MAPK docking motif in the Dictyostelium Gα2 subunit is required for aggregation and transcription factor translocation

Abstract

Some G protein alpha subunits contain a mitogen-activated protein kinase (MAPK) docking motif (D-motif) near the amino terminus that can impact cellular responses to external signals. The Dictyostelium Gα2 G protein subunit is required for chemotaxis to cAMP during the onset of multicellular development and this subunit contains a putative D-motif near the amino terminus. The Gα2 subunit D-motif was altered to examine its potential role in chemotaxis and multicellular development. In gα2⁻ cells the expression of the D-motif mutant (Gα2^D−) or wild-type subunit from high copy number vectors rescued cell aggregation but blocked the transition of mounds into slugs. This phenotype was also observed in parental strains with a wild-type gα2 locus indicating that the heterologous Gα2 subunit expression interferes with multicellular morphogenesis. Expression of the Gα2^D− subunit from a low copy number vectors in gα2⁻ cells did not rescue aggregation whereas the wild-type Gα2 subunit rescued aggregation efficiently and allowed wild-type morphological development. The Gα2^D− and Gα2 subunit were both capable of restoring comparable levels of cAMP stimulated motility and the ability to co-aggregate with wild-type cells implying that the aggregation defect of Gα2^D− expressing cells is due to insufficient intercellular signaling. Expression of the Gα2 subunit but not the Gα2^D− subunit fully restored the ability of cAMP to stimulate the translocation of the GtaC transcription factor suggesting the D-motif is important for transcription factor regulation. These results suggest that the D-motif of Gα2 plays a role in aggregation and other developmental responses involved with cAMP signaling.

1. Introduction

Eukaryotic cells sense many environmental signals through G protein-coupled receptors (GPCRs) and these receptors stimulate many downstream responses including changes in metabolism, gene expression, cell differentiation, and cell movement [1–7]. Signaling proteins commonly found downstream of G proteins include nucleotide cyclases, phosphodiesterases, kinases, and transcription factors and these proteins are often shared between different pathways, suggesting that signaling complexes can be important for providing mechanisms of pathway specificity [8, 9]. Interactions between signaling proteins can be directly mediated through docking sites or indirectly through scaffolding proteins. Heterotrimeric G protein Gα subunits can recruit Gβγ dimers to specific receptors and then release these dimers when activated by the receptor and some Gα subunits also interact directly with downstream effectors to initiate signaling [10, 11]. Some Gα subunits have been reported to interact with mitogen-activated protein kinases (MAPKs) in yeast and amoeba but these interactions are not necessary for MAPK activation [12–14]. The importance of these Gα subunit-MAPK interactions remain to be fully understood but the analysis of these interactions in genetically amenable organisms are beginning to provide clues to the role of these interactions.

The genetic and biochemical analyses of G protein-mediated signaling in the budding yeast mating response have provided useful insights into how G proteins control many different cellular responses [2]. Loss of the Gα subunit, Gpa1, in this pathway results in the continuous stimulation of downstream responses suggesting that Gα subunit regulates the release of the Gβγ subunit dimer, the primary transducer of downstream signaling [15, 16]. This role of the Gα subunit contrasts the analysis of some mammalian signaling pathways that suggest the Gα subunit can be the primary activator of effectors such as adenylyl cyclase but mammalian Gβγ dimers can also activate some effectors [17]. In yeast the Gpa1 Gα subunit has also been shown to interact with a MAPK, Fus3, even though the Gα subunit is not required for the protein kinase cascade that regulates Fus3 activation [12]. Genetic analysis has shown that Gpa1 interacts with Fus3 through a D-motif near the amino terminus, as indicated by alterations in the D-motif (Table 1). Phenotypic characterizations of this D-motif mutant have implied that the Gα subunit regulates Fus3 distribution between the nucleus and cytoplasm in adaptive responses to mating pheromone. More recent studies have indicated that this Gα subunit alteration or alterations in Fus3 interfere with the polarity of chemotropic growth of yeast cells in pheromone gradients [18]. Comparison of Gα subunit sequences in a wide range of eukaryotic organisms has revealed other Gα subunits with a putative D-motif in the amino terminus near the first highly conserved region of G proteins (G1) suggesting Gα subunit-MAPK interactions might not be restricted to yeast [14, 19].

Table 1.

