Analysis of gene expression in rapidly developing mutants of Dictyostelium discoideum

CUL-4A stimulates ubiquitylation and degradation of the HOXA9 homeodomain protein

The HOXA9 homeodomain protein is a key regulator of hematopoiesis and embryonic development. HOXA9 is expressed in primitive hematopoietic cells, and its prompt downregulation is associated with myelocytic maturation. Although transcriptional inactivation of HOXA9 during hematopoietic differentiation has been established, little is known about the biochemical mechanisms underlying the subsequent removal of HOXA9 protein. Here we report that the CUL-4A ubiquitylation machinery controls the stability of HOXA9 by promoting its ubiquitylation and proteasome-dependent degradation. The homeodomain of HOXA9 is responsible for CUL-4A-mediated degradation. Interfering CUL-4A biosynthesis by ectopic expression or by RNA-mediated interference resulted in alterations of the steady-state levels of HOXA9, mirrored by impairment of the ability of 32D myeloid progenitor cells to undergo proper terminal differentiation into granulocytes. These results revealed a novel regulatory mechanism of hematopoiesis by ubiquitin-dependent proteolysis.

Evidence that the RdeA protein is a component of a multistep phosphorelay modulating rate of development in Dictyostelium

We have isolated an insertional mutant of Dictyostelium discoideum that aggregated rapidly and formed spores and stalk cells within 14 h of development instead of the normal 24 h. We have shown by parasexual genetics that the insertion is in the rdeA locus and have cloned the gene. It encodes a predicted 28 kDa protein (RdeA) that is enriched in charged residues and is very hydrophilic. Constructs with the DNA for the c-Myc epitope or for the green fluorescent protein indicate that RdeA is not compartmentalized. RdeA displays homology around a histidine residue at amino acid 65 with members of the H2 module family of phosphotransferases that participate in multistep phosphoryl relays. Replacement of this histidine rendered the protein inactive. The mutant is complemented by transformation with the Ypd1 gene of Saccharomyces cerevisiae, itself an H2 module protein. We propose that RdeA is part of a multistep phosphorelay system that modulates the rate of development.

Developmental decisions in Dictyostelium discoideum

A few hours after the onset of starvation, amoebae of Dictyostelium discoideum start to form multicellular aggregates by chemotaxis to centers that emit periodic cyclic AMP signals. There are two major developmental decisions: first, the aggregates either construct fruiting bodies directly, in a process known as culmination, or they migrate for a period as "slugs." Second, the amoebae differentiate into either prestalk or prespore cells. These are at first randomly distributed within aggregates and then sort out from each other to form polarized structures with the prestalk cells at the apex, before eventually maturing into the stalk cells and spores of fruiting bodies. Developmental gene expression seems to be driven primarily by cyclic AMP signaling between cells, and this review summarizes what is known of the cyclic AMP-based signaling mechanism and of the signal transduction pathways leading from cell surface cyclic AMP receptors to gene expression. Current understanding of the factors controlling the two major developmental choices is emphasized. The weak base ammonia appears to play a key role in preventing culmination by inhibiting activation of cyclic AMP-dependent protein kinase, whereas the prestalk cell-inducing factor DIF-1 is central to the choice of cell differentiation pathway. The mode of action of DIF-1 and of ammonia in the developmental choices is discussed.

Overexpression of Dd PK2 protein kinase causes rapid development and affects the intracellular cAMP pathway of Dictyostelium discoideum

The Dd PK2 gene codes for a putative protein of 648 amino acids with a C-terminal half sharing high homology with protein kinase A catalytic subunits from other organisms. In order to find out more about the physiological role of the Dd PK2 kinase, its gene, and a version having a frame shift mutation in the middle of the catalytic region, were overexpressed in developing Dictyostelium cells. Both the intact gene (K-) and the frame shift mutant (Kdel-) caused rapid development with spores formed in 16–18 hours compared to the 24 hours required by their parent. This result was confirmed by the pattern of expression of some developmentally regulated genes. Other rapid developing strains (rde) are activated in the cAMP second messenger system. Both K- and Kdel-containing strains have lower cAMP levels than the parental strain during late development, thus resembling rdeC mutants. K-cells (but not Kdel-cells) produced bizarre fruiting bodies with many prostrate forms. The parallel with rde mutants was confirmed by demonstrating that K-cells are able to form spores in submerged monolayer culture. Furthermore, K-cells have about four times more protein kinase A (cAPK) activity than wild-type cells. These results indicate that the N-terminal domain of Dd PK2 is sufficient to influence cAMP levels and to provoke rapid development, whereas kinase activity seems to be required for the sporogenous phenotype. The association between elevated cAPK and Dd PK2 overexpression phenotype further indicates a role for cAPK in the formation of spores.

