor role in chemotactic responses to folate.

The expression of the constitutively active Gα4* subunit results in an aggregation defect that disappears if the D-motif is altered suggesting the defect in aggregation requires Gα4 subunit-MAPK interaction. Previous studies have shown that the Gα4* subunit inhibits cAMP accumulation and ERK2 increases cAMP accumulation in cells stimulated by extracellular cAMP (Maeda et al., 2004; Segall et al., 1995; Srinivasan et al., 1999). These observations suggest the Gα4* subunit impacts cell aggregation and cAMP accumulation possibly through interactions with ERK2. Our results indicate the Gα4* subunit has a relatively weak association with ERK2 but this association is further reduced by alterations in the D-motif suggesting that the Gα4* subunit could possibly inhibit ERK2 to limit the accumulation of cAMP during cell aggregation. Constitutively active Gα2* and Gα5* subunits also interfere with cell aggregation and both of these subunits contain putative D-motifs at their amino terminus (Okaichi et al., 1992; Srinivasan et al., 1999). Alteration of the amino terminal D-motif reduces the aggregation defect associated with the constitutively active Gα5* subunit (Raisley, et al., unpublished data).

Both folate and cAMP stimulate MAPK activation in Dictyostelium but many distinctions are emerging as these pathways become more characterized. Only folate-stimulated MAPK activation requires both the Gα4 and Gβ subunit function whereas neither the Gα2 or Gβ subunit are required in the response to cAMP stimulation (Brzostowski and Kimmel, 2006; Maeda et al., 1996; Maeda and Firtel, 1997). These results are surprising because the Gα2 and Gβ subunits have been shown essential for cAMP chemotactic responses and none of the other 11 Gα subunits have been directly associated with cAMP receptor. Even though the mechanism for cAMP-stimulated MAPK activation does not require the Gα2 subunit, the putative D-motif near the amino terminus of this subunit could possibly tether MAPKs and regulate downstream responses to cAMP. Defining the effect of potential Gα2 subunit-MAPK interactions on MAPK regulation and function will likely provide additional insights into the distinctions between folate and cAMP-stimulated MAPK regulation.

Acknowledgments

The author thanks B. Raisley for creation of His6-tagged ERK2 expression strain and Ruth Weidman, Meg Finley, and Nicole Clarkson for technical assistance. The research was supported by an NIH grant (R15 GM073698-01) awarded to J.A.H.

Abbreviations

PCR
polymerase chain reaction
MAP kinase
mitogen-activated protein kinase
MAPKK
MAP kinase kinase
ERK
extracellular-signal regulated kinase
CD
common docking site
D-motif
MAPK docking site
SDS
sodium dodecyl sulfate
cAMP
cyclic adenosine monophosphate

