Tissue self-organization based on collective cell migration by contact activation of locomotion and chemotaxis

Despite their central role in multicellular organization, navigation rules that dictate cell rearrangement remain largely undefined. Contact between neighboring cells and diffusive attractant molecules are two of the major determinants of tissue-level patterning; however, in most cases, molecular and developmental complexity hinders one from decoding the exact governing rules of individual cell movement. A primordial example of tissue patterning by cell rearrangement is found in the social amoeba Dictyostelium discoideum where the organizing center or the "tip" self-organizes as a result of sorting of differentiating prestalk and prespore cells. By employing microfluidics and microsphere-based manipulation of navigational cues at the single-cell level, here we uncovered a previously overlooked mode of Dictyostelium cell migration that is strictly directed by cell-cell contact. The cell-cell contact signal is mediated by E-set Ig-like domain-containing heterophilic adhesion molecules TgrB1/TgrC1 that act in trans to induce plasma membrane recruitment of the SCAR complex and formation of dendritic actin networks, and the resulting cell protrusion competes with those induced by chemoattractant cAMP. Furthermore, we demonstrate that both prestalk and prespore cells can protrude toward the contact signal as well as to chemotax toward cAMP; however, when given both signals, prestalk cells orient toward the chemoattractant, whereas prespore cells choose the contact signal. These data suggest a model of cell sorting by competing juxtacrine and diffusive cues, each with potential to drive its own mode of collective cell migration.

Prespore cell inducing factor, psi factor, controls both prestalk and prespore gene expression in Dictyostelium development

Prespore cell-inducing (psi, psi) factor (PsiA), encoded by the psiA gene of Dictyostelium, is a secreted signal glycoprotein that induces prespore cell differentiation when added to monolayer cultures. In situ hybridization during normal development showed that the psiA gene is highly expressed in scattered cells at the mound stage and in prespore cells at the onset of culmination. The conventional prespore-cell marker genes, cotC and pspA, were expressed normally in psiA(-) and psiA overexpressing strains. Expressions of rnrB and cudA are repressed in the prestalk cells of a wild type slug to render prespore specific pattern. However, a promoter-reporter fusion gene, rnrB:lacZ, showed an ectopic expression in the prestalk cells of the psiA(-) strain while cudA(psp):lacZ did so in those of the psiA overexpressing strain. Overexpression of psiA delayed expression of the prestalk specific gene, ecmB, during development, while knocking out psiA promoted its expression. In addition, overexpression inhibited DIF-1-induced stalk formation in monolayer cultures. Together with the known prespore inducing activity, the results indicate that PsiA regulates both prespore and prestalk/stalk cell differentiation. These results indicate that PsiA is also involved in prestalk cell differentiation.

Loss of SMEK, a novel, conserved protein, suppresses MEK1 null cell polarity, chemotaxis, and gene expression defects

MEK/extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase signaling is imperative for proper chemotaxis. Dictyostelium mek1(-) (MEK1 null) and erk1(-) cells exhibit severe defects in cell polarization and directional movement, but the molecules responsible for the mek1(-) and erk1(-) chemotaxis defects are unknown. Here, we describe a novel, evolutionarily conserved gene and protein (smkA and SMEK, respectively), whose loss partially suppresses the mek1(-) chemotaxis phenotypes. SMEK also has MEK1-independent functions: SMEK, but not MEK1, is required for proper cytokinesis during vegetative growth, timely exit from the mound stage during development, and myosin II assembly. SMEK localizes to the cell cortex through an EVH1 domain at its N terminus during vegetative growth. At the onset of development, SMEK translocates to the nucleus via a nuclear localization signal (NLS) at its C terminus. The importance of SMEK's nuclear localization is demonstrated by our findings that a mutant lacking the EVH1 domain complements SMEK deficiency, whereas a mutant lacking the NLS does not. Microarray analysis reveals that some genes are precociously expressed in mek1(-) and erk1(-) cells. The misexpression of some of these genes is suppressed in the smkA deletion. These data suggest that loss of MEK1/ERK1 signaling compromises gene expression and chemotaxis in a SMEK-dependent manner.