A density-sensing factor controls development in Dictyostelium

For an unknown reason, several genes expressed during Dictyostelium development are regulated by cell density. This is mediated by an 80-kD glycoprotein, conditioned medium factor (CMF), which is slowly secreted and simultaneously sensed by starved cells. To examine further this eukaryotic cell density-sensing mechanism, we have isolated a cDNA encoding CMF. The derived amino acid sequence of CMF shows no obvious similarity to any known protein and thus may represent a new class of eukaryotic intercellular signal. CMF antisense transformants do not aggregate, whereas normal development is restored by the addition of purified CMF protein. This suggests that CMF might synchronize the onset of development in Dictyostelium by triggering aggregation when a majority of the cells in a given area have starved, as signaled by CMF secretion.

Analysis of G alpha 4, a G-protein subunit required for multicellular development in Dictyostelium

The Dictyostelium G alpha 4 gene encodes a G-protein alpha subunit that is primarily expressed during the multicellular stages of development. g alpha 4 null mutants, created by gene disruption, show aberrant morphological differentiation, reduced levels of prespore gene expression, and a loss of the ability to produce spores. These developmental phenotypes can be rescued by complementation with the wild-type gene. Cells that overexpress the G alpha 4 gene (G alpha 4HC) also show reduced spore production but display an aberrant morphological phenotype distinct from that of g alpha 4 cells. The g alpha 4 phenotype can be partially rescued by the presence of wild-type or G alpha 4HC cells in chimeric organisms, suggesting that G alpha 4-expressing cells produce an intercellular signal that is essential for multicellular development.

A spatial gradient of expression of a cAMP-regulated prespore cell-type-specific gene in Dictyostelium

Previously, we identified a class of genes in Dictyostelium that are prespore cell-type specific in their expression in the multicellular aggregate and are inducible by cAMP acting through cell-surface cAMP receptors. In this paper, we report the cloning and analysis of the regulatory regions controlling the expression of one such gene that encodes a spore coat protein, SP60. By use of a fusion of the firefly luciferase gene and Escherichia coli lacZ [expresses beta-galactosidase (beta-gal)], we have identified cis-acting regions required for proper spatial and temporal expression in multicellular aggregates and for cAMP induction in shaking cell culture. Deletion analysis suggests that a CA-rich element (CAE) and surrounding sequences present three times within the 5'-flanking sequence are required for proper regulation. SP60-lacZ fusions that include all three of these regions express lacZ only in the posterior approximately 85% of migrating slugs (prespore zone). Studies show that SP60 is expressed during mid to late aggregation, and SP60-lacZ-positive cells are spatially localized as a doughnut-shaped ring within the forming aggregate. Cells within the skirt that surrounds the aggregate and that are still migrating into the aggregate do not stain. Sequential 5' deletions of CAEs and surrounding regions affect the expression level of SP60-luciferase in response to developmental signals and cAMP, as well as the spatial pattern of SP60-lacZ. Deletion of the first (most 5') of these regions restricts the spatial expression of SP60-lacZ fusions to the anterior of the prespore zone. When both the first and second regions are removed, the expression level drops, and the staining is restricted to the prespore/prestalk boundary. Furthermore, the staining pattern that is seen with these two deletions is present as a gradient from anterior to posterior within the prespore zone. Deletion of all three regions results in a loss of both cAMP and developmentally induced expression. These results suggest the presence of a gradient within the prespore zone that differentially affects the activity of promoters containing different numbers of response elements.

