Search

The RBR family of ubiquitin ligases is a large and complex family with members present in animals, plants, fungi, and protists (17). RBR genes have been implicated in cancer and neurodegenerative diseases (e.g., parkin) (17-19, 21). Despite significant advances, much of the potential diversity in RBR protein function remains to be studied. Particularly interesting may be members of the Ariadne subfamily, since this subfamily appears to be the most ancient (1, 3, 17). In this report, we identify a putative Ariadne ubiquitin ligase, RbrA, in Dictyostelium. Dictyostelium discoideum Ax4 and DH1 were used as wild-type and parental strains. All strains were grown in HL5 with glucose (Qbiogene, Carlsbad, CA) or with Klebsiella aerogenes on agar plates as described previously (12). Multicellular development, preparation of conditioned starvation medium, phototaxis studies, Northern blot analysis, and preparation of cDNA clones were done as previously described (12, 16, 20). Restriction enzyme-mediated integration (REMI) was carried out as described previously (14). An rbrA mutant was found in a screen of REMI transformants as cells whose development is blocked at the finger/slug stage. rbrA mutant cells formed ripples by 2 h of development, which was earlier than wild type (Fig. ​(Fig.1);1); however, development then slowed compared to wild-type cells and fruiting bodies never formed. In mixtures of rbrA− (90%) and wild-type (10%) cells, early rippling was observed, but nearly all structures produced by the chimeras formed fruiting bodies (Fig. ​(Fig.1).1). The addition of wild-type conditioned medium to rbrA− cells did not cause a similar phenotype rescue. Thus, a small number of wild-type cells appeared to rescue the morphological developmental defect, and the rescue appeared to require cell-cell proximity. FIG. 1. Developmental phenotype of rbrA mutant cells. (A) Wild-type cells, rbrA mutant cells generated by REMI mutagenesis or rbrA mutant cells generated by replacement of a large fragment of the rbrA coding region with a blasticidin resistance cassette and ... When developed on filters, we observed that slugs formed from rbrA− cells were capable of migrating at least short distances. However, slugs were unable to phototax when developed on agarose in a dark chamber with a directional light source (Fig. ​(Fig.1B).1B). Slugs that formed from a mixture of rbrA− and wild-type (10%) cells migrated directionally toward the light source and traveled nearly as far as wild type (Fig. ​(Fig.1B).1B). These are likely chimeric slugs because at least 50% of the slugs that formed showed positive phototaxis, and other experiments showed that chimeric slugs do form (see Fig. ​Fig.44). FIG. 4. Cell sorting is altered in the rbrA mutant. (A) Wild-type and rbrA− cells were transformed with cell-type-specific GFP expression plasmids PsA-ubi-s65tGFP or 63-ubi-s65tGFP, which contained the GFP cDNA expressed by the prestalk-specific promoter ... Genomic rbrA DNA was cloned by digestion with BglII to generate plasmid pM10B (Fig. ​(Fig.2A).2A). The pM10B plasmid was linearized with BglII and used to transform a wild-type host strain (DH1). The resulting gene disruption by homologous recombination successfully recapitulated the original rbrA mutant phenotype. A second rbrA mutant was constructed by using a gene targeting plasmid, pRbrBsr, to replace a portion of the rbrA gene with a blasticidin-resistance cassette (Fig. ​(Fig.2B).2B). This rbrA− strain demonstrated a mutant phenotype nearly identical to the original rbrA mutant (Fig. ​(Fig.1).1). Expression of the rbrA cDNA complemented the defective developmental phenotype of rbrA− cells (Fig. ​(Fig.2C).2C). A 2.2-kb rbrA mRNA was present in vegetatively growing cells (0 h), increased in level at 9 h (Fig. ​(Fig.2D),2D), and remained at the 9-h level throughout the remainder of development (data not shown). rbrA mRNA was not present in rbrA− cells (Fig. ​(Fig.2D2D). FIG.2. rbrA gene. (A) Genomic map and REMI-generated mutation. REMI was carried out by electroporating EcoRI-linearized DIV2 (the mutagenic plasmid carrying the Dictyostelium pyr5-6 gene) into DH1 (pyr5-6−) cells along with the restriction enzyme MunI ... rbrA genomic and cDNA clones revealed a single long open reading frame interrupted by two introns (Fig. ​(Fig.2A).2A). The deduced amino acid sequence predicts a protein of 520 residues with a molecular mass of 61 kDa (Fig. ​(Fig.2E).2E). RbrA has strong sequence similarity to the Ariadne subfamily of the RBR family of E3 ubiquitin ligases (17). The characteristic series of acid-rich, RBR (for “RING, between rings, RING”), and coiled-coil/leucine rich motifs found in Ariadne subfamily members are present in RbrA (Fig. ​(Fig.2E).2E). RbrA has 34% identity (54% similarity) to human HHARI. These data suggest that rbrA encodes an E3 ligase. To