Gα subunit amino terminal D-motifs

Gα	(+)1–2 (x) 4–8ϕxϕ	G1 region
ScGpa1	MGCTVSTQTIGDESDPFLQNKRANDVIEQSLQLEKQRDKNEIKLLLLGAGES–
DdGα2	MGICASSMEGEKTNTDINLSIEKERKKKHNEVKLLLLGAGES–
DdGα5	MGCILTIEAKKSRDIDYQLRKEEGSKNETKLLLLGPGES–
DdGα11	MGSQFSVLNRKWLIERSIMIEKRKRRSNKLIKILMMGNENS–
HsGα15	MARSLTWRCCPWCLTEDEKAAARVDQEINRILLEQKKQDRGELKLLLLGPGES–
DdGα2D−	MGICASSMEGEATNTDINASAEKERKKKHNEVKLLLLGAGES–

Alignment of Ga subunit amino termini that contain putative MAPK docking sites (+_(1–2)-X_(4–8)-ϕ-X-ϕ) (shaded) from Dictyostelium discoideum (Dd), Saccharomyces cerevisiae (Sc) and humans (Hs). Positively charged (+) and hydrophobic residues (ϕ) are indicated in bold and the initial part of the conserved G protein G1 region is noted. The altered motif in the Gα2 subunit is indicated by the DdGα2^D− sequence.

The soil amoeba Dictyostelium discoideum has also been instrumental in understanding G protein-mediated signaling, including pathways activated by chemoattractants [20–22]. Dictyostelium chemotax to cAMP as a mechanism for nutrient-deprived solitary cells to find each other and form multicellular aggregates as the initial stage of the developmental life cycle [23, 24]. Cell surface receptors coupled to G proteins are stimulated by extracellular cAMP and in turn activate pathways that regulate cell movement, cell adhesion, and developmental gene expression [25–28]. Dictyostelium has three Gα subunits (Gα2, Gα5, and Gα11) with putative D-motifs close to the amino terminus [13, 14]. In the case of the Gα5 subunit, mutations that alter this motif lead to reduced Gα5 function [14]. Over expression of the wild-type Gα5 subunit is toxic to cell viability but alteration of the D-motif reduces this toxicity. Little is known about the signals that activate the Gα5 G protein or the downstream cellular responses but Gα5 function has been associated with the rate of multicellular development and the inhibition of chemotactic responses to folate [29]. The folate chemotactic responses are mediated through the Far1 receptor and a G protein that contains the Gα4 subunit [30, 31]. The putative D-motifs on the Gα2 or Gα11 subunits in Dictyostelium have not been previously examined. The role of Gα11 in Dictyostelium development or motility is unknown but the role of Gα2 has been extensively characterized with respect to pathways activated by extracellular cAMP [26, 32–40]. The Gα2 G protein is required for chemotaxis to cAMP and the aggregation of cells in response to starvation. The location of the D- motif in Gα2 is nearly identical to that of the D-motif in the yeast Gpa1 subunit suggesting these sites might have conserved functions in G proteins that direct the movement or growth of cells (Table 1).

The Dictyostelium genome encodes only two MAPKs, Erk1 and Erk2 [41]. Erk1 is a typical MAPK because its sequence similarity and activation mechanism are similar to many other MAPKs in a wide range of organisms [42]. Loss of Erk1 results in small aggregates upon starvation and precious development [43–45]. In contrast, Erk2 is considered an atypical MAPK based on its homology to a small sub group of MAPKs and an activation mechanism that does not require conventional MAPK kinases [42]. Erk2 is essential for chemotactic movement and cell aggregation and over 60 proteins phosphorylated in response to chemoattractants contain Erk2-specific phosphorylation motifs [42, 46]. Both MAPKs are activated in response to the stimulation of cell surface cAMP receptors (Car1 and Car3) but the timing of these activations are staggered [45, 47]. Erk2 activation occurs within 30 sec of receptor stimulation and a burst of Erk1 activation follows 2–3 minutes later in a secondary response when Erk2 activation terminates. The Gα2 subunit is not essential for the activation of Erk2 even though the activation requires the presence of a cAMP receptor [47]. Others have proposed that cAMP receptor activation of Erk2 occurs through a G protein independent mechanism but a redundancy in G protein function might also explain the activation in the absence of Gα2 [47, 48]. In a similar chemotactic response to folate, the Far1 receptor and Gα4 subunit are both essential for Erk2 activation [31, 45].

In this study, the role of the D-motif on the Gα2 subunit was examined by genetic analysis through the creation of a mutant allele that produces a subunit with an altered D-motif, Gα2^D−. Phenotypes of cells expressing the Gα2^D− protein were examined for aggregation, development, chemotaxis to cAMP, and the regulation of a transcription factor. Signaling deficiencies of these cells were also examined in chimeric populations. The results of these analyses suggest that the D-motif in Gα2 is important for some but not all Gα2-mediated functions.

2. Materials and Methods

2.1. Strains and media

All strains used in this study were isogenic derivatives of the parental axenic strain KAx3. Gα2 gene disruption using the thyA gene (also referred to as thy1) was conducted as previously described [26]. Strains were transformed by electroporation and grown in HL-5 medium or on bacterial lawns of Klebsiella aerogenes covering SM+/3 agar plates as previously described [49–51]. Cells were washed free of nutrients using a phosphate buffer (12mM NaH₂PO₄ adjusted to pH 6.1 with KOH). Multiple clones were initially analyzed from all electroporations to ensure consistency in observed phenotypes but select clones were used to generate the data reported.

