Cell type proportioning in Dictyostelium slugs: lack of regulation within a 2.5-fold tolerance range.

The proportion of prestalk and prespore cells in Dictyostelium discoideum slugs is often cited as an example of "almost perfect" regulation. The pattern is similar over a very wide range of cell number; furthermore, removal of either of the cell types leads to compensatory transdifferentiation. Several studies of Dictyostelium fruiting bodies, however, have suggested that proportioning in Dictyostelium differs systematically from true constancy. We have confirmed this in the slug stage using a short-lived beta-galactosidase as a reporter of the prestalk specific ecmA gene expression: the prestalk proportion decreases from 24+/-5% in slugs of 10(3) cells to 10+/-3% when 10(5) cells are present. Regeneration experiments suggest that this difference is not due to a modulation of the proportioning set-point by size, as one might have expected; instead there appears to be a regulatory "tolerance zone" at all sizes. After amputation of the whole posterior region, transdifferentiation stops after the fraction of prestalk has been reduced from 100% to 28+/-20%, well above the initial value of 10+/-3%, while after anterior removal the transdifferentiation endpoint is about 10%. Most strikingly, we find no regulation at all after partial amputations of the prespore region. It seems that any prestalk proportion is stable between a approximately 10% lower threshold and a approximately 30% upper threshold. To explain this, we propose a regulation mechanism based on a negative feedback plus cell type bistability. In both intact and regenerating slugs we find that the slug morphology is regulated so that the length-to-width ratio of the anterior region is constant.

Biphasic expression of rnrB in Dictyostelium discoideum suggests a direct relationship between cell cycle control and cell differentiation.

Cell differentiation in Dictyostelium is strongly affected by the cell cycle. Cell cycle control is well-understood in other systems, but this has had almost no impact on the study of Dictyostelium cell differentiation, in part because the cell cycle in Dictyostelium is unusual, lacking a G1 phase. Here we describe the cell-cycle regulated expression of rnrB, which codes for the small subunit of ribonucleotide reductase and is a marker of late G1 in many systems. There appear to be two expression peaks, one in mid-G2 and the other near the G2/M transition. Using Xgal/anti-BrdU double staining, we show that cells in asynchronously growing cultures express in both phases, with a gap between them during which the gene is transcriptionally silent. Cold-synchronized cells show exclusively G2/M expression, while mid-G2 expression is seen in high-density synchronized cells and can also be inferred in cells undergoing synchronization by either method. rnrB expression occurs in other systems shortly after cells pass a point (the "restriction point" or "start") at which they commit to complete their current cell cycle. We demonstrate a similar commitment point in Dictyostelium and show that this occurs shortly before the mid-G2 rnrB expression peak. The Dictyostelium cell cycle thus appears to include a well-defined though inconspicuous event, between early and mid-G2, with some features which are normally associated with the G1/S transition. Others have described a switch from stalk to spore differentiation preference at about this time. Since Dictyostelium cells switch back from spore to stalk preference approximately at the G2/M rnrB expression maximum, cell differentiation as well as rnrB expression may be regulated directly by fundamental cell cycle control processes.

Inducible expression of exogenous genes in Dictyostelium discoideum using the ribonucleotide reductase promoter.

We report here the development of a regulated gene expression system for Dictyostelium discoideum based on the DNA-damage inducibility of the rnrB gene. rnrB, which codes for the small subunit of the enzyme ribonucleotide reductase, responds to DNA-damaging agents at all stages of the D.discoideum life cycle. Doses that have little effect on development have previously been shown to increase the level of the rnrB transcript by up to 15-fold. Here we show that all elements necessary for DNA-damage induction are contained in a 450 bp promoter fragment. We used a fusion of the rnrB promoter with the gene encoding GFP to demonstrate an up to 10-fold induction at the RNA level, which appears in all aspects similar to induction of the endogenous rnrB transcript. Using a fusion with the lacZ gene we observed an up to 7-fold induction at the protein level. These results indicate that the rnrB promoter can be used to regulate the expression of specific genes in D.discoideum. This controllable gene expression system provides the following new characteristics: the induction is rapid, taking place in the order of minutes, and the promoter is responsive at all stages of the D.discoideum life cycle.

Green fluorescent proteins with short half-lives as reporters in Dictyostelium discoideum.

We describe two modifications of the popular reporter green fluorescent protein (GFP) which have short half-lives in our system, the cellular slime mould Dictyostelium discoideum. One of these bears an N-terminal ubiquitin; this GFP was originally planned to be a substrate of the "N-end-rule" pathway, but deubiquitination does not seem to occur, and a degradation by the UFD (ubiquitin-fusion-degradation pathway seems more probable. The protein half-life is about 3-5 h. The second construct has an N-terminus derived from the L11 ribosomal protein; it is transported to the nucleus and broken down much more rapidly than the ubiquitin fusion (protein half-life about 30 min). We show examples of the use of these reporters in the study of gene expression in Dictyostelium.