Inositol trisphosphate and diacylglycerol can differentially modulate gene expression in Dictyostelium

We have previously shown that several genes expressed during Dictyostelium development could be induced in shaking culture by exogenous cAMP, even though the accumulation of intracellular cAMP was inhibited. The use of selected cAMP analogs indicated that the exogenous cAMP functioned by activating the cell surface cAMP receptor and not by interacting with the regulatory subunit of the intracellular cAMP-dependent protein kinase. Although some genes in Dictyostelium appear to be regulated by intracellular cAMP, these data suggest that this is not the case for all genes regulated by cAMP. Intracellular second messengers other than cAMP may, therefore, promote the expression of these other genes. Here, we have examined inositol trisphosphate and diacylglycerol as candidates for such mediators of signal transduction. We have studied three genes that exhibit disparate modes of temporal and spatial expression during development of Dictyostelium. In shaking cultures, maximal levels of expression of each are dependent on the accumulation of or exposure to extracellular cAMP. We show that the addition of inositol trisphosphate and/or diacylglycerol to cells in shaking culture has distinct effects on the expression of each gene and, under specific conditions, can bypass the requirement for extracellular cAMP. These data suggest that extracellular cAMP interacting with its cell surface receptor may promote synthesis of inositol trisphosphate and diacylglycerol to regulate gene expression and aspects of differentiation in Dictyostelium.

Two-phase regulatory pathway controls cAMP receptor-mediated expression of early genes in Dictyostelium

Two classes of early genes in Dictyostelium are differentially regulated by extracellular pulses of cAMP interacting with its cell-surface receptor, conditions that also regulate chemotaxis and aggregation. The pulse-repressed genes, such as K5, are induced shortly after the onset of starvation and are repressed a few hr later during aggregation by cAMP pulses. The pulse-induced genes (including D2, M3, and those encoding contact sites A, the G alpha protein subunit G alpha 2, and the cell-surface cAMP receptor) are maximally induced just prior to aggregation by pulses of cAMP and are subsequently repressed by sustained moderate levels of cAMP--conditions that exist sequentially in development. In this manuscript, we further analyze the requirement for cAMP pulses and characterize a requirement for protein synthesis for the expression of these two classes of genes. Our results indicate that the control of expression of both the pulse-induced and pulse-repressed genes requires other developmentally regulated factors in addition to starvation and cAMP pulses. We also identified another early gene, F9, whose expression is stimulated upon starvation, is not responsive to cAMP, and is hyperstimulated by cycloheximide, in a manner similar to the cycloheximide stimulation of c-fos and other serum-induced genes in mammalian cells. Examination of the kinetics of expression of the pulse-induced genes in a mutant blocked in the cAMP relay pathway indicates that their expression is controlled by a two-phase process. The first phase requires starvation and CMF, an extracellular conditioned medium factor, and results in a low level of expression. The second phase requires establishment of the cAMP signal-relay system and induces the genes to a high level. Both phases require prior and concomitant protein synthesis. Some of the members of the pulse-induced class encode elements of the cAMP signal-relay system that controls aggregation, indicating a feedback autoregulation. The two-phase process might allow the "finetuning" of the level of expression of genes involved in aggregation.

Characterization of genes which are transiently expressed during the preaggregative phase of development of Dictyostelium discoideum

We have identified and characterized three genes, the I genes (I for induced), which are induced during the preaggregative phase of the developmental program of Dictyostelium discoideum. None of these genes are expressed in cells growing vegetatively on bacteria or in axenic broth, and their induction during early development is due to transcriptional activation. Developmental expression of I6, I8, and I11 occurs even in the absence of protein synthesis. Their induction is very rapid and occurs essentially at the onset of development. The expression is transient, peaking between 2 and 4 hr followed by a rapid loss of expression. These characteristics suggest that the induction of I6, I8, and I11 is a primary result of the initiation of development, and thus they represent the first such genes isolated. Although their expression behavior shares these characteristics, examination of their expression under various conditions of development and in a variety of aggregation-deficient mutant strains reveals that the details of the regulation and developmental control of these three genes are distinct.