LrrA, a novel leucine-rich repeat protein involved in cytoskeleton remodeling, is required for multicellular morphogenesis in Dictyostelium discoideum

Cell sorting by differential cell adhesion and movement is a fundamental process in multicellular morphogenesis. We have identified a Dictyostelium discoideum gene encoding a novel protein, LrrA, which composes almost entirely leucine-rich repeats (LRRs) including a putative leucine zipper motif. Transcription of lrrA appeared to be developmentally regulated with robust expression during vegetative growth and early development. lrrA null cells generated by homologous recombination aggregated to form loose mounds, but subsequent morphogenesis was blocked without formation of the apical tip. The cells adhered poorly to a substratum and did not form tight cell-cell agglomerates in suspension; in addition, they were unable to polarize and exhibit chemotactic movement in the submerged aggregation and Dunn chamber chemotaxis assays. Fluorescence-conjugated phalloidin staining revealed that both vegetative and aggregation competent lrrA(-) cells contained numerous F-actin-enriched microspikes around the periphery of cells. Quantitative analysis of the fluorescence-stained F-actin showed that lrrA(-) cells exhibited a dramatically increase in F-actin as compared to the wild-type cells. When developed together with wild-type cells, lrrA(-) cells were unable to move to the apical tip and sorted preferentially to the rear and lower cup regions. These results indicate that LrrA involves in cytoskeleton remodeling, which is needed for normal chemotactic aggregation and efficient cell sorting during multicellular morphogenesis, particularly in the formation of apical tip.

Talin B is required for force transmission in morphogenesis of Dictyostelium

Talin plays a key role in the assembly and stabilisation of focal adhesions, but whether it is directly involved in force transmission during morphogenesis remains to be elucidated. We show that the traction force of Dictyostelium cells mutant for one of its two talin genes talB is considerably smaller than that of wild-type cells, both in isolation and within tissues undergoing morphogenetic movement. The motility of mutant cells in tightly packed tissues in vivo or under strong resistance conditions in vitro was lower than that of wild-type cells, but their motility under low external force conditions was not impaired, indicating inefficient transmission of force in mutant cells. Antibody staining revealed that the talB gene product (talin B) exists as small units subjacent to the cell membrane at adhesion sites without forming large focal adhesion-like assemblies. The total amount of talin B on the cell membrane was larger in prestalk cells, which exert larger force than prespore cells during morphogenesis. We conclude that talin B is involved in force transmission between the cytoskeleton and cell exterior.

Three-dimensional in vivo analysis of Dictyostelium mounds reveals directional sorting of prestalk cells and defines a role for the myosin II regulatory light chain in prestalk cell sorting and tip protrusion

During cell sorting in Dictyostelium, we observed that GFP-tagged prestalk cells (ecmAO-expressing cells) moved independently and directionally to form a cluster. This is consistent with a chemotaxis model for cell sorting (and not differential adhesion) in which a long-range signal attracts many of the prestalk cells to the site of cluster formation. Surprisingly, the ecmAO prestalk cluster that we observed was initially found at a random location within the mound of this Ax3 strain, defining an intermediate sorting stage not widely reported in Dictyostelium. The cluster then moved en masse to the top of the mound to produce the classic, apical pattern of ecmAO prestalk cells. Migration of the cluster was also directional, suggesting the presence of another long-range guidance cue. Once at the mound apex, the cluster continued moving upward leading to protrusion of the mound's tip. To investigate the role of the cluster in tip protrusion, we examined ecmAO prestalk-cell sorting in a myosin II regulatory light chain (RLC) null in which tips fail to form. In RLC-null mounds, ecmAO prestalk cells formed an initial cluster that began to move to the mound apex, but then arrested as a vertical column that extended from the mound's apex to its base. Mixing experiments with wild-type cells demonstrated that the RLC-null ecmAO prestalk-cell defect is cell autonomous. These observations define a specific mechanism for myosin's function in tip formation, namely a mechanical role in the upward movement of the ecmAO prestalk cluster. The wild-type data demonstrate that cell sorting can occur in two steps, suggesting that, in this Ax3 strain, spatially and temporally distinct cues may guide prestalk cells first to an initial cluster and then later to the tip.

A novel Akt/PKB-related kinase is essential for morphogenesis in Dictyostelium

BACKGROUND

Dictyostelium Akt/PKB is homologous to mammalian Akt/PKB and is required for cell polarity and proper chemotaxis during early development. The kinase activity of Akt/PKB kinase is activated in response to chemoattractants in neutrophils and in Dictyostelium by the chemoattractant cAMP functioning via a pathway involving a heterotrimeric G protein and PI3-kinase. Dictyostelium contains several kinases structurally related to Akt/PKB, one of which, PKBR-1, is investigated here for its role in cell polarity, movement and cellular morphogenesis during development.

RESULTS

PKBR-1 has a kinase and a carboxy-terminal domain related to those of Akt/PKB, but no PH domain. Instead, it has an amino-terminal myristoylation site, which is required for its constitutive membrane localization. Like Akt/PKB, PKBR-1 is activated by cAMP through a G-protein-dependent pathway, but does not require PI3-kinase, probably because of the constitutive membrane localization of PKBR-1. This is supported by experiments demonstrating the requirement for membrane association for activation and in vivo function of PKBR-1. PKBR-1 protein is found in all cells throughout early development but is then restricted to the apical cells in developing aggregates, which are thought to control morphogenesis. PKBR-1 null cells arrest development at the mound stage and are defective in morphogenesis and multicellular development. These phenotypes are complemented by Akt/PKB, suggesting functional overlap between PKBR-1 and Akt/PKB. Akt/PKB PKBR-1 double knockout cells exhibit growth defects and show stronger chemotaxis and cell-polarity defects than Akt/PKB null cells.