2.2. Recombinant DNA constructs

The wild-type Gα2 open reading frame (ORF) from a cDNA vector was amplified using the oligos Gα2up and Gα2down (Table 2) and inserted as a XbaI-HindIII fragment into pBluescriptSKII. This fragment was inserted into the HindIII-XbaI sites of the expression vector, pDXA-GFP (removing the GFP sequence) conferring G418 resistance. An SpeI fragment of this vector (containing pact15:gα2 gene) was inserted into the same sites of pJH1075, pBluescriptSKII vector with a blasticidin resistance gene inserted at the PstI site [42, 52]. Both vectors are integrating but the G418 resistance vector requires many copies to confer drug resistance to 5 μg/ml G418 whereas a single copy of blasticidin resistance vector can confer drug resistance to 2 μg/ml blasticidin [42]. The Gα2 ORF was mutagenized using PCR and the oligonucleotides Gα2^D−sens and Gα2^D−anti (Table 2) converting key residue codons of the D-motif to alanine codons and incorporating a BglII restriction site (silent mutations) for identification. The coding region changes were verified by sequencing analysis and the open reading frame was inserted into the same expression vectors used for the wild-type gα2 allele. Extra-chromosomal expression vectors for GFP-Erk1 or GFP-Erk2 conferring G418 resistance have been previously described [53]. Expression vector pHC326 for the GFP-GtaC transcription factor fusion has been previously described [54]. The GFP expression vector pTX-GFP2 was used to label cells in developing chimeric populations [55].

Table 2.

Oligonucleotide list

Oligo	Sequence
Gα2up	GCCGGCAAGCTTAAAAAATGGGTATTTGTGCATCATCAATGGAAGGAG
Gα2down	CGGCGCTCTAGATTAAGAATATAAACCAGCTTTCATAACACATTG
Gα2D−sens	CAACCAATACTGATGCTGCAGCATCTATTGAAAAAGAAAG
Gα2D−anti	CAGCATCAGTATTGGTTGCTTCTCCTTCC

Oligonucleotides used for amplifying the ga2 open reading frame (Gα2up, Gα2down) and creating the altered D-motif (Gα2^D−sens, Gα2^D−anti).

2.3. Development and chemotaxis analysis

Cells were grown in fresh HL-5 medium for one day prior to harvesting by centrifugation, washing in phosphate buffer and plating cell suspension droplets (5 μl of 2 × 10⁷ cells/ml to give 5 mm diameter circular patches) on non-nutrient agar plates (1.5% agar in phosphate buffer). Developing cell aggregates on non-nutrient plates or bacterial lawns were observed using a dissecting microscope. Above agar chemotaxis assays were conducted as previously described by plating droplets of cell suspensions (1 × 10⁷ cells/ml) on non-nutrient agar plates and then plating a droplets of cAMP solution approximately 2 mm away from the cell droplet [53]. Images were acquired at the start of the assay and 3 hr later and needle-punctured scars in the agar were used to align images from the different periods. Chemotaxis was measured by determining the leading edge of migrating cells in the direction of the chemoattractant. In each assay at least five droplets were analyzed for each strain and strains were compared side-by-side on the same agar plates for consistency. Multiple independent assays were conducted for each strain. Chimeric population development was conducted using one population labeled with a GFP expression vector (pTXGFP2) mixed with an unlabeled population in a 1:10 ratio prior to plating on a non-nutrient plate. For the aggregation assay, conditioned medium from KAx3 strain was prepared by starving KAx3 cells at 9×10⁶ cells/ml in developmental buffer for 17 hr as similarly described by [56]. Cells were pelleted and the supernatant was collected and stored on ice. Cells expressing Gα2^D− or Gα2 and gα2⁻ strains were grown overnight in shaking cultures of fresh HL5, pelleted, washed and suspended in developmental buffer at 5 × 10⁷ cells/ml before being spotted in 3 μl aliquots onto nonnutrient plates. Every hr during the first 6 hr 2 μl droplets of conditioned medium, 100 μM cAMP, or both were spotted adjacent to the cell droplets.

2.4. Transcription factor shuttling

Strains expressing the GFP-GtaC transcription factor fusion protein were grown in fresh medium one day prior to their analysis. Cells were harvested from the plates and allowed to settle on coverslips attached to 60 mm petri dishes containing a 10 mm diameter hole. Excess cells and medium were removed and replaced with fresh medium for at least five min before the cells were washed and then covered with developmental buffer (phosphate buffer containing 2 mM MgCl₂ and 1 mM CaCl₂). A solution of 1 μM cAMP was added to cells to reach a final concentration of 100 nM. Time-lapse imaging of the GFP-GtaC protein was recorded every 30 sec using spinning disk confocal fluorescence microscopy and CellSens software. The nuclear and cytoplasmic distribution of the GFP-GtaC reporter, as determined by mean pixel intensity, was analyzed in the images using FIJI (ImageJ) software. Occasionally cell movement on the coverslip transiently results in nuclei moving out of the focal plane and so the data plotted represented the mean of ratios within the middle 50% of values to reduce this variability.