Genetics of early Dictyostelium discoideum development

Images

Structure and expression of the cAMP cell-surface receptor

Using antibodies specific for the 3',5'-cyclic AMP (cAMP) cell surface receptor of Dictyostelium discoideum, we have screened lambda gtll expression libraries and isolated a series of cDNAs derived from cAMP receptor mRNA during early development. The identity of the cDNA clones was verified by multiple criteria: 1) beta-galactosidase fusion proteins synthesized by isolated cDNA clones stain intensely with cAMP receptor directed antiserum, 2) these fusion proteins affinity purify antibodies specific for the cAMP receptor, 3) the cDNA probes hybridize to a 2 kb mRNA whose change in relative level of abundance during development parallels that of receptor mRNA as assayed by in vitro translation, 4) the 2 kb mRNA size equals that of receptor mRNA as determined by in vitro translation of size fractionated poly (A)+ RNA, and 5) RNA transcribed in vitro from cDNAs containing the entire protein-coding region produces a polypeptide by in vitro translation with an apparent molecular weight in close agreement with that of nascent cAMP receptor protein produced by in vitro translation of cellular RNA. The DNA sequence predicts an open reading frame of 392 amino acids. The deduced amino acid sequence contains seven domains enriched in hydrophobic residues. A model is proposed in which the cAMP cell-surface receptor traverses the lipid bilayer seven times in a pattern similar to that of other receptors, such as rhodopsin, which interact with G-proteins. The structural similarities suggest a gene family of related surface receptors from such evolutionarily diverse species as Dictyostelium, yeast, and mammals.

Different molecular mechanisms for cAMP regulation of gene expression during Dictyostelium development

Alterations in cAMP concentrations have been implicated in developmentally regulated gene expression in Dictyostelium. Using a variety of culture conditions to control the metabolism of cAMP during cytodifferentiation, I have examined the role of the cyclic nucleotide in development. Conditions which allow intracellular synthesis of cAMP promote the normal developmental repression of gene M4-1 by a mechanism which is completely independent of the formation of multicellular aggregates. If, however, cells are inhibited in their ability to activate adenylate cyclase and, thus, intracellular cAMP signaling, they prove unable to repress M4-1, even in the presence of exogenous cAMP. In contrast, expression of genes which exhibit maximal activity after aggregate formation depends upon accumulation of extracellular cAMP. Inhibition of intracellular cAMP signaling does not prevent the expression of these genes if cultures are simultaneously exposed to high levels of exogenously added extracellular cAMP. These results indicate that there are at least two independent mechanisms involved in the developmental regulation of gene expression by cAMP in Dictyostelium. I discuss plausible molecular mechanisms through which cAMP might alter gene expression.

Identification of the sequences controlling cyclic AMP regulation and cell-type-specific expression of a prestalk-specific gene in Dictyostelium discoideum

We have cloned and analyzed a developmentally and spatially regulated prestalk cell-specific gene from Dictyostelium discoideum. The gene encodes a protein highly homologous to the lysosomal cysteine proteinases cathepsin H and cathepsin B. Amino acid comparisons between these enzymes showed that the active-site amino acids were conserved, as were amino acids known to be important for catalysis and residues which form the intramolecular cysteine bridges. We have constructed a series of internal deletions, duplications, and linker scanner mutations within the region 300 base pairs 5' to the cap site. Analysis of expression of the mutations in transformants identified a approximately 35-base pair GC-rich region containing a dAdC/dGdT palindromic repeat and a G-rich box which is homologous to the 3' GT half of the palindromic repeat. Deletion or disruption of the G box resulted in a approximately 50-fold drop in the level of expression of the gene fusion in transformants in response to cyclic AMP in single-cell culture but did not affect the temporal pattern of regulation or control by cyclic AMP. The expression of such constructs during normal multicellular differentiation paralleled that of the endogenous gene; however, the level of RNA from the constructs was only approximately 10-fold lower than that of constructs containing the G box. Deletion of the 3' half of the palindromic sequence and the G box region resulted in a dramatic decrease in the level of transcription, although the constructs still showed proper temporal expression. These results suggest that this 35-base-pair region acts as an important part of the regulatory region for cell type and cyclic AMP regulation.

