Assaying Rho GTPase–dependent processes in Dictyostelium discoideum

The model organism D. discoideum is well suited to investigate basic questions of molecular and cell biology, particularly those related to the structure, regulation, and dynamics of the cytoskeleton, signal transduction, cell-cell adhesion, and development. D. discoideum cells make use of Rho-regulated signaling pathways to reorganize the actin cytoskeleton during chemotaxis, endocytosis, and cytokinesis. In this organism the Rho family encompasses 20 members, several belonging to the Rac subfamily, but there are no representatives of the Cdc42 and Rho subfamilies. Here we present protocols suitable for monitoring the actin polymerization response and the activation of Rac upon stimulation of aggregation-competent cells with the chemoattractant cAMP, and for monitoring the localization and dynamics of Rac activity in live cells.


Dictyostelium discoideum as a model organism
Dictyostelium discoideum is a simple eukaryotic microorganism whose natural habitat is the deciduous forest soil where free-living amoebas feed on bacteria and multiply by equal mitotic division. Exhaustion of the food source triggers a developmental program in which more than 100,000 cells aggregate by chemotaxis toward cAMP to form a multicellular structure. Differentiation and morphogenesis result in the formation of a fruiting body composed of a mass of spores supported by a stalk made of vacuolized dead cells. Numerous advantages make D. discoideum a widely accepted and well-suited model organism to investigate basic questions of molecular and cell biology, particularly those related to the structure, regulation and dynamics of the cytoskeleton, signal transduction, cell-cell adhesion and development (1). More recently D. discoideum is being employed as a model to study the mechanisms of infection by pathogenic bacteria and the molecular causes of human diseases (2). D. discoideum has a short life cycle, is easy to cultivate, grows in inexpensive media, can be harvested in large amounts and is amenable to a variety of biochemical and molecular and cell biological techniques (3,4). Its genome, completely sequenced and largely annotated, can be easily manipulated by means of a growing list of molecular genetics techniques that have allowed considerable advances in the study of biological processes.

Rho GTPases in D. discoideum
Like other eukaryotic cells D. discoideum makes use of Rho-regulated signal transduction pathways to reorganize its cytoskeleton during chemotaxis and other processes such as endocytosis and cytokinesis. In D. discoideum the Rho family encompasses 20 members. Rho signaling is complemented in this organism with more than 100 regulators (exchange factors, GTPase activating proteins and guanine nucleotide dissociating inhibitors) and about 80 effectors or components of effector complexes (Scar, WASP, formins, PAK kinases,IQGAP-related proteins, etc). Based on phylogenetic analyses, several Rho GTPases can be grouped in the Rac The final publication is available at Springer via https://doi.org/10.1007/978-1-4939-8612-5_25 4 subfamily: Rac1a, Rac1b, Rac1c, RacF1, RacF2, and more loosely RacB and the GTPase domain of RacA (the D. discoideum RhoBTB ortholog). All other D.
discoideum Rho GTPases were named Rac for historical reasons but they do not have a clear affiliation, although some are closer to Rac than to members of other subfamilies. There are no representatives of the Cdc42, Rho or other subfamilies in D. discoideum (5).
Like in other organisms, knockout, overexpression and gain-of-function mutants have been used to investigate the physiological roles of D. discoideum Rho GTPases.
However, the presence of multiple Rac proteins, some of them as very closely related isoforms, has presented a challenge for the elucidation of their function, because of potential functional redundancy. Nevertheless, the evidence accumulated so far implicates Rho GTPases in the regulation of chemotaxis and cell motility, endocytosis and vesicle trafficking, cytokinesis and development in D. discoideum.
The reader is referred to a recent comprehensive review on Rho signalling in D.

Assaying Rac-dependent processes in D. discoideum
Because it would be impossible to provide detailed methods to investigate all relevant processes in which Rho GTPases have been implicated in D. discoideum, here we will restrict ourselves to protocols for monitoring the actin polymerization response, the activation of Rac upon cAMP stimulation of aggregation competent cells, and the localization and dynamics of Rac activity in live cells. For more detailed investigation of actin localization, chemotaxis, endocytosis and other processes the reader is referred to other sources (3, 4).