3. Results

3.1. Overexpression of the Gα2 or Gα2^D− subunit

To examine the contribution of the putative D-motif in the amino terminal region of the Gα2 subunit, the signature residues of the motif were changed to alanines (Table 1). Similar alterations have been used to examine D-motifs in other Gα subunits [12–14]. The expression of this D-motif mutant subunit, Gα2^D−, or the wild-type Gα2 subunit from the relatively constitutive act15 promoter on extrachromosomal vectors conferring G418 drug resistance was initially used to assess whether or not mutant subunit could rescue developmental defects of gα2⁻ cells. During synchronous starvation on non-nutrient agar plates gα2⁻ cells expressing either Gα2^D− or Gα2 were able to aggregate but most aggregates remained at the mound stage rather than progressing to slugs and fruiting bodies like the parental strain (KAx3) containing only an endogenous gα2 allele (Fig. 1A). Clonal transformants growing on bacterial lawns created plaques in which cell aggregates were observed for both Gα2 and Gα2^D− expressing cells but rather than continuing through the multicellular developmental life cycle the mounds typically disaggregated (data not shown). Only in rare cases for both the Gα2 and Gα2^D− strains an individual mound continued to develop further into fruiting bodies. To determine if the heterologous expression of the Gα2 subunits interferes with developmental progression, the Gα2 and Gα2^D− expression vectors were introduced into the parental strain for phenotypic analysis. Clones containing either expression vector displayed cell aggregation in response to synchronous starvation but the transition between the mound and slug stages of development was severely delayed or completely blocked (Fig. 1B). This impairment of development suggests that heterologous expression of the Gα2 or Gα2^D− subunit from a high copy number vector can provide sufficient Gα2 function to complete aggregation but then lead to a delay or block in subsequent developmental processes.

Figure 1. Developmental phenotypes of strains with high copy number expression vectors.

(A) gα2⁻ and (B) parental (KAx3) strains with or without Gα2 or Gα2^D− expression from high copy number vectors (hc) were washed free of nutrients and plated on non-nutrient plates. Images of developmental morphology were recorded at the times indicated. Only the 6 hr image of the gα2⁻ strain is shown because the cells remained aggregation deficient at later times. All images are at the same magnification.

3.2. Gα2^D− expression from a low copy number vector does not rescue aggregation efficiently

As an alternative to the high copy number vectors, the Gα2 and Gα2^D− subunits were expressed from a previously characterized low copy number integrating vector conferring blasticidin resistance [42]. The expression of Gα2 subunit from this vector in gα2⁻ cells rescued aggregation and subsequent development without a noticeable delay in the transitions from mounds to slugs suggesting the lower copy number of Gα2 vector provides a more physiological relevant level of Gα2 expression than the high copy number vector (Fig. 2). This phenotype was consistently observed for independent clones implying that most random genomic integration sites allow for appropriate gene dosage. In contrast, gα2⁻ cells expressing the Gα2^D− subunit from the same vector were defective in aggregation. This aggregation defect was not observed when the Gα2^D− vector was expressed in parental cells containing the endogenous gα2 allele indicating that ectopic expression of the Gα2^D− subunit does not result in a dominant aggregation defect. However, the observed aggregate size for this strain was smaller than that of wild-type cells without the Gα2^D− expression vector. Expression of the wild-type Gα2 subunit from the low copy vector was sufficient for aggregation and development on bacterial lawns. However, Gα2^D− expressing cells on bacterial lawns were aggregation deficient (Fig. 2). The aggregation defect of gα2⁻ cells expressing the Gα2^D− subunit exhibited some phenotypic instability because extended culturing (i.e., multiple weeks) of clonal populations resulted in a small subset of cells capable of aggregation when plated for development at high densities (5 × 10⁷ cells/ml). The basis of this instability (e.g., possible suppressing mutations or alterations of the expression vector) remains to be determined but such populations typically remained aggregation deficient when grown on bacterial lawns. All subsequent analyses of the Gα2 and Gα2^D− subunits were conducted in gα2⁻ clones containing the low copy number expression vectors.

Figure 2. Developmental phenotypes of strains with low copy number expression vectors.

