Genetic Engineering of Dictyostelium discoideum Cells Based on Selection and Growth on Bacteria

Dictyostelium discoideum is an intriguing model organism for the study of cell differentiation processes during development, cell signaling, and other important cellular biology questions. The technologies available to genetically manipulate Dictyostelium cells are well-developed. Transfections can be performed using different selectable markers and marker re-cycling, including homologous recombination and insertional mutagenesis. This is supported by a well-annotated genome. However, these approaches are optimized for axenic cell lines growing in liquid cultures and are difficult to apply to non-axenic wild-type cells, which feed only on bacteria. The mutations that are present in axenic strains disturb Ras signaling, causing excessive macropinocytosis required for feeding, and impair cell migration, which confounds the interpretation of signal transduction and chemotaxis experiments in those strains. Earlier attempts to genetically manipulate non-axenic cells have lacked efficiency and required complex experimental procedures. We have developed a simple transfection protocol that, for the first time, overcomes these limitations. Those series of large improvements to Dictyostelium molecular genetics allow wild-type cells to be manipulated as easily as standard laboratory strains. In addition to the advantages for studying uncorrupted signaling and motility processes, mutants that disrupt macropinocytosis-based growth can now be readily isolated. Furthermore, the entire transfection workflow is greatly accelerated, with recombinant cells that can be generated in days rather than weeks. Another advantage is that molecular genetics can further be performed with freshly isolated wild-type Dictyostelium samples from the environment. This can help to extend the scope of approaches used in these research areas.


Introduction
The Dictyostelium genus are soil-living social amoeba that mainly feed on bacteria. Being placed in the phylum Amoebozoa, a large number of species have been isolated that can be grouped into four different clades 1 . The species Dictyostelium discoideum (D. discoideum) has become a popular model organism to study complex cellular processes such as cell migration and phagocytosis. To control and standardize experimental conditions, axenic cell lines have been developed that are able to grow in complex or defined liquid medium in the absence of bacteria 2 . Of particular importance are the Ax2, Ax3, and Ax4 strains, which were all generated in the 1970s and ultimately derived from a single wild isolate NC4 3 . Tools for genetic engineering were developed in these axenic strains, resulting in the first published knockout in 1987 4,5 . Protocols were further developed and optimized for use under axenic conditions 6,7 . Adaptation of these protocols to wild-type D. discoideum strains that are not able to grow in liquid broth have been attempted by several laboratories. However, this has not become fully successful since the transfection protocols are complex and lack efficiency, in part due to the capacity of the bacteria to act as a sink for the selective reagents 8,9 . As a result, essentially all molecular data on D. discoideum comes from descendants of a single wild-type isolate. We wanted to overcome this limitation and develop a method to genetically modify D. discoideum cells independent of their ability to grow in liquid medium. The need for such a method can be explained by the observation that it was assumed in the past that the mutations allowing axenic growth were mainly neutral and did not impair cell physiology. This supposition is only partially correct. In general, there are two notable differences; first, between the different isolated axenic strains, and second, when these axenic strains are compared with non-axenic wild isolates . The deletion of the enzyme in all axenic strains leads to excessive Ras activity manifested as the formation of large active Ras patches. These enlarged Ras patches lead to the accumulation of PIP3 in the plasma membrane. Those coincidental appearing patches of PIP3 and active Ras are a template for the formation of a circular ruffle that eventually closes and leads to the formation of macropinosomes 10 . The consequence is an excessive increase in macropinocytic activity. Macropinocytosis is an actin-driven process. A competition for cytoskeletal components for the formation of either macropinosomes or pseudopods is the result. Its impact on cell behaviour is reflected in the almost complete prevention of chemotaxis of vegetative cells to folate 11 . The massively enlarged PIP3 patches are very persistent. Even in starved cells, the PIP3 patches remain and can be misinterpreted as pseudopods, which can cause problems interpreting studies on chemotaxis to cAMP.
In some cases, the NF1 mutation is experimentally useful. This leads us to a second motivation for developing a transfection method for bacterially grown D. discoideum cells, since the increase in macropinocytosis rate makes axenic cells valuable for investigating fundamental aspects of this process 12 . However, mutation in the genes required for macropinocytosis, such as Ras and PI3-kinases 10 , have almost abolished axenic growth, making it necessary to manipulate these cells through growth on bacteria. Another reason that renders bacteria-based transfections valuable is the increasing use of Dictyostelids to explore questions in the evolution of multi-cellularity 13,14 , kin recognition 15,16 , and altruistic cellular behaviour, which mainly depend on the use of freshly isolated wild-type isolates 17 . All mentioned research areas can be facilitated by efficient methods for genetic manipulation of wild isolates, which are non-axenic and do not grow in liquid broth.
Our protocols allow for overcoming of the described limitations. Taken together, the possibility of performing genetic manipulations with bacterial grown D. discoideum cells holds benefits for all Dictyostelium researchers, even if it is just the increased speed of the selection process due to faster growth of the amoeba (4 h doubling time) on bacteria compared to growth in axenic media (10 h doubling time). NOTE: Due to their faster growth rate, the bacteria initially produce a confluent lawn, and the amoebae subsequently "clear" the plate of the bacteria. The time required for this process can differ depending on the strain background and ability of mutant cells to grow on bacteria.