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • Anjard C, Su Y, Loomis WF. Steroids initiate a signaling cascade that triggers rapid sporulation in Dictyostelium. Development (Cambridge, England) 2009;136:803–12. [PMC free article] [PubMed]
  • Arnold K, Bordoli L, Kopp J, Schwede T. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics (Oxford, England) 2006;22:195–201. [PubMed]
  • Belcheva MM, Coscia CJ. Diversity of G protein-coupled receptor signaling pathways to ERK/MAP kinase. Neurosignals. 2002;11:34–44. [PMC free article] [PubMed]
  • Blackwell E, Halatek IM, Kim HJ, Ellicott AT, Obukhov AA, Stone DE. Effect of the pheromone-responsive G(alpha) and phosphatase proteins of Saccharomyces cerevisiae on the subcellular localization of the Fus3 mitogen-activated protein kinase. Mol Cell Biol. 2003;23:1135–50. [PMC free article] [PubMed]
  • Brandon MA, Voglmaier S, Siddiqi AA. Molecular characterization of a Dictyostelium G-protein alpha-subunit required for development. Gene. 1997;200:99–105. [PubMed]
  • Brzostowski JA, Johnson C, Kimmel AR. Galpha-mediated inhibition of developmental signal response. Current biology. 2002;12:1199–208. [PubMed]
  • Brzostowski JA, Kimmel AR. Nonadaptive regulation of ERK2 in Dictyostelium: implications for mechanisms of cAMP relay. Molecular biology of the cell. 2006;17:4220–7. [PMC free article] [PubMed]
  • Caunt CJ, Finch AR, Sedgley KR, McArdle CA. Seven-transmembrane receptor signalling and ERK compartmentalization. Trends in endocrinology and metabolism: TEM. 2006;17:276–83. [PubMed]
  • Chen RE, Thorner J. Function and regulation in MAPK signaling pathways: lessons learned from the yeast Saccharomyces cerevisiae. Biochimica et biophysica acta. 2007;1773:1311–40. [PMC free article] [PubMed]
  • Conklin BR, Bourne HR. Structural elements of G alpha subunits that interact with G beta gamma, receptors, and effectors. Cell. 1993;73:631–41. [PubMed]
  • Conklin BR, Farfel Z, Lustig KD, Julius D, Bourne HR. Substitution of three amino acids switches receptor specificity of Gq alpha to that of Gi alpha. Nature. 1993;363:274–6. [PubMed]
  • Gaskins C, Clark AM, Aubry L, Segall JE, Firtel RA. The Dictyostelium MAP kinase ERK2 regulates multiple, independent developmental pathways. Genes Dev. 1996;10:118–28. [PubMed]
  • Gaskins C, Maeda M, Firtel RA. Identification and functional analysis of a developmentally regulated extracellular signal-regulated kinase gene in Dictyostelium discoideum. Molecular and cellular biology. 1994;14:6996–7012. [PMC free article] [PubMed]
  • Goldberg JM, Manning G, Liu A, Fey P, Pilcher KE, Xu Y, Smith JL. The Dictyostelium kinome--analysis of the protein kinases from a simple model organism. PLoS genetics. 2006;2:e38. [PMC free article] [PubMed]
  • Goldsmith ZG, Dhanasekaran DN. G protein regulation of MAPK networks. Oncogene. 2007;26:3122–42. [PubMed]
  • Grewal S, Molina DM, Bardwell L. Mitogen-activated protein kinase (MAPK)-docking sites in MAPK kinases function as tethers that are crucial for MAPK regulation in vivo. Cellular signalling. 2006;18:123–34. [PMC free article] [PubMed]
  • Hadwiger JA. Developmental morphology and chemotactic responses are dependent on G alpha subunit specificity in Dictyostelium. Developmental biology. 2007;312:1–12. [PMC free article] [PubMed]
  • Hadwiger JA, Firtel RA. Analysis of Gα4, a G-protein subunit required for multicellular development in Dictyostelium. Genes Dev. 1992;6:38–49. [PubMed]
  • Hadwiger JA, Lee S, Firtel RA. The G alpha subunit G alpha 4 couples to pterin receptors and identifies a signaling pathway that is essential for multicellular development in Dictyostelium. Proc Natl Acad Sci U S A. 1994;91:10566–70. [PMC free article] [PubMed]
  • Hadwiger JA, Srinivasan J. Folic acid stimulation of the Galpha4 G protein-mediated signal transduction pathway inhibits anterior prestalk cell development in Dictyostelium. Differentiation. 1999;64:195–204. [PubMed]
  • Hadwiger JA, Wittenberg C, Mendenhall MD, Reed SI. The Saccharomyces cerevisiae CKS1 gene, a homolog of the Schizosaccharomyces pombe suc1+ gene, encodes a subunit of the Cdc28 protein kinase complex. Mol Cell Biol. 1989a;9:2034–41. [PMC free article] [PubMed]
  • Hadwiger JA, Wittenberg C, Richardson HE, de Barros Lopes M, Reed SI. A family of cyclin homologs that control the G1 phase in yeast. Proc Natl Acad Sci U S A. 1989b;86:6255–9. [PMC free article] [PubMed]
  • Hadwiger JH, Natarajan K, Firtel RA. Mutations in the Dictyostelium heterotrimeric G protein α subunit Gα5 alter the kinetics of tip morphogenesis. Development. 1996;122:1215–1224. [PubMed]
  • Hamm HE. The many faces of G protein signaling. The Journal Of Biological Chemistry. 1998;273:669–72. [PubMed]
  • Knetsch ML, Epskamp SJ, Schenk PW, Wang Y, Segall JE, Snaar-Jagalska BE. Dual role of cAMP and involvement of both G-proteins and ras in regulation of ERK2 in Dictyostelium discoideum. The EMBO journal. 1996;15:3361–8. [PMC free article] [PubMed]