UGUS, a reporter for use with destabilizing N-termini.

Due to constraints in vector construction, reporter polypeptides often carry N-terminal sequences of extraneous origin. Since protein half-life can be influenced by small determinants in the N-terminus, such foreign sequences can destabilize proteins and compromise results of reporter-based studies. We provide a real-life example of this problem (destabilizing sequences derived from a ribosomal protein) and show that it can be solved with the ubiquitin fusion technique, in which ubiquitin sequences are placed upstream of the reporter, in our case beta-glucuronidase. Post-translational processing by characterized pathways removes the ubiquitin together with destabilizing sequences, generating a stable reporter whose N-terminus is constant.

Centrosome positioning and directionality of cell movements.

In several cell types, an intriguing correlation exists between the position of the centrosome and the direction of cell movement: the centrosome is located behind the leading edge, suggesting that it serves as a steering device for directional movement. A logical extension of this suggestion is that a change in the direction of cell movement is preceded by a reorientation, or shift, of the centrosome in the intended direction of movement. We have used a fusion protein of green fluorescent protein (GFP) and gamma-tubulin to label the centrosome in migrating amoebae of Dictyostelium discoideum, allowing us to determine the relationship of centrosome positioning and the direction of cell movement with high spatial and temporal resolution in living cells. We find that the extension of a new pseudopod in a migrating cell precedes centrosome repositioning. An average of 12 sec elapses between the initiation of pseudopod extension and reorientation of the centrosome. If no reorientation occurs within approximately 30 sec, the pseudopod is retracted. Thus the centrosome does not direct a cell's migration. However, its repositioning stabilizes a chosen direction of movement, most probably by means of the microtubule system.

Wild-type strains of Dictyostelium discoideum can be transformed using a novel selection cassette driven by the promoter of the ribosomal V18 gene.

Almost all methods for transformation of the social ameba Dictyostelium discoideum rely on axenic growth, that is, growth in a synthetic medium, for at least part of the procedure. Axenic growth requires several mutations. Here we describe a procedure that can be used to transform wild-type strains which are able to grow only on the natural food source, bacteria. The method relies on a new selection cassette driven by the V18 promoter, a promoter that we show is substantially more active during growth on bacteria than the actin-6 promoter, which is widely used for axenic transformation. The procedure gives transformation frequencies of about 10(-5) with both strains Ax2 (capable of axenic growth) and NC4 (capable of growth only on bacteria). Using this vector, we have obtained NC4 strains carrying several beta-galactosidase reporter cassettes. Our vector can also be used in axenic transformations.

Cell-density-dependent repression of discoidin in Dictyostelium discoideum.

When Dictyostelium discoideum cells are grown on bacteria, their natural food source, the discoidin genes are induced by cell-density-sensing factors before the food supply is exhausted [11, 18], and expression increases continuously thereafter. This regulation pattern is changed when cells are grown in axenic medium: the discoidins are induced at a considerably lower cell density and are no longer expressed in stationary phase [13]. We have investigated this phenomenon further and show that repression begins when cells are still in exponential growth. It occurs at the level of transcription and involves an element of the discoidin I gamma promoter for which no function has previously been described. Since the effect of high cell density can be mimicked by conditioned medium, it appears that the repression is due to an extracellular signal. This signal is neither ammonia, nor folate, nor cAMP, the known repressors of discoidin expression.

A protein kinase from Dictyostelium discoideum with an unusual acidic repeat domain.

DdKinX codes for 1093 amino acids which are organized in four regions: the N-terminal catalytic domain, a region containing 30% acidic amino acids, tandem repeats of the motif VKVEEPVEE and the C-terminus. Identity with other protein kinases is 25 to 30%. Descendent trees show that DdKinX does not belong to any of the known kinase branches.

The 'prespore-like cells' of Dictyostelium have ceased to express a prespore gene: analysis using short-lived beta-galactosidases as reporters.

In transgenic strains of Dictyostelium discoideum that express beta-galactosidase under the control of a prespore-specific promoter, only early slugs show reporter confined to the prespore zone. As slugs migrate beta-galactosidase-positive cells accumulate in the prestalk zone; ultimately, there may be so many that the prestalk-prespore boundary is no longer distinguishable (Harwood, A., Early, A., Jermyn, K. and Williams, J. (1991) Differentiation 46, 7-13). It is not clear whether these 'anomalous' reporter-positive cells currently express prespore genes; another possibility is that they are ex-prespore cells that have transformed to prestalk and sorted to the prestalk zone (Sternfeld, J. (1993) Roux Archiv. Dev. Biol. 201, 354-363), while retaining their previously produced reporter. To test the activity of the prespore genes in these cells, we have made prespore reporter constructs whose products decay quickly; these are based on constructs used to investigate protein turnover in yeast (Bachmair, A., Finley, D. and Varshavsky, A. (1986) Science 234, 179-186). In strains bearing such constructs, beta-galactosidase-positive cells do not appear in the prestalk zone. The apparent deterioration of the prestalk/prespore pattern in older slugs is thus an artefact of reporter stability.