Selection of transfectants: depending on whether transfection is aimed to generate a knock-out, knock-in, or act5 knock-in, or
to express a fluorescent reporter protein from an extrachromosomal plasmid, carry out the selection process through one of the following described methods. NOTE: Unlike axenically grown cells, bacterially grown cells are highly resistant to blasticidin. Selection, therefore, is always carried out using G418 or hygromycin (see discussion). 1. Knock-outs, knock-ins, and act5 knock-ins 1. Detach the cells carefully from the Petri dish by repeatedly forcing the liquid from a pipette onto the surface. 2. Set up three dilutions in the SorMC K. aerogenes suspension (OD 600 = 2) and add the selective agent according to the resistance used (see Figure 1). 2. Extrachromosomal plasmids 1. Add the selectable marker directly to the 10 mL dish (see Figure 1). NOTE: There is no need to set up dilutions, since there is no desire for clonal populations. The selection process for extrachromosomal plasmids is faster due to high copy numbers present in D. discoideum cells. Transfectants can be expected after 32 h to 2 days. CAUTION: The antibiotics used as selectable marker are toxic. Wear gloves.

Screen the obtained clones for positive tranfectants. For knock-out or knock-in attempts, follow the instructions in step 2.5.1. Check transfection success for extrachromosomal plasmids using the instructions in step 2.5.2.
1. For knock-outs, knock-ins, and act5 knock-ins, perform the initial screen via PCR to confirm integration of the construct into the correct genomic locus. NOTE: To maximize the likelihood of clonal populations, use the highest dilution possible that yields transfectants after selection. Aim for plates that are a maximum one-third of wells occupied. 1. To expand clonal populations, transfer clones that have grown up after selection from the 96-well tissue-culture plate into a 12-well tissue-culture plate to grow enough cells for the isolation of genomic DNA. Supply each well with 1 mL of SorMC K. aerogenes (OD 600 = 2) and fresh selectable marker. NOTE: 1 day is usually sufficient to obtain a confluent well suitable for DNA isolation. 2. To perform mini genomic DNA isolation, harvest the cells of a confluent well and isolate genomic DNA using a mini DNA extraction kit following the manufacturer's instructions. 3. Use the isolated genomic DNA together with suitable primers and Taq-polymerase (see Table of