Filamentous actin content
The filamentous (F-)actin content is quantitated by fluorescently labeled phalloidin staining of formaldehyde-fixed pelleted material, an adaptation to D. discoideum cells (7) of the method originally devised for neutrophils (8). The method has been applied extensively to monitoring the actin polymerization response induced by cAMP in aggregation competent cells in a variety of circumstances. F-actin exhibits a characteristic biphasic curve upon agonist stimulation. During the first peak, at 5 sec, the amount of F-actin approximately doubles. This peak is very short and correlates with the cringe reaction in which the cells round up and produce a uniform cortical accumulation of F-actin. The second peak is significantly broader and lower, shows a maximum at 30-60 sec and corresponds to the emergence of pseudopods and cell movement. The actin polymerization response parallels in time and magnitude that of Rac activation, on which it depends (9)(10)(11) and therefore constitutes an excellent tool to monitor alterations in Rac-dependent signaling pathways (Fig. 1A).

Rac activation assay
The Rac activation (or pull-down) assay is an adaptation for D. discoideum of the method described by Benard et al. (12) for the determination of Rac and Cdc42 activation in chemoattractant-stimulated human neutrophils. The method makes use of a fusion protein of glutathione-S-transferase (GST) and the Rho GTPase binding domain of an effector molecule to capture active Rac in cell lysate. In D. discoideum this method has been applied to the determination of activated Rac1, RacB and RacC (9)(10)(11)13). In principle the method could be adapted to determine the activity of any other Rac, provided a GST fusion with a specific GTPase-binding domain (GBD) and a specific antibody to detect the Rac by Western blot are available. However, Rho GTPase-binding domains of potential use have been identified for a very limited number of D. discoideum Racs (Rac1a/1b/1c, RacA, RacB, RacC and RacF1/2) only, and they include WASP and related proteins, PAK kinases and IQGAPs (Table 1 and   ref. 14). We have tested the GTPase binding domains of D. discoideum WASP and PAKb as well as human PAK1 for the pull-down assay of activated Rac1 and have obtained better results with WASP (11,13). WASP has also been used by Han (9), an approach that will be very useful if applied in future to the other members of the Rac family.

Monitoring Rac activity in live cells
A limitation of the pull-down assay is that it provides a measurement of the total activity of a given Rac in a population of cells and therefore does not permit the spatial and temporal monitoring of Rac activation within individual living cells. Probes based on fluorescence resonance energy transfer (FRET) have been developed and applied to D. discoideum (10,(16)(17)(18)(19). The first FRET probe for monitoring of RacC activity in vivo consisted of cyan fluorescent protein (CFP)-RacC and yellow fluorescent protein (YFP)-B-GBD (the GBD of WASP) (10). The main obstacle encountered when attempting to image FRET probes in live cells is a prohibitively long integration time needed to obtain an acceptable signal-to-noise ratio (19). In order to be able to follow the Rac dynamics during fast intracellular processes, investigators resorted to the use of simpler probes based on fluorescently labeled Rac-binding domains (Table 1). These probes were mostly used to monitor the activity of Rac1 and RacC, but appear to have different degrees of specificity.
Interestingly, the fluorescently labeled Rac1 effector DGAP1 displayed an opposite localization to PAK-GBD-based probes, suggesting that the entire population of active Rac is not detected by the use of standard probes (14).

Materials
All media and buffers are prepared using deionized water (dH2O), filtered through an ion-exchange unit.

Culture and selection of D. discoideum cells
D. discoideum cells are cultured in nutrient medium in suspension on a rotary shaker (160 rpm) at 21°C. Cells are harvested and washed at room temperature by 7 centrifugation at 500×g for 3 min. When analyzing different strains, it is recommended that cells be cultivated under the same conditions for some days preceding the assays. For the assays described here cells should be taken from axenically growing or "log phase" cultures (up to 2×10 6 cells/mL). Dilute cultures the day before in order to have the necessary number of log phase cells. Autoclave.