Mutants of Dictyostelium discoideum with defects in the regulation of discoidin I expression.

Discoidins are proteins, coded by a multigene family, which are regulated by extracellular factors during growth and development of Dictyostelium discoideum. In this paper we describe the isolation and characterization of mutants which misregulate the expression of the discoidin I subgroup. One mutant (III29) induces discoidin I during late growth phase but does not express it during development. Another mutant (VI41) has significantly reduced discoidin levels under all conditions tested, while two mutants (VI88 and X27) express discoidin early during exponential growth and accumulate more discoidin protein than the wild type. The defects are due to abnormal regulation of transcription in all mutants except VI41. Experiments in which mutants and wild type are mixed suggest that the mutant phenotypes are not caused by changes in extracellular signals. Since multiple members of the multigene family are affected, it can be concluded that the intracellular signals regulating discoidin expression are changed rather than the genes themselves. The mutants are thus likely to have defects in the reception or intracellular processing of environmental signals.

Cell sorting within the prespore zone of Dictyostelium discoideum.

Dictyostelium discoideum forms elongate cell aggregates called "slugs" which migrate over the substrate before completing their conversion into fruiting bodies. Prespore cells are found in a zone which occupies the rear four-fifths of the slug. Both front- and rear-prespore cells, marked by a bacterial beta-galactosidase gene, sort out to their original positions in experiments in which slugs are reconstituted from disaggregated tissue. When cells from the rear of the prespore zone are transplanted to the middle or front, sorting is also observed: the transplanted cells return rapidly to the rear. Cells from the front of the prespore zone, however, were not observed to "home" to the front after transplantation to the rear. Since front-prespore cells sort out in disaggregation/reaggregation experiments, but fail to do so after transplantation to the rear, it is possible that the transplanted cells are converted to rear-prespore cells by extracellular signals present in the rear of the slug. In an experiment designed to test this hypothesis, front cells were transplanted to the rear and the host and transplant together then subjected to disaggregation/reaggregation. The results showed that front-prespore cells had not been converted to rear-prespore cells. Instead, there was an unanticipated effect: cells placed in the rear of the prespore zone underwent an anterior shift in positional preference, while cells placed in the front of the prespore zone showed a posterior shift. The specific sorting properties of front- and rear-prespore cells thus do not appear to result from the action of positional signals; positional signals destabilize rather than reinforce sorting preferences. Our observations are consistent with a model in which innate differences among cells bias them to differentiate as front-prespore or rear-prespore types, but the proportions of these types are also modulated by a negative-feedback mechanism.

Use of a transactive regulatory mutant of Dictyostelium discoideum in a eucaryotic expression system.

The discoidin proteins of Dictyostelium discoideum are highly expressed during development. The Disc I gamma promoter allows the regulation of heterologous protein expression by experimental conditions. We report conditions under which the promoter activity is efficiently repressed during growth in the wildtype strain AX2. In addition we show that a mutant which overexpresses the discoidins also overexpresses the reporter genes beta-galactosidase, luciferase and CAT 10- to 100-fold when these are placed under the control of a Disc I gamma promoter. This system may be generally useful for the overexpression of genes in Dictyostelium, both for functional studies in vivo and for the production of heterologous proteins for purification.

Cell sorting within the prestalk zone of Dictyostelium discoideum.

The prestalk zone of slugs of Dictyostelium discoideum has been shown to contain three subregions in which the extracellular matrix genes ecmA and ecmB are differentially expressed; it is generally thought that these regions are defined by extracellular signals. Using beta-galactosidase as a cell marker, we have shown that cells can sort specifically to all three regions. Cells from the posterior-prestalk zone ("prestalk 0 zone") which are injected into the slug tip move within 60 min back to their position of origin. When cells from the anterior prestalk zone (presumably containing a mixture of ecmA and ecmB expressers) are transplanted to the posterior prestalk zone, they move to the tip ("prestalk A zone") within 1 h and about 30 min subsequently are often found in a cone-shaped region within the tip ("prestalk B zone"). Cells transplanted to their own positions do not move significantly within this period. Since the subregions of the prestalk zone can be defined by sorting, it is possible that they are normally formed in this way rather than by position-dependent signals. Cells transplanted without a change in anterior-posterior position and cells which have sorted back to their positions of origin eventually spread out throughout the prestalk zone. This suggests that sorting preferences of cells are respecified. When posterior prestalk cells are transplanted to the prespore zone, respecification of sorting preference is suspended until the cells return to the prestalk zone and anterior-prestalk cells acquire posterior-prestalk sorting preferences.