CONCLUSIONS

Our results expand the previously known functions of Akt/PKB family members in cell movement and morphogenesis during Dictyostelium multicellular development. The results suggest that Akt/PKB and PKBR-1 have overlapping effectors and biological function: Akt/PKB functions predominantly during aggregation to control cell polarity and chemotaxis, whereas PKBR-1 is required for morphogenesis during multicellular development.

The Dictyostelium LIM domain-containing protein LIM2 is essential for proper chemotaxis and morphogenesis

We have identified limB, a gene encoding a novel LIM domain-containing protein, LIM2, in a screen for genes required for morphogenesis. limB null cells aggregate, although poorly, but they are unable to undergo morphogenesis, and the aggregates arrest at the mound stage. limB null cells exhibit an aberrant actin cytoskeleton and have numerous F-actin-enriched microspikes. The cells exhibit poor adhesion to a substratum and do not form tight cell-cell agglomerates in suspension. Furthermore, limB null cells are unable to properly polarize in chemoattractant gradients and move very poorly. Expression of limB from a prestalk-specific but not a prespore-specific promoter complements the morphogenetic defects of the limB null strain, suggesting that the limB null cell developmental defect results from an inability to properly sort prestalk cells. LIM2 protein is enriched in the cortex of wild-type cells, although it does not colocalize with the actin cytoskeleton. Our analysis indicates that LIM2 is a new regulatory protein that functions to control rearrangements of the actin cytoskeleton and is required for cell motility and chemotaxis. Our findings may be generally applicable to understanding pathways that control cell movement and morphogenesis in all multicellular organisms. Structure function studies on the LIM domains are presented.

In vivo observations of myosin II dynamics support a role in rear retraction

To investigate myosin II function in cell movement within a cell mass, we imaged green fluorescent protein-myosin heavy chain (GFP-MHC) cells moving within the tight mound of Dictyostelium discoideum. In the posterior cortex of cells undergoing rotational motion around the center of the mound, GFP-MHC cyclically formed a "C," which converted to a spot as the cell retracted its rear. Consistent with an important role for myosin in rotation, cells failed to rotate when they lacked the myosin II heavy chain (MHC-) or when they contained predominantly monomeric myosin II (3xAsp). In cells lacking the myosin II regulatory light chain (RLC-), rotation was impaired and eventually ceased. These rotational defects reflect a mechanical problem in the 3xAsp and RLC- cells, because these mutants exhibited proper rotational guidance cues. MHC- cells exhibited disorganized and erratic rotational guidance cues, suggesting a requirement for the MHC in organizing these signals. However, the MHC- cells also exhibited mechanical defects in rotation, because they still moved aberrantly when seeded into wild-type mounds with proper rotational guidance cues. The mechanical defects in rotation may be mediated by the C-to-spot, because RLC- cells exhibited a defective C-to-spot, including a slower C-to-spot transition, consistent with this mutant's slower rotational velocity.

Possible cooperation of differential adhesion and chemotaxis in mound formation of Dictyostelium

In the mound stage of Dictyostelium discoideum, pre-stalk cells sort and form a tip at the apex. How this pattern forms is as yet unknown. A cellular level model allows us to simulate both differential cell adhesion and chemotaxis, to show that with differential adhesion only, pre-stalk cells move to the surface of the mound but form no tip. With chemotaxis driven by an outgoing circular wave only, a tip forms but contains both pre-stalk and pre-spore cells. Only for a narrow range of relative strengths between differential adhesion and chemotaxis can both mechanisms work in concert to form a tip containing only pre-stalk cells. The simulations provide a method to determine the processes necessary for patterning and suggest a series of further experiments.

Cell-type-specific rescue of myosin function during Dictyostelium development defines two distinct cell movements required for culmination