investigate the function of RbrA, we examined the expression of developmentally regulated genes. Transcript levels for the prestalk-cell-specific ecmA gene (23) were lower in rbrA− cells compared to wild-type cells (Fig. ​(Fig.3).3). The prespore-specific pspD (25) was regulated temporally in developing rbrA− cells as in wild-type cells, but at much higher than wild-type levels (Fig. ​(Fig.3).3). In addition, transcript levels for gpaD (required for spore formation) (10) were elevated and gpaE (tip formation) (11) levels were lower during rbrA− cell development (Fig. ​(Fig.33). FIG. 3. Gene expression in rbrA− cells. Total RNA was prepared from wild-type cells (Ax4) at the indicated stages of development (in hours). RNA (5 μg) was resolved by electrophoresis through 1.2% agarose-formaldehyde gels, blotted onto nylon ... To investigate cell-type proportioning and cell sorting, rbrA− and parental cells were transformed with cell-type-specific green fluorescent protein (GFP) expression plasmids (5). When the ecmAO promoter was used to express GFP in wild-type cells, fluorescence appeared in the anterior prestalk region, as expected (6, 13) (Fig. ​(Fig.4A);4A); however, rbrA− slugs showed a very low level of fluorescence throughout the slug. GFP expressed from the pspA promoter marked the posterior prespore region in wild-type slugs, whereas the entire rbrA− slug was marked (Fig. ​(Fig.4A).4A). Thus, prespore cells were prevalent and appeared to dominate the entire rbrA− slug. To determine whether the initial cell-autonomous differentiation was altered in rbrA− cells (2, 8, 22), cells were starved at low cell density and the numbers of CP2-positive prestalk cells and SP70-positive prespore cells were counted (9). The percentage of CP2- and SP70-positive wild-type cells was similar to what we previously observed (4, 24). There was essentially no difference in the initial differentiation of rbrA− cells compared to wild-type cells (Table ​(Table11). TABLE 1. Cell-type proportioning is altered in rbrA− slugsa To determine the percentage of prestalk and prespore cells in slugs, wild-type and rbrA− slugs were dissociated into single cells, fixed, and stained for prespore (SP70) or prestalk (CP2) markers (7). The percentage of CP2- and SP70-positive wild-type cells was similar to what we previously observed (4). rbrA− slugs had a percentage of SP70-positive prespore cells that was similar to that of the wild type but had threefold-fewer CP2-positive prestalk cells than did the wild type (Table ​(Table1).1). Together, the data suggest that the initial cell-autonomous differentiation of rbrA− cells into CP2-positive prestalk cells and SP70-positive prespore cells is normal and that rbrA− slugs contain an abnormally low number of CP2-positive prestalk cells. To determine the effect of wild-type cells on rbrA− cell-fate choice, mixtures of wild-type and rbrA− cells were prepared in which one of the strains contained the GFP gene driven by the constitutively expressed actin-15 promoter (15). Wild-type and rbrA− cells were mixed in various proportions and developed on filters. Fluorescence imaging of chimeric slugs indicated that the wild-type cells preferentially located to the tip region, which is occupied by prestalk cells in pure wild-type slugs (Fig. ​(Fig.4B).4B). The rbrA− cells preferentially located to the posterior of the slug, which normally is populated with prespore cells. To investigate the ability of rbrA− cells to form spores during chimeric development, spore assays were carried out on fruiting bodies produced by chimeras (6). Experiments with chimeras containing 3 to 40% rbrA− cells produced less than 1% detergent-resistant spores with the rbrA− phenotype, suggesting that rbrA− spores were sensitive to detergent treatment. We therefore carried out additional experiments in which spores were plated before and after detergent treatment (Table ​(Table2).2). The results showed that rbrA− cells did not efficiently survive the terminal differentiation stages. TABLE 2. rbrA− cells do not efficiently produce viable sporesa The predicted amino acid sequence of Dictyostelium RbrA suggests that it is a member of the Ariadne subfamily of ubiquitin ligases. RbrA appears to be required for normal cell-type proportioning and cell sorting during multicellular development. Prestalk cell numbers are reduced in rbrA− slugs and these prestalk cells do not localize to the tip of slugs. Development terminates at the finger or slug stage, and these slugs do not phototax, possibly because the tip region does not properly form. In addition to being necessary for a normal percentage of prestalk cells and the organization of the slug, RbrA is also necessary for spore cell viability. RbrA thus affects multiple processes such as prestalk cell differentiation, pattern formation, and spore cell maturation.