Hydra transplantation phenomena and the mechanism of hydra head regeneration. I. Properties of the head inhibition.

Measurements were made of the "head inhibition" and the "head inhibition gradient" in Hydra. Following decapitation, the head inhibition decays with a half-time of 2-3 hr. The slope of the inhibition gradient increases about twofold when the temperature is raised from 20 to 24 degrees C; the increase (at the higher temperature) occurs with a half-time of about 1.2 hr. These results are consistent with the idea that the head inhibition and the head inhibition gradient are due to a diffusing substance, made in the head and broken down in the body of Hydra, with a half-life of about 2 hr at 20 degrees C and about 1 hr at 24 degrees C. Calculations suggest a diffusion constant of 1-3 X 10(-6) cm2/sec and a gradient range (ratio of maximum to minimum concentration) of about 2. During head regeneration, the head inhibition returns much more slowly than one would expect if its production were switched on at the time of "head determination." The slow return of the inhibition can be explained if one assumes that determination is due to the "activation" of a few cells, while the restoration of inhibition depends on the lateral expansion of the activated region. This behavior is observed in a "proportion-regulating" diffusion-reaction model proposed by Gierer and Meinhardt, 1972, Kybernetik 12, 30-39. Transplants to beheaded animals demonstrate a substantial inhibition gradient even at 6 hr after decapitation, a time at which most of the inhibitor from the original head should have decayed and no substantial amount of inhibitor is expected from the regenerated head. Experiments in which parts of a hydra's body are removed suggest that this inhibition gradient is due to the production of some inhibitor in the body, with larger amounts produced in apical portions.

Hydra transplantation phenomena and the mechanism of Hydra head regeneration. II. Properties of the head activation.

Measurements were made of the "head activation" in transplanted fragments of Hydra tissue. The measurements confirm that head activation is graded in the body and that activation increases during head regeneration. Experiments of two different sorts show that regeneration-specific activation increase is confined to the presumptive head zone. When head formation is initiated and subsequently blocked, the activation decays back to its original level in about 12 hr; this is three times faster than the activation decay observed in body tissue moved to a more basal position. Experiments on foot formation phenomena show a similar lability difference. It thus appears that regeneration-specific activation increase involves a mechanism different from the one responsible for the gradient of activation in the body. Cutting produces a local increase of the head and/or foot activation level; this increase decays in about 12 hr. These results and those of the preceding paper are consistent with a version of the Gierer-Meinhardt model which also accounts for the regulation of the head/body proportion in Hydra.

The prestalk-prespore pattern in cellular slime molds.

In cellular slime molds the slugs become divided into two regions with different properties, and anterior prestalk-zone and a posterior prespore zone. Although the cells in these zones are normally destined to form the stalk cells and spores of the fruiting body, respectively, they are not irreversibly committed to one sort of differentiation or the other during the slug stage. The volume ratio of the two zones remains almost constant over a wide range of slug sizes. If the prestalk-prespore pattern is distrubed by removing tissue from the slug, conversion of tissue from prestalk to prespore or vice versa occurs as necessary to restore a normal pattern with normal proportions. Conversions also occur in both directions during normal development. The initial formation of the prestalk-prespore pattern may well involve sorting-out, but other mechanisms must be invoked to account for regulation. We describe three different models of the generation of the prestalk-prespore pattern, the'cell-contact model' of McMahon, in which pattern is created by interactions of cells with their immediate neighbors, the 'positional-information model' of various authors, in which pattern formation involves an overall gradient and a gradient-reading mechanism, and the 'activator-inhibitor model' of Gierer and Meinhardt, in which the prestalk-prespore pattern is formed by a system of diffusible substances that affect one another's production. The activator-inhibitor model is the most successful of the models at describing the known features of the prestalk-prespore pattern. The various models lead to a number of distinctive predictions. According to the cell-contact model, small transplants may cause gross changes in the prestalk-prespore pattern, and mutants may exist which severely disrupt pattern formation even if diluted with a large excess of wild-type cells. Positional-information models predict the existence of 'gradient-reading mutants'; slugs that are a mixture of such mutants and wild-type cells would show two prestalk-prespore boundaries, one at the mutant and one at the normal position. Both the activator-inhibitor model and some versions of the positional-information model predict that small transplants will sometimes induce accessory prestalk or prespore zones; the quantitative characteristics of these effects may allow one to make a case in favor of one or other of the two models. Finally, the activator-inhibitor model leads one to expect that mutants may be isolated which normally show accessory prestalk or prespore zones. A search for these phenomena may help determine whether the activator-inhibitor model will continue to enjoy its present preeminent position.