Rac activity assay
1. E. coli strain (e.g. XL-1 blue or BL21) carrying a pGEX series vector (GE Healthcare) with the GBD domain of D. discoideum WASP cloned in frame to GST. The plasmid is available from the authors or can be easily constructed using standard molecular biology procedures. A cDNA encoding WASP is available from the Dicty Stock Centre (http://dictybase.org/StockCenter/StockCenter.html).
2. Luria broth (LB): dissolve 10 g of Bacto-tryptone, 5 g of yeast extract and 10 g of NaCl in 600 mL of dH2O, adjust pH to 7.5 with 1 N NaOH, complete to 1 L with dH2O, autoclave.

Determination of the relative F-actin content
Using this method cells are permeabilized immediately by Triton X-100, allowing very rapid fixation (within 1 sec) by formaldehyde. TRITC-phalloidin binds specifically to Factin. TRITC-phalloidin is methanol-extracted from the Triton insoluble pellet and quantitated fluorimetrically, providing a measurement of the amount of F-actin present in the cell sample. Here we describe the determination of the basal F-actin content of a cell strain relative to a control strain.
1. Harvest and resuspend cells at 2×10 7 cells/mL in fresh nutrient medium. Allow cells to recuperate for 30 min.

Actin polymerization response upon cAMP stimulation
The same principle as in Subheading 3.1 is applied to the determination of the time course of F-actin formation upon agonist stimulation.

Pull-down assay for activated Rac
The method makes use of a fusion protein of GST and the Rho GTPase binding domain of an effector molecule. The fusion protein is expressed recombinantly in E. coli and is purified by attachment to glutathione agarose. The glutathione Sepharose beads carrying the fusion protein are applied to a cell lysate. The beads are then separated from the lysate followed by washing and elution of the captured active Rac in sample loading buffer. The active Rac is detected by Western blot. The protocol given below is designed for the determination of activated Rac1 using the CRIB (Cdc42 and Rac interactive binding) domain of WASP (11), but can in principle be adapted for the detection of any activated Rho GTPase provided a specific effector is available and the GTPase can be identified on a Western blot.

Production of GST-WASP CRIB
1. Innoculate 50 mL of LB/ampicillin (50 µg/mL) with a single colony of E. coli transformed with the pGEX-WASP CRIB plasmid. Grow overnight at 37°C.

Pull-down of activated Rac upon cAMP stimulation
Following protocol is devised for the determination of activated Rac1 in one D.  (Fig. 1B).

Monitoring Rac activity in live cells
In order to monitor the localization and activity of Rho GTPases in living D. discoideum cells, a number of GBDs fused to fluorescent proteins have been used (see Subheading 1.3.3 and Table 1). Here we describe the construction of a probe consisting of DYFP fused C-terminally to the GBD of protein kinase DPAKa (aa 731-890). We then describe how to determine the relative intensity distribution of the DPAKa(GBD)-DYFP probe in the plasma membrane, and illustrate its use in estimating Rac1 activation upon stimulation of cells with chemoattractant pulses.

Construction of the DPAKa(GBD)-DYFP probe
The probe for active Rac1 is constructed by two step cloning: (a) insertion of DYFP in the extrachromosomal pDM304 vector (20) to obtain a vector with C-terminal DYFP tag followed by (b) insertion of the GBD domain of DPAKa in pDM304 C-terminal DYFP vector. Optionally, this protocol can be adapted to create probes with any tag.
It is possible to carry out step (b) on predesigned vectors that already contain a tag of choice, if available (see Note 13).
1. Design and order a PCR primer pair for amplification of the DYFP tag (see Note 14).
2. Amplify DYFP in a 100-µL PCR reaction from 1 ng of template vector according to standard procedures (see Note 15).
3. Resolve the PCR reaction on a 1% agarose gel electrophoresis. Stain the gel with ethidium bromide to determine the amount and size of the PCR product. 4. Using a UV transilluminator and a scalpel, cut out the band corresponding to the PCR product and extract it from the gel using a gel extraction kit according to the manufacturer's instructions. 5. Digest the extracted PCR product with SpeI and XbaI restriction endonucleases in a total reaction volume of 100 µL according to manufacturer's instructions. 6. Purify the digested PCR product using a PCR purification kit according to the manufacturer's instructions. 7. Run a 1% agarose gel electrophoresis with a small volume of purified digested PCR product and stain gel with ethidium bromide to estimate its amount.
Alternatively measure the concentration spectrophotometrically. Store the prepared PCR product at -20°C.