Cell suspensions of the parental (KAx3) or gα2⁻ cells with or without Gα2 or Gα2^D− expression from low copy number vectors were plated at the same density on non-nutrient plates and images were recorded at the time indicated (upper panels, same magnification for all images). Strains were also spotted on a lawn of Klebsiella aerogenes and images of the plaques and developing aggregates were taken after 4 days (lower panels, same magnification for all images). Plaques on the edge of the images are from different strains. Mature fruiting bodies are indicated by arrows. Images are also shown for the gα2⁻(Gα2^D−) strain after extended culturing (ext.).

3.3. Gα2^D− expression restores motility in response to cAMP stimulation

The defective aggregation associated with the Gα2^D− subunit could possibly result from cell autonomous defective responses to cAMP stimulation or from the inability of the cells to generate sufficient intercellular signals such as cAMP. Chemotaxis assays of gα2⁻ cells expressing the Gα2 or Gα2^D− subunit indicated both subunits allow for increased movement to exogenous cAMP stimulation compared to gα2⁻ cells without either expression vector (Fig. 3). In both strains, cells moved greater distances in the presence of cAMP but a directional bias in this movement to was not evident in either strain compared to that observed for the parental wild-type strain (KAx3). In these chemotaxis assays the gradient of cAMP is not static because the diffusion of cAMP over time allows for cell movement in all directions from the cell droplet but a greater proportion of wild-type cells consistently move in the direction of the cAMP source (Fig. S1). The Gα2 and Gα2^D− expressing cells exhibited similar movement in all directions rather than a bias toward the cAMP source and the movement was much greater in the presence of cAMP compared to the absence of cAMP. This movement is more characteristic of chemokinesis rather than chemotaxis.

Figure 3. Chemotaxis of strains expressing the Gα2 or Gα2^D− subunit to cAMP.

Cells were prepared and assayed for cAMP chemotaxis as described in the Methods section. Relative movement indicates migration distance from the original droplet perimeter to the leading edge of cells after 3 hr. Cell movement toward (forward), away from (reverse), or in the absence of cAMP are displayed. Values represent the mean of one chemotaxis assay of six cell droplets for each strain and the error bars represent the standard deviation. Chemotaxis data is representative of at least 3 assays. Differences indicated by ‘NS’ or the distance differences of all gα2⁻ strains were assessed by Student’s unpaired t-test and determined to be not significantly different (P>0.1). All other pairings were significantly different (P<0.002). Representative images of chemotaxis assays are displayed in Fig. S1.

3.4. Alteration of the D-motif results in defective intercellular signaling

Aggregation deficiencies of mutant strains can sometimes be rescued by the intercellular signaling of wild-type cells in chimeric populations if the mutant strain does not secrete sufficient levels of intercellular signals [57]. GFP-labeled Gα2 or Gα2^D− subunit expressing cells were mixed with parental KAx3 cells prior to development and the presence of GFP expressing cells in aggregates was determined using fluorescence microscopy. The Gα2 and Gα2^D− subunit expressing cells were capable of co-aggregating with the wild-type cells suggesting the presence of wild-type intercellular signaling is sufficient to rescue the aggregation of Gα2^D− expressing cells (Fig. 4). The distribution of the Gα2 and Gα2^D− cells in chimeric slugs were similar in that both were underrepresented in the extreme anterior regions but otherwise these cells were found throughout the other regions. Chimeric populations with gα2⁻ cells showed limited co-aggregation with parental KAx3 cells suggesting that some of these mutant cells can been carried into the aggregate without having chemotactic responses to cAMP.

Figure 4. Coaggregation of gα2⁻ strains expressing the Gα2 or Gα2^D− subunit with parental strain KAx3.

Cells expressing the Gα2 or Gα2^D− subunit or no Gα2 subunit were labeled with a GFP expression vector and mixed with KAx3 cells at a ratio of 1:10 before development on non-nutrient plates. Images of developing slug structures were recorded using fluorescence microscopy (slug anterior on right side of image). All images are at the same magnification except the brightfield inset images that are at 25% magnification.

The co-aggregation of Gα2^D− subunit expressing cells with parental KAx3 cells could possibly be mediated by secreted factors, including cAMP. Therefore Gα2^D− subunit expressing cells were developed in the presence of conditioned medium obtained from the extracellular fluid of developing KAx3 cells or the presence of exogenous cAMP. Neither the conditioned medium, cAMP, or the combination of both (data not shown) was sufficient to rescue the aggregation of Gα2^D− subunit expressing cells (Table 3, Fig. S2). This result suggests that the intercellular signaling provided by the KAx3 strain might require kinetic or concentration parameters not present in this aggregation assay or the signaling might be mediated through cell-cell contact.

Table 3.

Aggregation of strains supplemented with exogenous factors.

	Aggregate formation with supplements
Strain	None	CM	cAMP
gα2 −	−	−	−
gα2−(Gα2D−)	−	−	−
gα2−(Gα2)	+	+	+

Cell droplets on non-nutrient agar plates were treated with developmental buffer (none), conditioned medium (CM), or 100 μM cAMP every hr for 6hr. Representative images of cell droplets after 11 hr are displayed in Fig. S2.

