The Dictyostelium MAP kinase kinase DdMEK1 regulates chemotaxis and is essential for chemoattractant-mediated activation of guanylyl cyclase
1997; Springer Nature; Volume: 16; Issue: 14 Linguagem: Inglês
10.1093/emboj/16.14.4317
ISSN1460-2075
Autores Tópico(s)Retinal Development and Disorders
ResumoArticle15 July 1997free access The Dictyostelium MAP kinase kinase DdMEK1 regulates chemotaxis and is essential for chemoattractant-mediated activation of guanylyl cyclase Hui Ma Hui Ma Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Marianne Gamper Marianne Gamper Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Carole Parent Carole Parent Department of Biological Chemistry, Johns Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, MD, 21205 USA Search for more papers by this author Richard A. Firtel Corresponding Author Richard A. Firtel Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Hui Ma Hui Ma Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Marianne Gamper Marianne Gamper Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Carole Parent Carole Parent Department of Biological Chemistry, Johns Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, MD, 21205 USA Search for more papers by this author Richard A. Firtel Corresponding Author Richard A. Firtel Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA Search for more papers by this author Author Information Hui Ma1, Marianne Gamper1, Carole Parent2 and Richard A. Firtel 1 1Department of Biology, Center for Molecular Genetics, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA, 92093-0634 USA 2Department of Biological Chemistry, Johns Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, MD, 21205 USA *Corresponding author. E-mail: [email protected] The EMBO Journal (1997)16:4317-4332https://doi.org/10.1093/emboj/16.14.4317 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info We have identified a MAP kinase kinase (DdMEK1) that is required for proper aggregation in Dictyostelium. Null mutations produce extremely small aggregate sizes, resulting in the formation of slugs and terminal fruiting bodies that are significantly smaller than those of wild-type cells. Time-lapse video microscopy and in vitro assays indicate that the cells are able to produce cAMP waves that move through the aggregation domains. However, these cells are unable to undergo chemotaxis properly during aggregation in response to the chemoattractant cAMP or activate guanylyl cyclase, a known regulator of chemotaxis in Dictyostelium. The activation of guanylyl cyclase in response to osmotic stress is, however, normal. Expression of putative constitutively active forms of DdMEK1 in a ddmek1 null background is capable, at least partially, of complementing the small aggregate size defect and the ability to activate guanylyl cyclase. However, this does not result in constitutive activation of guanylyl cyclase, suggesting that DdMEK1 activity is necessary, but not sufficient, for cAMP activation of guanylyl cyclase. Analysis of a temperature-sensitive DdMEK1 mutant suggests that DdMEK1 activity is required throughout aggregation at the time of guanylyl cyclase activation, but is not essential for proper morphogenesis during the later multicellular stages. The activation of the MAP kinase ERK2, which is essential for chemoattractant activation of adenylyl cyclase, is not affected in ddmek1 null strains, indicating that DdMEK1 does not regulate ERK2 and suggesting that at least two independent MAP kinase cascades control aggregation in Dictyostelium. Introduction MAP kinase cascades are comprised of three evolutionarily conserved kinases, a MEK kinase (MAP kinase kinase kinase), a MEK (MAP kinase kinase) and a MAP kinase, that function sequentially (Davis, 1993; Errede and Levin, 1993; Herskowitz, 1995; Levin and Errede, 1995; Marshall, 1995; Bokemeyer et al., 1996). These cascades are activated by various cell surface receptors, including receptor tyrosine kinases, receptors coupled to cytoplasmic tyrosine kinases, G protein-coupled receptors and histidine kinases that function as part of two-component systems. They control a diversity of intracellular responses that include: pathways leading to cell growth and cell type differentiation, mating in yeast, activation of adenylyl cyclase in Dictyostelium and cell viability after stresses such as osmotic or heat shock, low oxygen tension and UV radiation (Herskowitz, 1995; Levin and Errede, 1995; Marshall, 1995; Segall et al., 1995; Bokemeyer et al., 1996). In the yeast Saccharomyces cerevisiae, five independent MAP kinase cascades have been identified (Levin and Errede, 1995). Some of these cascades have overlapping components, but at least one component of each cascade is unique to a specific response pathway. In Dictyostelium, formation of the multicellular organism