Analysis of the relative Rac activity in the cell membrane
The DPAKa(GBD)-DYFP probe is strongly enriched in the cell plasma membrane where it binds to the active, GTP-bound form of Rac1 GTPases (Fig. 2). Its average fluorescence intensity is more than 4 times higher in the membrane compared to the cytoplasmic background, and can attain up to 20 times higher values. In the following we describe how to monitor the DPAKa(GBD)-DYFP-expressing cells using confocal microscopy, and quantitatively analyse its distribution along the two-dimensional cross-section of the plasma membrane. 3. Analyse the cell images by image-processing software capable of automatically extracting the bright contour of the cell that corresponds to the plasma membrane. We use the ImageJ plugin QuimP, a software package which enables extraction of the cell contour, and sampling of the fluorescence intensity at an adjustable number of points along the contour (Fig. 3A) (22). QuimP is capable of analysing long series of images with a small extent of manual intervention (see Note 23). It consists of four sub-plugins that correspond to four stages of analysis.
a. Cell segmentation, as the first stage, is performed by the BOA sub-plugin.
Outputs are global measures, such as displacement of the cell centroid.
b. ECMM Mapping sub-plugin performs membrane tracking in which cell outlines are mapped between frames to extract local membrane velocities.
c. Measuring fluorescence intensities is performed by the ANA sub-plugin.
d. Finally, data analysis is performed by the Q Analysis sub-plugin.
Alternatively, data from the output files generated by BOA, ECMM and ANA plugins can be directly read into Matlab™, where custom scripts can be used for more in-depth analysis. For example, it is possible to plot the data representing the relative fluorescence intensity along the membrane either for single cell images (Fig. 3B) or for the entire time series in the form of a kymograph (Fig. 3C).

Membrane recruitment of active Rac upon chemoattractant stimulation
In addition to analyzing the activation of Rac1 upon agonist stimulation using a pull- In this type of experiment it is also necessary to determine the total fluorescence intensity in the cytoplasm for each frame, which is a feature integrated into QuimP.

1.
Caution should be exerted while working with phalloidin and methanol. Phalloidin

8.
Bacterial pellets obtained after protein expression can be stored at -20°C for processing at a later step.

9.
Parameters for sonication need to be adjusted to the device available. Bacterial lysis can be achieved by other methods (freezing and thawing, French press).
10. It is more convenient to store bacterial supernatant for attachment to Sepharose beads freshly prior to the pull-down assay rather than to store a batch of purified fusion protein, as this would need to be dialyzed to remove glutathione and will have to be attached to beads anyway. It is also not advisable to store coupled beads.
11. The amount of fusion protein pulled down using glutathione Sepharose beads can be estimated by comparison to a bovine serum albumin standard resolved on the same SDS-PAGE, followed by Coomassie staining.
12. Several exposure times should be used when documenting the detected Rac1 in the Western blots by conventional X-ray film development so that the density of the protein bands varies linearly with the amount of protein on the blot.
Alternatively, a quantitative detection system based on secondary fluorescent antibodies may be used.  which can be avoided by using objective-cooling devices. Also, in order to minimize phototoxicity, we recommend to use low-fluorescence growth medium, which can be obtained, e.g., from Formedium. 22. Such sub-Nyquist pixel size will also facilitate a possible digital post-processing of images, e.g. using deconvolution algorithms. In order to minimize phototoxicity and photobleaching, keep the illumination intensity at a practical minimum, while maintaining an acceptable signal-to-noise ratio in the image. 23. Results of the automatic contour extraction should be inspected frame-by-frame and corrected if necessary using tools provided by QuimP. For typical cells, we find manual intervention needed in less than 5% of analyzed images.