3.5. D-motif alteration impacts transcription factor translocation

Many G protein-mediated signaling pathways modulate gene expression through the regulation of transcription factors. A recently characterized Dictyostelium response to external cAMP is the transient translocation of the transcription factor GtaC from the nucleus to the cytoplasm and this shuttling of GtaC is thought to be an important process for gene regulation during the aggregation phase of development [54, 58]. The movement of GtaC from the nucleus to the cytoplasm can be monitored through the use of a GFP-GtaC reporter construct [54]. To test whether or not the D-motif on Gα2 impacted the shuttling of the GtaC transcription factor between the nucleus and the cytoplasm, a GFP-GtaC reporter vector was introduced into strains expressing Gα2 or the Gα2^D− subunit. Both strains exhibited translocation of the reporter from the nucleus to the cytoplasm after cAMP stimulation but the translocation in the Gα2^D− expressing strain was slower and less extensive than in the Gα2 expressing strain but more extensive than the translocation in the gα2⁻ strain (Fig. 5, see Fig. S3 for videos). This observation suggests that the D-motif on Gα2 contributes to GtaC translocation and therefore it might play a role in developmental gene regulation.

Figure 5. Translocation of the GtaC transcription factor in gα2⁻ cells expressing Gα2 or Gα2^D− subunits.

Strains transformed with the GFP-GtaC reporter vector were prepared as described in the Methods section and then stimulated with 100 nM cAMP. Parental KAx3 cells containing the same reporter are shown as a control. (A) Images at the start of cAMP stimulation and 2 min after stimulation are shown. Time-lapse videos can be found in Fig. S3. (B) Quantification of the change in nuclear to cytoplasm ratio fluorescence in response to cAMP stimulation. Data represent the mean change in ratios within one standard deviation of the strains gα2⁻(Gα2) (open circles, n=52), gα2⁻(Gα2^D−) (closed circles, n=52, 100nM cAMP), and gα2⁻ (closed squares, n=75, 100nM cAMP). Parental strain KAx3 (open squares, n=16). Error bars represent the standard deviation. See Methods for statistical analysis.

3.6. Alteration of the Gα2 D-motif does not change the overall cellular distribution of MAPKs

The translocation of transcription factors between the nucleus and cytoplasm is often dependent on MAPKs or other protein kinases and some recent studies suggest MAPKs are involved with this regulation (Cai and Hadwiger, unpublished results). In a previous study, we demonstrated that Dictyostelium Erk1 and Erk2 can be found in both the cytoplasm and nucleus based on the expression and distributions of GFP tagged Erk1 and Erk2 proteins [53]. This study qualitatively indicated that chemoattractant stimulation with either cAMP or folate did not cause major changes in the distribution of the GFP-Erk1 or GFP-Erk2 in cells. We re-examined MAPK distribution using confocal fluorescence microscopy to better quantify possible changes. The absence of major changes in the ratio of nuclear/cytoplasmic ratio over the course of cAMP stimulation in the parental wild-type cells (KAx3) confirmed previous conclusions that the overall distribution of GFP-Erk2 does not undergo a pronounced change in response to cAMP stimulation (Fig. 6, S4 videos). The nuclear/cytoplasmic ratio of GFP-Erk2 distribution also did not change in Gα2 or Gα2^D− expressing cells stimulated with cAMP. The same analysis of GFP-Erk1 was more challenging due to the limited number of cells with a detectable level of fluorescence in all strains. The basis of this limited expression remains to be determined but could possibly represent toxicity associated with the GFP-Erk1 protein in most cells. This lack of change was also observed in the very few Gα2 or Gα2^D− expressing cells with a detectable levels of GFP-Erk1 (Fig. S5). While no major changes in the nuclear/cytoplasmic ratios were observed in response to cAMP stimulation it is possible that a small portion of the MAPKs can translocate between these compartments during the response. The possibility that overexpression or tagging of MAPKs in Dictyostelium alters MAPK movement through the cell cannot be ruled out either.

Figure 6. Cellular distribution of GFP-Erk2 during cAMP stimulation.

Strains transformed with the GFP-Erk2 reporter vector were prepared and analyzed as described in Fig. 5. Parental KAx3 cells transformed with the same vector are shown as a control. (A) Representative images at the start of cAMP stimulation and 4 min post stimulation (upper panels). Time-lapse videos can be found in Fig. S4.

(B) Quantification of the change in nuclear to cytoplasm ratio fluorescence in response to cAMP stimulation. Data represent the mean of ratios within the middle 50% of values for the strains gα2⁻(Gα2) (open circles, n=52), gα2⁻(Gα2^D−) (closed circles, n=52), KAx3 (open squares, n=16), and gα2⁻ (closed squares, n=75).