results from the chemotactic aggregation of up to 105 cells. This process is controlled by a series of integrated signal transduction pathways that are activated by oscillatory pulses of extracellular cAMP interacting with G protein-coupled, serpentine receptors (cARs) (Devreotes, 1994; Firtel, 1995; Van Haastert, 1995; Chen et al., 1996). As an aggregation domain forms, cAMP waves are initiated by a central core of cells, or oscillator, and move outward with a spiral or concentric pattern and a periodicity of ∼6 min during the height of aggregation (Siegert and Weijer, 1991). Approximately 20 of these waves are required for the formation of a loose aggregate. cAMP binding to the receptors results in the activation of adenylyl and guanylyl cyclases. Activation of adenylyl cyclase results initially in a rise in intracellular cAMP, which is then released into the extracellular medium, relaying the signal from cell to cell. The transient rise in intracellular cGMP in response to the activation of guanylyl cyclase has been linked directly to chemotactic movement, and the pathways that control chemotaxis in Dictyostelium are similar to those used in mammalian neutrophils (Devreotes and Zigmond, 1988). Dictyostelium mutations that are unable to activate guanylyl cyclase do not undergo chemotaxis (Kuwayama et al., 1993, 1995; Liu et al., 1993). A mutation in a cytosolic cGMP-specific phosphodiesterase (Ross and Newell, 1981; Newell, 1995), which results in a more extended cAMP-stimulated rise in cGMP, produces a protracted period of chemotactic movement. Both the adenylyl and guanylyl cyclase pathways are followed by a period of adaptation. Clearing of the extracellular cAMP signal by phosphodiesterase allows these pathways to become resensitized and the cells to respond to the subsequent wave of cAMP (Hall et al., 1993). Activation of cell surface receptors also results in the induction of aggregation-stage gene expression (Firtel, 1995). Most of these genes require cAMP oscillations for expression and removal of the cAMP signal or the addition of a continuous signal, which results in an adaptation of the receptors and repression of gene expression (Mann and Firtel, 1987, 1989; Wu et al., 1995a). Other genes, such as the extracellular phosphodiesterase, are induced by either a pulsatile or continuous signal (Hall et al., 1993; Wu et al., 1995a). Genetic and biochemical analyses have resulted in the identification of many of the components that regulate various aspects of the aggregation-stage signaling pathways (Devreotes, 1994; Firtel, 1995; Van Haastert, 1995; Chen et al., 1996). The activation of both adenylyl and guanylyl cyclases requires the cAMP receptor-coupled G protein containing the Gα subunit Gα2, but the mechanisms by which the two pathways are activated are distinct. Data indicate that the Gα subunit is thought to activate guanylyl cyclase and phospholipase C, while the βγ subunits are directly involved in the activation of adenylyl cyclase (Okaichi et al., 1992; Bominaar et al., 1994; Valkema and Van Haastert, 1994; Wu et al., 1995b; Chen et al., 1996). In addition, other components, such as CRAC (cytosolic regulator of adenylyl cyclase), are essential for activation of adenylyl, but not guanylyl, cyclase (Insall et al., 1994; Lilly and Devreotes, 1994, 1995). The MAP kinase ERK2 is also essential for activation of adenylyl cyclase (Segall et al., 1995). erk2 null cells have normal levels of adenylyl cyclase but do not exhibit receptor-mediated activation. ERK2 is activated rapidly in response to cAMP, but a significant component of this activation does not require receptor-coupled G proteins (Knetsch et al., 1996; Maeda et al., 1996). In addition to being required for aggregation, ERK2 is also essential for proper morphogenesis and pre-spore, but not pre-stalk, cell differentiation during the multicellular stages of development (Gaskins et al., 1996). ERK2 activation and adaptation are negatively regulated by Ras and positively regulated by cAMP-dependent protein kinase (PKA), which is also required for other aspects of aggregation (Simon et al., 1989; Firtel and Chapman, 1990; Schulkes and Schaap, 1995; Mann et al., 1997). Analysis of these and other mutants has produced a picture of aggregation that is significantly more complex than might have been thought initially, containing multiple activation and feedback pathways that allow the organism to control the ability of the cells to undergo chemotaxis and produce an appropriate-sized multicellular organism (Chen et al., 1996). Two independently functioning MAP kinases have been identified in Dictyostelium, ERK1 (Gaskins et al., 1994) and ERK2. Overexpression studies of wild-type and mutant ERK1 expressed in a variety of genetic backgrounds indicate that ERK1 plays an important role in aggregation