Identification of the sequences controlling cyclic AMP regulation and cell-type-specific expression of a prestalk-specific gene in Dictyostelium discoideum

We have cloned and analyzed a developmentally and spatially regulated prestalk cell-specific gene from Dictyostelium discoideum. The gene encodes a protein highly homologous to the lysosomal cysteine proteinases cathepsin H and cathepsin B. Amino acid comparisons between these enzymes showed that the active-site amino acids were conserved, as were amino acids known to be important for catalysis and residues which form the intramolecular cysteine bridges. We have constructed a series of internal deletions, duplications, and linker scanner mutations within the region 300 base pairs 5' to the cap site. Analysis of expression of the mutations in transformants identified a approximately 35-base pair GC-rich region containing a dAdC/dGdT palindromic repeat and a G-rich box which is homologous to the 3' GT half of the palindromic repeat. Deletion or disruption of the G box resulted in a approximately 50-fold drop in the level of expression of the gene fusion in transformants in response to cyclic AMP in single-cell culture but did not affect the temporal pattern of regulation or control by cyclic AMP. The expression of such constructs during normal multicellular differentiation paralleled that of the endogenous gene; however, the level of RNA from the constructs was only approximately 10-fold lower than that of constructs containing the G box. Deletion of the 3' half of the palindromic sequence and the G box region resulted in a dramatic decrease in the level of transcription, although the constructs still showed proper temporal expression. These results suggest that this 35-base-pair region acts as an important part of the regulatory region for cell type and cyclic AMP regulation.

Cellular and subcellular distribution of a cAMP-regulated prestalk protein and prespore protein in Dictyostelium discoideum: a study on the ontogeny of prestalk and prespore cells

We have analyzed a developmentally and spatially regulated prestalk-specific gene and a prespore-specific gene from Dictyostelium. The prestalk gene, pst-cathepsin, encodes a protein highly homologous to the lysosomal cysteine proteinases cathepsin H and cathepsin B. The prespore gene encodes a protein with some homology to the anti-bacterial toxin crambin and has been designated beejin. Using the lambda gtll system, we have made polyclonal antibodies directed against a portion of the protein encoded by pst-cathepsin and other antibodies directed against the beejin protein. Both antibodies stain single bands on Western blots. By immunofluorescence and Western blots, pst-cathepsin is not present in vegetative cells or developing cells during the first approximately 10 h of development. It then appears with a punctate distribution in a subset of developing cells. Beejin is detected only after approximately 15 h of development, also in a subset of cells. Pst-cathepsin is distributed in the anterior approximately 1/10 of migrating slugs and on the peripheral posterior surfaces of slugs. Beejin is distributed in the posterior region of slugs. Expression of both pst-cathepsin and beejin can be induced in subsets of isolated cultured cells by a combination of conditioned medium and extracellular cAMP in agreement with the regulation of the mRNAs encoding these proteins. We have used the antibodies as markers for cell type to examine the ontogeny and the spatial distribution of prestalk and prespore cells throughout multicellular development. Our findings suggest that prestalk cell differentiation is independent of position within the aggregate and that the spatial localization of prestalk cells within the multicellular aggregate arises from sorting of the prestalk cells after their induction. We have also found a class of cell in developing aggregates that contains neither the prestalk nor the prespore markers.

Regulation of late gene expression in a temperature-sensitive cohesion-defective mutant of Dictyostelium discoideum

We have analyzed the expression of a series of developmentally regulated genes in the Dictyostelium discoideum strain JC-5. This strain has been previously described as a temperature-sensitive, cohesion-defective derivative of FR17, itself a temporally deranged mutant of wild-type NC-4. At restrictive temperature (27 degrees C), JC-5 initially acquires EDTA-resistant cell contacts but at the time of tip formation (12 hr) loses the ability to make specific cell-cell associations and regresses to an amorphous mound of cells. WE have found that genes preferentially expressed in either prespore or prestalk cells are expressed prior to the appearance of the cohesion defect in JC-5; the specific cell contact system defective in this strain is necessary for neither the proper initiation nor maintenance of expression of either prespore of prestalk genes. We have also found, by use of an in vitro cell suspension system, that JC-5 is temperature-sensitive with respect to gene expression several hours before the defect in cell cohesion is observable. Our data suggest that the defect in JC-5 is due to a specific lesion not in the late cohesion system but rather in a more general component that is required earlier in the developmental process.