4. Discussion

The results from this study indicates that putative D-motif in N-terminal region of the Dictyostelium Gα2 subunit contributes to the function of this subunit in the aggregation of cells during the developmental response to starvation. While the requirements for Gα2 subunit function in aggregation and cAMP chemotaxis were established many years ago from the analysis of strains with a disrupted gα2 locus, the phenotype of cells expressing the Gα2^D− subunit is distinct in multiple ways. First, the aggregation deficiency phenotype of Gα2^D− expressing cells can be genetically unstable in some synchronously starved populations even though aggregation is absent in plaques when cells are grown on bacterial lawns. Second, the defective aggregation is not directly associated with a loss of cAMP stimulated movement because the Gα2^D− expressing cells display movement in response to cAMP in a manner similar to that of Gα2 expressing cells but different than gα2⁻ cells. Last, cells expressing the Gα2^D− allele can co-aggregate with wild-type cells suggesting that the altered Gα2 subunit confers aggregation competence when supplied with intercellular signals or interactions (Fig. 7). Overall, the Gα2^D− subunit is capable of some but not all Gα2 functions and some of these functions are limited to synchronous starvation. These observations suggest the D-motif of the Gα2 subunit might contribute to signaling complexes that involve MAPKs and that regulate intercellular signaling, such as the diffusion of external cAMP. While it remains to be determined which if any MAPKs associate with the Gα2 subunit, the analysis of the putative D-motif indicates that this region is important for the function of this G protein subunit. As proposed for MAPK-Gα subunit interactions in yeast, the amino terminus of the Gα subunit is likely to interact with MAPKs when the subunit is activated and dissociated from the Gβγ dimer [12]. The amino terminus of some Gα subunits have been reported to interact with Gβγ dimers but in Dictyostelium the single conventional Gβγ dimer functions with multiple Gα subunits, such as the Gα2 and Gα4 subunits, that do not have similar amino terminal sequences [13, 59]. Therefore, the D-motif in Gα2 is unlikely to be critical for Gβγ dimer interactions.

Figure 7. Model for the role of the Gα2 D-motif in developmental signaling.

Stimulation of cAMP receptors (cARs) leads to the activation of the Gα2 G protein and a variety of cellular responses including chemotaxis, gene regulation, and intercellular signaling. Canonical signaling suggests that multiple steps occur between the G protein and MAPK activation but the D-motif in Gα2 contributes to MAPK regulated processes during aggregation such as the production of intercellular signals and the translocation of the GtaC transcription factor. Heterologous expression of the Gα2 subunit (red arrow) can lead to the inhibition of developmental progression.

The impaired translocation of the GFP-GtaC reporter in Gα2^D− expressing cells in response to cAMP implies that Gα2 D-motif can contribute to this process. An earlier report demonstrated that GtaC is required for aggregation and the translocation of GFP-GtaC occurs independently of the Gβ subunit that functions to mediate chemotaxis responses to cAMP suggesting that the signaling pathway that regulates this translocation can occur independently of the Gβ subunit [54]. The reduced translocation of the GFP-GtaC reporter in gα2⁻ cells suggests that the Gα2 subunit is important but not essential in the regulation of this process. The translocation is fully rescued by heterologous expression of the Gα2 but not Gα2^D− subunit underscoring a the role for the D-motif in this process. Previously, GtaC has been shown to be phosphorylated in cells responding to cAMP and therefore it is possible that the D-motif of the Gα2 subunit impacts the function or distribution of protein kinases that regulate GtaC (Fig. 7). While neither the Gα2 or the Gβ subunits are essential for MAPK activation in response to exogenous cAMP these subunits could possibly contribute to MAPK regulation in combination with other signaling proteins. MAPK function has been demonstrated to be important for gene expression, chemotaxis, and intercellular signaling in Dictyostelium but whether MAPKs regulate these processes in parallel or sequentially remains to be understood [42, 43, 57, 60, 61].

The lack of detectable changes in the nuclear/cytoplasmic distribution of GFP-Erk1 or GFP-Erk2 in response to cAMP stimulation suggests that large changes in the distribution of MAPKs are not required for the rapid translocation of the GtaC transcription factor. The assessment of MAPK distribution in other organisms have also utilized MAPKs tagged with fluorescent proteins and in some cases changes in distribution have been associated with cell stimulation. In mammals the typical MAPKs, Erk1 and Erk2, show increased presence in the nucleus in response to cell stimulation [62]. In the yeast mating response, the Fus3 MAPK becomes enriched in the nucleus and this enrichment is impacted by alterations in the D-motif of the Gpa1 Gα subunit [12, 63]. Interestingly, the Kss1 MAPK can also be activated in response to mating pheromone but this MAPK increases in the cytoplasm suggesting that MAPKs do not follow a universal mechanism in response to external stimuli [64]. The translocation of Dictyostelium MAPKs between the cytoplasm and nucleus in response to cAMP stimulation cannot be entirely ruled out because small portions of the MAPK population could move undetected and it is also possible that the GFP-tagged version of the MAPKs might not be subjected to the same regulation as the endogenous MAPKs.