and later multicellular stages (Gaskins et al., 1994). Overexpression in wild-type cells produces multicellular aggregates with abnormal morphology which initiates at the slug stage. Overexpression of ERK1 in a protein tyrosine phosphatase 2 (PTP2) null background (ptp2 null cells) results in large aggregation streams that break up into multiple, tiny aggregates that developmentally arrest (Gaskins et al., 1994; Howard et al., 1994). The activity of ERK1 is developmentally regulated, as indicated by measuring the activity of the immunoprecipitated ERK1 isolated from cells at various times during Dictyostelium development (Gaskins et al., 1994). The inability to knock out the expression of ERK1 by either homologous recombination or antisense constructs has led to the suggestion that ERK1 function may be essential for growth. Here we describe the identification and function of a Dictyostelium MAP kinase kinase designated DdMEK1 that is essential for proper aggregation. ddmek1 null cells created by homologous recombination form very small aggregates. These cells show the normal wave patterning of cAMP signaling during the initial stages of aggregation, but the cells do not migrate towards the aggregation centers and are unable to migrate towards cAMP in standard chemotaxis assays. Instead, the aggregation domains break down and small groups of cells coalesce to form aggregates that are significantly smaller than those of wild-type cells. These aggregates continue to develop and show normal morphological differentiation, with the exception that the fruiting bodies are very small. These cells exhibit cAMP-stimulated adenylyl cyclase activity that is not significantly different from that in wild-type cells, but show a very severe defect in the cAMP-mediated activation of guanylyl cyclase. However, activation of guanylyl cyclase in response to osmotic stress is normal, suggesting that the intrinsic guanylyl cyclase activity is unaffected. Complementation of the ddmek1 null strain with a putative Ddmek1 temperature-sensitive mutation results in normal-sized aggregates at the permissive temperature and very small aggregates at the non-permissive temperature. Temperature shift experiments indicate that DdMEK1 activity is required at the time of cAMP stimulation for guanylyl cyclase activation and continuously throughout aggregation. A mutation in which the conserved serine/threonine residues that are phosphorylated in response to MEKK stimulation are changed to alanine residues is unable to complement the null cells and functions as a dominant-negative mutation in wild-type cells, producing a phenotype like that of ddmek1 null cells. ddmek1 null or wild-type cells expressing DdMEK1 containing either glutamate or aspartate at the conserved positions of serine/threonine phosphorylation form almost normal sized aggregates, but many of these aggregates arrest at the mound stage. Furthermore, we show that the ddmek1 null mutant does not affect the activation of the MAP kinase ERK2, which is also essential for aggregation and regulates the activation of adenylyl cyclase, suggesting that at least two independent MAP kinase cascades control aggregation in Dictyostelium. Our results identify a new and novel MAP kinase cascade that is essential for proper chemotaxis in Dictyostelium and appears to function by regulating receptor-mediated activation of guanylyl cyclase that is essential for aggregation. Results Cloning of DdMEK1 PCR was used to amplify expected conserved domains and components of MAP kinase pathways using Dictyostelium genomic DNA and cDNA libraries (see Materials and methods). One of the PCR products, evaluated after sequencing, was used to screen a λZap cDNA library. cDNA clones were isolated and sequenced. The derived amino acid sequence of these clones suggested that the gene encoded a MEK (MAP kinase kinase) and was designated DdMEK1. The full open reading frame (ORF) and a comparison of the putative kinase domain with that of known MEKs are shown in Figure 1A and B, respectively. The amino acid sequence shows strong homology to known MEKs present in the GenBank database and includes the conserved serine/threonine residues that are phosphorylated by the upstream MEK kinase during activation in other MEKs. The highest level of homology is to the murine MEK2 (Brott et al., 1993). Northern blot analysis indicates that DdMEK1 is expressed at a low level in vegetative cells, with mRNA levels increasing during early development, peaking during late aggregation, and then decreasing during the later multicellular stages (Figure 2). Figure 1.Sequence of DdMEK1. (A) The amino acid sequence of DdMEK1 is shown. The boxed region depicts the kinase core. Underlined regions show homopolymer and homopolymer-rich runs. The boxed amino acids with numbering are positions