  • Kosaka C, Pears CJ. Chemoattractants induce tyrosine phosphorylation of ERK2 in Dictyostelium discoideum by diverse signalling pathways. The Biochemical journal. 1997;324(Pt 1):347–52. [PMC free article] [PubMed]
  • Kumagai A, Hadwiger JA, Pupillo M, Firtel RA. Molecular genetic analysis of two G alpha protein subunits in Dictyostelium. J Biol Chem. 1991;266:1220–8. [PubMed]
  • Levi S, Polyakov M, Egelhoff TT. Green fluorescent protein and epitope tag fusion vectors for Dictyostelium discoideum. Plasmid. 2000;44:231–8. [PubMed]
  • Lilly P, Wu L, Welker DL, Devreotes PN. A G-protein beta-subunit is essential for Dictyostelium development. Genes Dev. 1993;7:986–95. [PubMed]
  • Maeda M, Aubry L, Insall R, Gaskins C, Devreotes PN, Firtel RA. Seven helix chemoattractant receptors transiently stimulate mitogen-activated protein kinase in Dictyostelium - Role of heterotrimeric G proteins. J Biol Chem. 1996;271:3351–3354. [PubMed]
  • Maeda M, Firtel RA. Activation of the mitogen-activated protein kinase ERK2 by the chemoattractant folic acid in Dictyostelium. J Biol Chem. 1997;272:23690–5. [PubMed]
  • Maeda M, Lu S, Shaulsky G, Miyazaki Y, Kuwayama H, Tanaka Y, Kuspa A, Loomis WF. Periodic signaling controlled by an oscillatory circuit that includes protein kinases ERK2 and PKA. Science (New York, NY) 2004;304:875–8. [PubMed]
  • Manstein DJ, Schuster HP, Morandini P, Hunt DM. Cloning vectors for the production of proteins in Dictyostelium discoideum. Gene. 1995;162:129–34. [PubMed]
  • Metodiev MV, Matheos D, Rose MD, Stone DE. Regulation of MAPK function by direct interaction with the mating-specific Galpha in yeast. Science. 2002;296:1483–6. [PubMed]
  • Milligan G, Kostenis E. Heterotrimeric G-proteins: a short history. British journal of pharmacology. 2006;147(Suppl 1):S46–55. [PMC free article] [PubMed]
  • Natarajan K, Ashley CA, Hadwiger JA. Related Gα subunits play opposing roles during Dictyostelium development. Differentiation. 2000;66:136–46. [PubMed]
  • Neer EJ, Clapham DE. Roles of G protein subunits in transmembrane signalling. Nature. 1988;333:129–34. [PubMed]
  • Neves SR, Ram PT, Iyengar R. G protein pathways. Science. 2002;296:1636–9. [PubMed]
  • Okaichi K, Cubitt AB, Pitt GS, Firtel RA. Amino acid substitutions in the Dictyostelium Galpha subunit Galpha2 produce dominant negative phenotypes and inhibit the activation of adenylyl cyclase, guanylyl cyclase, and phospholipase-C. Mol Biol Cell. 1992;3:735–747. [PMC free article] [PubMed]
  • Remenyi A, Good MC, Bhattacharyya RP, Lim WA. The role of docking interactions in mediating signaling input, output, and discrimination in the yeast MAPK network. Molecular cell. 2005;20:951–62. [PubMed]
  • Segall JE, Kuspa A, Shaulsky G, Ecke M, Maeda M, Gaskins C, Firtel RA, Loomis WF. A MAP kinase necessary for receptor-mediated activation of adenylyl cyclase in Dictyostelium. J Cell Biol. 1995;128:405–13. [PMC free article] [PubMed]
  • Simon MI, Strathmann MP, Gautam N. Diversity of G proteins in signal transduction. Science. 1991;252:802–8. [PubMed]
  • Sobko A, Ma H, Firtel RA. Regulated SUMOylation and ubiquitination of DdMEK1 is required for proper chemotaxis. Developmental cell. 2002;2:745–56. [PubMed]
  • Srinivasan J, Gundersen RE, Hadwiger JA. Activated Galpha Subunits Can Inhibit Multiple Signal Transduction Pathways during Dictyostelium Development. Dev Biol. 1999;215:443–452. [Record as supplied by publisher] [PubMed]
  • Tanoue T, Adachi M, Moriguchi T, Nishida E. A conserved docking motif in MAP kinases common to substrates, activators and regulators. Nature cell biology. 2000;2:110–6. [PubMed]
  • Tanoue T, Nishida E. Molecular recognitions in the MAP kinase cascades. Cellular signalling. 2003;15:455–62. [PubMed]
  • Wang Y, Liu J, Segall JE. MAP kinase function in amoeboid chemotaxis. J Cell Sci. 1998;111:373–83. [PubMed]
  • Watts DJ, Ashworth JM. Growth of myxameobae of the cellular slime mould Dictyostelium discoideum in axenic culture. The Biochemical journal. 1970;119:171–4. [PMC free article] [PubMed]
  • Wedegaertner PB. Lipid modifications and membrane targeting of G alpha. Biological signals and receptors. 1998;7:125–35. [PubMed]
  • Zhang N, Long Y, Devreotes PN. Ggamma in Dictyostelium: its role in localization of gbetagamma to the membrane is required for chemotaxis in shallow gradients. Molecular Biology Of The Cell. 2001;12:3204–13. [PMC free article] [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • Pathways + GO
    Pathways + GO
    Pathways, annotations and biological systems (BioSystems) that cite the current article.
  • PubMed
    PubMed
    PubMed citations for these articles
  • Substance
    Substance
    PubChem Substance links
  • Taxonomy
    Taxonomy
    Related taxonomy entry
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

    Your browsing activity is empty.

    Activity recording is turned off.

    Turn recording back on

    See more...