The ubiquitin processing protease, UbpA (a homolog of yeast Ubp14 and human IsoT/USP5), is required for the growth-to-development transition of Dictyostelium. When ubpA¯ cells starve, they accumula...

Initial differentiation in Dictyostelium involves both asymmetric cell division and a cell cycle-dependent mechanism. We previously identified a gene, rtoA, which when disrupted randomizes the cell cycle-dependent mechanism without affecting either the underlying cell cycle or asymmetric differentiation. We find that in wild-type cells, RtoA levels vary during the cell cycle. Cytosolic pH, which normally varies with the cell cycle, is randomized in rtoA cells. The middle 60% of the RtoA protein is 10 tandem repeats of an 11 peptide-long serine-rich motif, which we find has a random coil structure. This domain catalyzes the fusion of phospholipid vesicles in vitro. Conversely, rtoA cells have a defect in the fusion of endocytic vesicles. They also have a decreased exocytosis rate, a decreased pH of endocytic/exocytic vesicles, and an increased average cytosolic pH. Our data indicate that the serine-rich domain of RtoA can mediate membrane fusion and that RtoA can increase the rate of vesicle fusion during processing of endoctyic vesicles. We hypothesize that RtoA modulates initial cell type choice by linking vegetative cell physiology to the cell cycle.

Although cell differentiation usually involves synthesis of new proteins, little is known about the role of protein degradation. In eukaryotes, conjugation to ubiquitin polymers often targets a protein for destruction. This process is regulated by deubiquitinating enzymes, which can disassemble ubiquitin polymers or ubiquitin-substrate conjugates. We find that a deubiquitinating enzyme, UbpA, is required for Dictyostelium development.ubpA cells have normal protein profiles on gels, grow normally, and show normal responses to starvation such as differentiation and secretion of conditioned medium factor. However,ubpA cells have defective aggregation, chemotaxis, cAMP relay, and cell adhesion. These defects result from low expression of cAMP pulse-induced genes such as those encoding the cAR1 cAMP receptor, phosphodiesterase, and the gp80 adhesion protein. Treatment ofubpA cells with pulses of exogenous cAMP allows them to aggregate and express these genes like wild-type cells, but they still fail to develop fruiting bodies. Unlike wild type, ubpAcells accumulate ubiquitin-containing species that comigrate with ubiquitin polymers, suggesting a defect in polyubiquitin metabolism. UbpA has sequence similarity with yeast Ubp14, which disassembles free ubiquitin chains. Yeast ubp14 cells have a defect in proteolysis, due to excess ubiquitin chains competing for substrate binding to proteasomes. Cross-species complementation and enzyme specificity assays indicate that UbpA and Ubp14 are functional homologs. We suggest that specific developmental transitions inDictyostelium require the degradation of specific proteins and that this process in turn requires the disassembly of polyubiquitin chains by UbpA.