of point mutations described in the text. (B) Amino acid sequence comparison of the kinase core domain of DdMEK1 with other MAP kinase kinases from a variety of eukaryotes. Hu-Mpk1, human MEK Mpk1, accession no. Q02750; Mu-MEK2, murine MEK MEK2, accession no. S68267; Dm-SOR1, Drosophila MEK SOR1, accession no. A45176; Sp-BYR1, Schizosaccharomyces pombe MEK BYR1, accession no. P10506; Sc-PBS2, Saccharomyces cerevisiae MEK PBS2, accession no. P08018; Sc-MKK1, S.cerevisiae MEK MKK1, accession no. P32490; S.cerevisiae MEK Ste7, accession no. P06784. Asterisks indicate conserved residues in the ATP-binding site and sites of MEK activating phosphorylation by MEKKs. Download figure Download PowerPoint Figure 2.Developmental kinetics of expression of DdMEK1 in wild-type and ddmek1 null cells. RNA was isolated from developing cells, at the times indicated, from wild-type and ddmek1 null cells, size fractionated on a denaturing gel, blotted and hybridized with the DdMEK1 probe. ‘Veg’ is RNA from log-phase vegetatively growing cells. Download figure Download PowerPoint Phenotype of ddmek1 null cells DdMEK1 was disrupted by homologous recombination by insertion of a blasticidin resistance cassette between codons 377 and 378 of the ORF. The disruption was confirmed by Southern blot analysis of randomly selected clones (data not shown; see Materials and methods). Northern blot analysis of a representative clone shows the loss of the full-length DdMEK1 transcript and the presence of new, shorter transcripts (Figure 2). ddmek1 null cells grow normally either in axenic medium or in association with bacteria. When these cells are plated on Na/K-buffered agar to allow multicellular development to ensue, the cells do not show normal aggregation streams but produce aggregates, slugs and fruiting bodies that are very small compared with those of wild-type organisms. A good indication of the size differential is seen in Figure 3A, which shows photographs of wild-type and ddmek1 null mounds and slugs at the same magnifications. As seen in the top panels, the mound sizes of the ddmek1 null cells are significantly smaller than those of wild-type cells. As depicted in the wild-type control (lower panel), the one large slug is representative of the size of wild-type slugs when cells are plated at this density. In addition, there are a few smaller slugs composed of cells that did not aggregate into and become part of a normal, larger aggregate. As is evident, the sizes of all the ddmek1 null strain aggregates and slugs are significantly smaller than the wild-type slugs when the strains are plated on the agar at the same density. Except for the size, the overall morphologies of the ddmek1 null slugs, fruiting bodies and other multicellular stages are indistinguishable from those of wild-type cells (data not shown). [Note that in wild-type strains, normal morphology and spatial patterning is observed in aggregates that range in size over more than three orders of magnitude, with a maximum size of ∼105 cells (Schaap, 1986).] This small-aggregate ddmek1 null phenotype is fully complemented by expression of the full-length DdMEK1 cDNA from the constitutively expressed Actin 15 (Act15) promoter (Knecht et al., 1986; data not shown). Overexpression of DdMEK1 from the Act15 promoter in wild-type cells does not produce an observable phenotype. Some mutants affecting aggregation can be rescued or partially rescued by first being pulsed with cAMP (Insall et al., 1996), which maximizes the expression of the aggregation-stage cAMP receptor and other components of the signaling pathway. When both wild-type and ddmek1 null cells are pulsed for 4.5 h and then plated for development, the aggregate size of the ddmek1 null and wild-type strains are unchanged compared with unpulsed cells (Figure 3B), indicating that prior pulsing of ddmek1 cells cannot rescue the null phenotype. Figure 3.Developmental morphology of wild-type and ddmek1 null cells. (A) The figure shows the developmental morphology of wild-type and ddmek1 null cells at 9 h (upper panels) and 17 h of development (the slug stage; bottom panels). The magnification of the two images in the top row and the two images in the bottom row is the same. (B) Log-phase wild-type and ddmek1 null vegetative cells were washed and resuspended in 12 mM Na/KPO4 (pH 6.2) at 5×106 cells/ml and pulsed with 30 nM cAMP at 6 min intervals for 4.5 h. Cells were then washed and plated for development on Na/KPO4-containing agar at the same density. Photographs were taken 7 h after plating on agar at the time of initial slug formation. (Note: aggregation is significantly more rapid in cells that are pulsed with cAMP.) Both photographs are at the same magnification. Download figure Download PowerPoint Northern blot analysis was performed to determine if the ddmek1 null