Representative Results
Extrachromosomal plasmids are used for reporter studies, which aim to identify the localization of certain proteins inside a cell or changes in cellular structure of mutant cells. For many approaches, such as monitoring of the cell cycle, it is crucial to express two reporters at the same time. This is now possible using our dual reporter extrachromosomal plasmid system ( Table 1). On day 1, cells were transfected before adding the selectable marker G418 after 5 h (Figure 1). In the example, NC4, DdB, Ax2, and the independently derived wild isolates V12M2 and WS2162 (Supplementary Table 1) were transfected with the plasmid pPI289, which encodes for GFP-TubulinA, a marker for microtubules and mCherry-PCNA, a protein that is used for monitoring the cell cycle (Figure 2A). After 32 h, the cells were observed under the microscope. The majority of cells expressed both fluorescent-labeled fusion proteins, consistent with previous reports that expression of two reporters from the same plasmid shows similar expression levels, which is nearly impossible when using two different plasmids 18,19 . A representative cell for each cell line (NC4, DdB, Ax2, V12M2, and WS2162) expressing the desired dual reporter is shown in Figure 2B. The transfection efficiencies are summarized in Figure2C. NC4-derived cell lines show the best transfection efficiencies. However, for cell lines V12M2 and WS2162, a considerably high number of transfectants was obtained.      , preventing their general adoption. The generated materials and protocols presented here can be used for any D. discoideum strain independent of its ability to grow in liquid medium. It should be mentioned that this protocol is optimized for NC4-derived cell lines. Transfection efficiencies for freshly isolated strains from the wild differ from NC4, as we have observed before and are shown here for V12M2 and WS2162 8,9 . The electroporation conditions seem in particular to have a considerable influence on transfection efficiencies and may require further optimization for some strains. In general, a sufficient amount of transfectants have been observed in all strains tested so far, showing that these methods are workable. The number of positive transfectants obtained using NC4derived strains is higher compared to other non-axenic wild-type strains, but in all cases, sufficient numbers of transfectants are obtained to allow for further experiments. This, plus the simplicity of our protocol, is the major improvement compared to earlier attempts 8,9 .
These new non-axenic protocols cover all standard genetic procedures and add further advantages, since transfections can be performed rapidly and efficiently in parallel. Positive clones can be obtained in days rather than weeks, since growth on bacteria halves the division time. The newly established plasmids also work under axenic conditions and can be routinely used under both growth conditions, which is another advancement of this methology 22 . As introduced in the protocol, the plasmid used for selection on bacteria have special requirements that are crucial for the success of the transfection. In particular, the promoters driving the expression and resistance cassettes are important for successful transfection. The often used act6 (actin 6) promoter 24 for driving the resistance or expression cassettes lacks efficiency when cells are grown on bacteria.
In our plasmid system, the highly active act15 (actin 15) promoter drives all expression cassettes, while the resistance cassettes are under the control of a coA (coactosin A) promoter, both of which are active under axenic conditions and in cells grown on bacteria. Requirements for the expression of reporter constructs as well as knock-in and knock-out constructs make it necessary to use our plasmid repertoire, but unfortunately limits the use of vectors already created that use the inefficient act6 promoter.
Promoter efficiencies are especially critical for transfections that depend on a single integration event into the correct genomic locus. Due to our improvement of promoters driving the resistance gene expression, enough resistance protein is produced from single locus integrations. A major concern was the favoring of multiple integrations into the genome when using hygromycin or G418 as described. No favoring of multiple integrations has been observed so far, as reported previously for G418 selections using older promoter systems 25 . This means that both hygromycin and G418 are suitable selectable markers for the generation of clean knock-outs and knock-ins. Unfortunately, the selectable marker blasticidin does not work under non-axenic conditions. This is a major disadvantage of our method, since the blasticidin resistance cassette is routinely used to generate knock-out constructs in D. discoideum. Constructs that were already generated with blasticidin resistance cassettes will need to be rederived using one of the workable selectable markers. Another possibility to overcome this limitation is to combine the recently established axenic D. discoideum CRISPR technology with this transfection protocol 26 . The generation of suitable, single guide RNAs (sgRNAs) is simpler and faster than the reconstruction of complete knock-out constructs. For future directions, the possibility of generating multiple knockouts using CRISPR/Cas9 combined with this transfection protocol is appealing and may pave the way for many researchers in the Dictyostelium community. However, the workability of the established transient expression system used for CRISPR/Cas9 in axenic grown cells should be carefully examined.
The act5 knock-in system presented offers a reliable and safe integration system for the generation of stable cell lines in D. discoideum, with the advantage of similar sites established in other organisms 22 . It also depends on a single integration event in the genome and offers many possibilities for different research directions. The act5 promoter is strongly active and guarantees almost homogenous expression 27 independent of the cultivation conditions. Fluorescent reporter proteins can easily be integrated into this safe haven locus using the engineered targeting plasmids. This can be useful, for example, for cell-tracking purposes, as shown here in a mix experiment. The cells show minimal cell-to-cell expression variability, which can assist in automated cell tracking. Importantly, the insertion of a desired sequence into the act5 locus appears phenotypically neutral 28,29 . As the expression does not depend on a selectable marker to maintain protein expression, as seen in random integrants or extrachromosomal vector-bearing cell lines, the act5 system can be useful for rescue experiments, as well. Since the cell lines are homogenous, all cells can be analyzed. This is not possible using an extrachromosomal plasmid for expression, in which there is considerable heterogeneity in expression.
In addition to these general improvements for all strains, the ability to use non-axenic wild-type cells for molecular research allows an assessment of the effects of accumulated mutations in current laboratory strains. Multiple genetic rearrangements has been found since the adoption of the Ax2 strain in 1970, as shown by previously published microarray analyses 26 . Synthetic phenotypes are observed because of the presence of these mutations. For example, a reported phg2 mutant shows decreased adhesion in one laboratory strain background and increased adhesion in another 3 . Such conflicting data can now be resolved by repeating the experiment in the common ancestor strain (in this case, DdB). The strength of this approach has been recently shown for the small GTPase RasS 30,31 . The ability to perform genetic experiments in any D. discoideum strain expands the potential of new research directions that depend on the use of wild isolates, notably those involving social evolution, kin recognition, and the sexual cycle 22 .

Disclosures
The authors have no conflicts of interest to disclose.