We have developed a mutagenesis technique that uses antisense cDNA to identify genes required for development in Dictyostelium discoideum. We transformed Dictyostelium cells with a cDNA library made from the mRNA of vegetative and developing cells. The cDNA was cloned in an antisense orientation immediately downstream of a vegetative promoter, so that in transformed cells the promoter will drive the synthesis of an antisense RNA transcript. We find that individual transformants typically contain one or occasionally two antisense cDNAs. Using this mutagenesis technique, we have generated mutants that fail to aggregate, aggregate but fail to form fruiting bodies, or aggregate but form abnormal fruiting bodies. The individual cDNA molecules from the mutants were identified and cloned using PCR. Initial sequence analysis of the PCR products from 35 mutants has identified six novel Dictyostelium genes, each from a transformant with one antisense cDNA. When the PCR-isolated antisense cDNAs were ligated into the antisense vector and the resulting constructs transformed into cells, the phenotypes of the transformed cells matched those of the original mutants from which each cDNA was obtained. We made homologous recombinant gene disruption transformants for three of the novel genes, in each case generating mutants with phenotypes indistinguishable from those of the original antisense transformants. Shotgun antisense thus is a rapid way to identify genes in Dictyostelium and possibly other organisms.

Dictyostelium discoideum initiates development when cells overgrow their bacterial food source and starve. To coordinate development, the cells monitor the extracellular level of a protein, conditioned medium factor (CMF), secreted by starved cells. When a majority of the cells in a given area have starved, as signaled by CMF secretion, the extracellular level of CMF rises above a threshold value and permits aggregation of the starved cells. The cells aggregate using relayed pulses of cAMP as the chemoattractant. Cells in which CMF accumulation has been blocked by antisense do not aggregate except in the presence of exogenous CMF. We find that these cells are viable but do not chemotax towards cAMP. Videomicroscopy indicates that the inability of CMF antisense cells to chemotax is not due to a gross defect in motility, although both video and scanning electron microscopy indicate that CMF increases the frequency of pseudopod formation. The activations of Ca2+ influx, adenylyl cyclase, and guanylyl cyclase in response to a pulse of cAMP are strongly inhibited in cells lacking CMF, but are rescued by as little as 10 s exposure of cells to CMF. The activation of phospholipase C by cAMP is not affected by CMF. Northern blots indicate normal levels of the cAMP receptor mRNA in CMF antisense cells during development, while cAMP binding assays and Scatchard plots indicate that CMF antisense cells contain normal levels of the cAMP receptor. In Dictyostelium, both adenylyl and guanylyl cyclases are activated via G proteins. We find that the interaction of the cAMP receptor with G proteins in vitro is not measurably affected by CMF, whereas the activation of adenylyl cyclase by G proteins requires cells to have been exposed to CMF. CMF thus appears to regulate aggregation by regulating an early step of cAMP signal transduction.

When the unicellular eukaryote Dictyostelium discoideum starves, it senses the local density of other starving cells by simultaneously secreting and sensing a glycoprotein called conditioned medium factor (CMF). When the density of starving cells is high, the corresponding high density of CMF permits signal transduction through cAR1, the chemoattractant cAMP receptor. cAR1 activates a heterotrimeric G protein whose alpha-subunit is Galpha2. CMF regulates cAMP signal transduction in part by regulating the lifetime of the cAMP-stimulated Galpha2-GTP configuration. We find here that guanosine 5'-3-O-(thio)triphosphate (GTPgammaS) inhibits the binding of CMF to membranes, suggesting that the putative CMF receptor is coupled to a G protein. Cells lacking Galpha1 (Galpha1 null) do not exhibit GTPgammaS inhibition of CMF binding and do not exhibit CMF regulation of cAMP signal transduction, suggesting that the putative CMF receptor interacts with Galpha1. Work by others has suggested that Galpha1 inhibits phospholipase C (PLC), yet when cells lacking either Galpha1 or PLC were starved at high cell densities (and thus in the presence of CMF), they developed normally and had normal cAMP signal transduction. We find that CMF activates PLC. Galpha1 null cells starved in the absence or presence of CMF behave in a manner similar to control cells starved in the presence of CMF in that they extend pseudopods, have an activated PLC, have a low cAMP-stimulated GTPase, permit cAMP signal transduction, and aggregate. Cells lacking Gbeta have a low PLC activity that cannot be stimulated by CMF. Cells lacking PLC exhibit IP3 levels and cAMP-stimulated GTP hydrolysis rates intermediate to what is observed in wild-type cells starved in the absence or in the presence of an optimal amount of CMF. We hypothesize that CMF binds to its receptor, releasing Gbetagamma from Galpha1. This activates PLC, which causes the Galpha2 GTPase to be inhibited, prolonging the lifetime of the cAMP-activated Galpha2-GTP configuration. This, in turn, allows cAR1-mediated cAMP signal transduction to take place.