mutation affected the temporal or quantitative level of expression of cell type-specific genes. This analysis included the pre-stalk-specific gene ecmA (pre-stalk) (Williams et al., 1987), the pre-spore-specific gene SP60 (Mehdy et al., 1983; Haberstroh and Firtel, 1990) and the sporulation-specific gene SpiA (Richardson et al., 1994). No differences were observed (data not shown). Analysis of cell type proportioning and spatial localization of pre-stalk and pre-spore cells within the multicellular organisms using cell type-specific promoters expressing the lacZ reporter also showed no observable differences between ddmek1 null cells and wild-type cells (data not shown). Point mutations of DdMEK1 suggest that DdMEK1 encodes a MAP kinase kinase MAP kinase kinases, or MEKs, are activated by phosphorylation at conserved serine/threonine residues by upstream MEK kinases (Mansour et al., 1994; Pages et al., 1994; Zheng and Guan, 1994). Mutations of these conserved serine/threonine residues to glutamates or aspartates can result in constitutively active forms of the enzyme, whereas mutations of these residues to alanines result in an enzyme that cannot be activated by phosphorylation and can yield mutations that are dominant-negative when overexpressed in an otherwise wild-type background (Mansour et al., 1994; Pages et al., 1994; Zheng and Guan, 1994; Huang et al., 1995; Yashar et al., 1995; Errede and Ge, 1996). Figures 1A and 4 show the amino acid sequence around these conserved residues and the various mutations that were made by site-directed mutagenesis. The different mutant DdMEK1 expression constructs were transformed into both wild-type and ddmek1 null cells. Clonal isolates of stable transformants were obtained. As shown in Figure 5 and summarized in Table I, overexpression of DdMEK1 with the double glutamate or aspartate substitutions (DdMEK1S444E,T448E; DdMEK1S444D,T448D) in ddmek1 null cells complemented the ddmek1 null phenotype with respect to its aggregate size defect, although, on average, the aggregates were slightly smaller than those of ddmek1 null cells complemented with the wild-type DdMEK1. ddmek1 null cells expressing DdMEK1S444D,T448D show defects in cAMP-mediated signaling pathways essential for aggregation (see below). While ddmek1 null cells expressing DdMEK1S444E,T448E form almost normal sized aggregates, many of these aggregates arrested at the mound stage (Figure 5). Approximately 30–50% of the cells continued through development and formed mature fruiting bodies. A similar phenotype was observed when the DdMEK1S444E,T448E mutant protein was expressed in wild-type cells, indicating that the effect is dominant. The same phenotypes were observed for the overexpression of DdMEK1 with the aspartate (DdMEK1S444D,T448D) mutations (data not shown). Expression of the double alanine mutant (DdMEK1S444A,T448A) or a mutant in which the lysine residue required for ATP binding was converted to an alanine (DdMEK1K321A) did not complement the ddmek1 null cells and resulted in a dominant-negative phenotype when expressed in wild-type cells producing a ddmek1 null phenotype (very small aggregate size; Figure 5; data for DdMEK1K321A not shown). The phenotypes of these mutants are consistent with Dictyostelium DdMEK1 encoding a MAP kinase kinase and these residues being essential for proper functioning of the protein. Table I summarizes the observed phenotypes of the different mutations. Figure 4.Amino acid sequence substitutions in DdMEK1. (A) The sequence surrounding the putative activation/phosphorylation sites of DdMEK1, Ste7 and human MEK1. Point mutations made in DdMEK1 are indicated below. (B) The sequence surrounding sites of proline→serine mutation that results in temperature-sensitive mutants in yeast Cdc2, Drosophila DmMEK1 and Dictyostelium DdERK2. DdMEK1 is shown with the point mutation described in the text. (C) The sequence around another conserved domain in yeast Ste7 and Dictyostelium DdMEK1. Sites of proline→serine or threonine changes are indicated. Download figure Download PowerPoint Figure 5.Morphologies of DdMEK1 mutations expressed in wild-type and ddmek1 null cells. Morphologies of ddmek1 null and wild-type cells expressing wild-type and mutant DdMEK1 proteins. The strains and time in development are indicated. All images are at the same magnification, except ddmek1 null and wild-type cells expressing DdMEK1S44E,T448E at 24 h, which are at a lower magnification to show a larger field of view. Download figure Download PowerPoint Table 1. Summary of expression of mutant DdMEK1 on the small aggregate size phenotype DdMEK1 mutant Wild-type ddmek1 null S444, T448→E, E almost wild-type aggregate almost wild-type aggregate S444, T448→D, D partial mound arrest partial mound arrest S444, T448→A, A null phenotype null phenotype (dominant-negative) K321→A null phenotype null phenotype (dominant-negative) P458→S temperature sensitive temperature sensitive (weak) (weak) P458→T temperature sensitive temperature sensitive (strong) (strong) (dominant-negative) P417→S non-ts, wild-type wild-type DdMEK1ΔN null phenotype null phenotype (dominant-negative) The region N-terminal to the catalytic core has long homopolymers that are not uncommon in Dictyostelium proteins and are present, for example, in the catalytic subunit of cAMP-dependent protein kinase and the adenylyl cyclase ACA (Mann and Firtel, 1991; Pitt et al., 1992). In addition, there is a domain near the N-terminus with weak homology to the MAP kinase-interacting domains that have been identified (Bardwell and Thorner, 1996; Bardwell et al., 1996). It is possible that this region is involved in designating DdMEK1 specificity. A DdMEK1 construct carrying a deletion of the region upstream from the catalytic core cannot complement the ddmek1 phenotype and inhibits proper aggregation and later morphogenesis in ddmek1 null and wild-type strains (Figure 5, Table I). ddmek1 null cells are defective in chemotaxis to cAMP As indicated in the Introduction, aggregation is controlled by a series of integrated signal transduction pathways that are activated by extracellular cAMP interacting with G protein-coupled receptors. To obtain better insight into the inability of ddmek1 null cells to undergo normal aggregation, the formation of multicellular aggregates in these cells was followed by time-lapse video microscopy using a phase contrast microscope. The videos were then compared with a similar analysis of wild-type cells. Figure 6A shows images from the time-lapse video recording of wild-type cells at different stages of the aggregation process taken at two different magnifications. Small, dark areas within the field of brighter rings of cells represent the initial formation of aggregation centers (see arrowheads). As the cAMP wave moves outward, responding cells elongate, producing lighter bands (Newell, 1995). As a cell adapts, the cell's shape becomes random and this area darkens. Moving waves can thus be seen as alternating light and dark bands moving through the field and as ’undulations' in the field of cells that can only be visualized by directly viewing the time-lapse video. At 3 h into development, the field appears uniform, although wave patterns can be seen in the video recordings. By 3.5–4 h, aggregation domains can be seen in the still images, which are very clear by 4 h 20 min (see arrowheads). In the 4 h 20 min and 4 h 40 min images, alternating light and dark domains can be identified. The time-lapse shows an oscillatory wave emitting through these various aggregation domains. Chemotaxis then initiates rapidly and, within 1 h, the cells are almost fully aggregated. In the mature mounds seen at 6 h, larger concentric rings due to cell movement and oscillations are visible. The oscillations can be observed in the video. By 7–8 h, the mounds start to spin rapidly as the tight aggregate forms and a tip starts to initiate. Figure 6.Time-lapse video microscopy of wild-type and ddmek1 null cells. The morphology of Dictyostelium cells was followed by time-lapse video phase microscopy taken with a 4× objective and recorded using an SVHS recorder. Zero time is the time of plating of the cells on agar. The time compression factor was 120. Shown are computer-grabbed images from the videotape at the approximate times in hours and minutes indicated. Images shown were captured from the video tape at ∼20 min intervals of real time, except as indicated. (A) Wild-type cells. The solid arrowheads mark the center of the aggregation center, which is always dark, as these cells are always adapted. The open arrowheads mark the outer limits of the aggregation domain. (B) ddmek1 null cells. In the images 3:30–5:10, the innermost open arrowheads mark the limits of the cells that show the strongest refractory changes. The arrowhead further out indicates the outer limits of the aggregation domain as determined by wave movement in the time-lapse video. In images 6:10 and 6:50, the open arrowheads mark the center of the aggregation domain that is splitting into multiple small aggregates. Download figure Download PowerPoint Video imaging of ddmek1 null cells shows a similar pattern to that of wild-type cells, with waves moving through the lawn of cells at the start of aggregation, although the bright central region of the aggregation domains is slightly smaller (Figure 6B, see arrowheads). However, videos of the wave patterns show that the waves extend outward to a size similar to that seen in wild-type cells (see arrowheads in Figure 6B, images at 3 h 50 min through 5 h 10 min). Moreover, the cells in the outer regions of these domains do not ap
Referência(s)