A surprising result from the analysis of the heterologous Gα2 subunit expression from high copy number vectors was the inhibition of developmental progression beyond the mound stage (Fig. 7). While not previously reported in the complementation of gα2⁻ mutants, this phenotype probably is likely the result of higher gene copy number of Gα2 expression vectors because the low copy number vectors with the identical Gα2 genes did not inhibit this developmental transition. The contributions of the act15 promoter to this developmental phenotype cannot be ruled out because even the low copy number Gα2 expression vectors with this promoter affected directional cell movement in the cAMP chemotaxis assays. An earlier study has shown that expression of the gα2 gene from a prespore- or prestalk-specific promoter did not alter development but these reporters are expressed later in development compared to the act15 promoter [65]. Heterologous expression of the Gα2 subunit could impair signaling processes at these later stages especially if there are competitions with other related Gα subunits for interactions with other signaling proteins such as receptors. Two of the four cAMP receptors and possibly other G protein-coupled receptors in Dictyostelium are expressed after aggregation [66–68]. Heterologous expression of other Gα subunits can also affect developmental progression. Overexpression of the Gα1 subunit, the closest paralog to the Gα2 subunit, results in aberrant morphogenesis in the last stages of multicellular development [29, 69]. Overexpression of the Gα4 subunit, a mediator of folate chemotaxis, can delay the aggregation phase and block development after aggregate formation by preventing prestalk cell development [29, 70, 71]. Conversely, overexpression of the Gα5 subunit, most closely related to the Gα4 subunit, has the reverse effect by promoting prestalk cell development and precocious tip formation on mounds [29, 50]. The basis of developmental phenotypes associated with Gα subunit overexpression is not well understood but the phenotypes are specific to individual Gα subunits. Therefore the developmental delay associated with the high copy number Gα2 expression vectors is not likely the result of general Gα subunit overexpression but rather due to Gα2 subunit expression.

5. Conclusion

The results of this study suggest the region corresponding to the putative D-motif of Gα2 serves important roles in intercellular signaling during aggregation and the translocation of the GtaC transcription factor. The heterologous expression of the Gα2 subunit can also compromise the directed movement of cells to cAMP and inhibit developmental morphogenesis suggesting that G protein subunit stoichiometry is an important factor throughout the developmental life cycle. While not directly contributing to the activation of MAPKs, Gα subunits might contribute to signaling pathways by associating with complexes that contain MAPKs. The presence of D-motif sequences in some of the Gα subunits found in other organisms suggests such Gα subunits possibly play a role in the regulation of cellular responses in organisms other than yeast and Dictyostelids.

Supplementary Material

Figure S3-1.avi

Figure S3. Videos of GFP-GtaC translocation in strains gα2⁻(Gα2), gα2⁻(Gα2^D−), gα2⁻and parental strain KAx3 after stimulation with 100 nM cAMP. Each video covers an 8 min period with images acquired every 30 sec. Red dot on second frame designates the addition of cAMP.

Figure S4-1.avi

Figure S4. Videos of GFP-Erk2 distribution in strains gα2⁻(Gα2), gα2⁻(Gα2^D−), gα2⁻and parental strain KAx3 after stimulation with 100 nM cAMP. Each video covers an 8 min period with images acquired every 30 sec. Red dot on second frame designates the addition of cAMP.

Figure S5-1.avi

Figure S5. Videos of GFP-Erk1 distribution in strains gα2⁻(Gα2) and gα2⁻(Gα2^D−) after stimulation with 100 nM cAMP. Each video covers an 8 min period with images acquired every 30 sec. Red dot on second frame designates the addition of cAMP.

Figure S1

Images of chemotaxis assay to cAMP. Original cell droplet perimeter (inner circle) and 3 hr perimeter (dashed outer circle). Orientation of cAMP source is the upper side of each image. is All images are the same magnification.

Figure S2.

Images of aggregation assays with supplements. Cell suspensions of gα2⁻ cells with or without Gα2 or Gα2^D− expression from low copy number vectors were plated at the same density on non-nutrient plates and droplets of supplements developmental buffer (none), conditioned medium (CM), or 100 μM cAMP were added adjacently hourly for 6 hr. Images were recorded after 11 hr (same magnification for all images).

Highlights:

• MAPK docking motifs are present in the amino terminus of some Gα subunits

• Gα2 subunit MAPK docking motif mediates Dictyostelium aggregation signaling

• Altered MAPK docking motif impairs translocation of the GtaC transcription factor

Abbreviations:

MAPK

mitogen activated protein kinase

cAMP

cyclic adenosine monophosphate

GFP

green fluorescent protein