Mutant Dictyostelium cells lacking any of the component polypeptides of myosin II exhibit developmental defects. To define myosin's role in establishing Dictyostelium's developmental pattern, we have rescued myosin function in a myosin regulatory light chain null mutant (mlcR-) using cell-type-specific promoters. While mlcR- cells fail to progress beyond the mound stage, expression of RLC from the prestalk promoter, ecmA, produces culminants with normal stalks but with defects in spore cell localization. When GFP-marked prestalk and prespore cells expressing ecmA-RLC are mixed with wild-type cells, the mislocalization of prestalk cells, but not prespore cells, is rescued. Time-lapse video recording of ecmA-RLC cells showed that the posterior prespore zone failed to undergo a contraction important for the upward movement of prespore cells. Prespore cells marked with green fluorescent protein (GFP) failed to move toward the tip with the spiral motion typical of wild type. In contrast, expression of RLC in prespore cells using the psA promoter produced balloon-like structures reminiscent of sorocarps but lacking stalks. GFP-labeled prespore cells showed a spiral movement toward the top of the structures. Expression of RLC from the psA promoter restores the normal localization of psA-GFP cells, but not ecmA-GFP cells. These results define two distinct, myosin-dependent movements that are required for establishing a Dictyostelium fruiting body: stalk extension and active movement of the prespore zone that ensures proper placement of the spores atop the stalk. The approach used in these studies provides a direct means of testing the role of cell motility in distinct cell types during a morphogenetic program.

Propagating waves control Dictyostelium discoideum morphogenesis

The morphogenesis of Dictyostelium results from the coordinated movement of starving cells to form a multicellular aggregate (mound) which transforms into a motile slug and finally a fruiting body. Cells differentiate in the mound and sort out to form an organised pattern in the slug and fruiting body. During aggregation, cell movement is controlled by propagating waves of the chemo-attractant cAMP. We show that mounds are also organised by propagating waves. Their geometry changes from target or single armed spirals during aggregation to multi-armed spiral waves in the mound. Some mounds develop transiently into rings in which multiple propagating wave fronts can still be seen. We model cell sorting in the mound stage assuming cell type specific differences in cell movement speed and excitability. This sorting feeds back on the wave geometry to generate twisted scroll waves in the slug. Slime mould morphogenesis can be understood in terms of wave propagation directing chemotactic cell movement.

Coexpression of a constitutively active plasma membrane calcium pump with GFP identifies roles for intracellular calcium in controlling cell sorting during morphogenesis in Dictyostelium

To examine the potential role of calcium in regulating Dictyostelium development, we reduced free cytosolic and total cell Ca2+ in Dictyostelium cells by expressing a constitutively active form of a human erythrocyte plasma membrane calcium pump. The pump-expressing cells lacked a thapsigargin-mediated increase in cytoplasmic calcium, consistent with a reduced level of total cellular Ca2+. During aggregation, the cells initially formed a large number of aggregation centers, many of which coalesced to form mounds that were smaller than those of wild-type cells, and the cells did not exhibit the normal formation of elongated aggregation streams. The majority of the mounds either arrested at this stage with the formation of small protrusions or formed very aberrant finger-like structures, indicating an essential role for cellular calcium in morphogenesis. We used pump and wild-type cells differentially labeled by expressing different wavelength (green and blue) forms of green fluorescent protein and three-dimensional (3-D) reconstruction of serial fluorescent imaging to visualize the movement of pump and wild-type cells within the aggregate. The results showed that the pump cells exhibited very aberrant cell movement and sorting within the forming mound, suggesting that the reduced cytosolic calcium affects movement required for tip formation. When allowed to form chimeric organisms with wild-type cells, pump cells preferentially localized to two bands, one at the prestalk/prespore boundary and the other in the very posterior of the organism, suggesting that pump cells are unable to properly sort. Expression of the calcium pump had little effect on the induction of prestalk- or prespore-specific genes, whereas extended treatment with EGTA blocked induction of both classes of cell-type-specific genes. Our results suggest a role for intracellular Ca2+ in controlling cell sorting and morphogenesis in Dictyostelium.

The role of the cortical cytoskeleton: F-actin crosslinking proteins protect against osmotic stress, ensure cell size, cell shape and motility, and contribute to phagocytosis and development

We generated Dictyostelium double mutants lacking the two F-actin crosslinking proteins alpha-actinin and gelation factor by inactivating the corresponding genes via homologous recombination. Here we investigated the consequences of these deficiencies both at the single cell level and at the multicellular stage. We found that loss of both proteins severely affected growth of the mutant cells in shaking suspension, and led to a reduction of cell size from 12 microns in wild-type cells to 9 microns in mutant cells. Moreover the cells did not exhibit the typical polarized morphology of aggregating Dictyostelium cells but had a more rounded cell shape, and also exhibited an increased sensitivity towards osmotic shock and a reduced rate of phagocytosis. Development was heavily impaired and never resulted in the formation of fruiting bodies. Expression of developmentally regulated genes and the final developmental stages that were reached varied, however, with the substrata on which the cells were deposited. On phosphate buffered agar plates the cells were able to form tight aggregates and mounds and to express prespore and prestalk cell specific genes. Under these conditions the cells could perform chemotactic signalling and cell behavior was normal at the onset of multicellular development as revealed by time-lapse video microscopy. Double mutant cells were motile but speed was reduced by approximately 30% as compared to wild type. These changes were reversed by expressing the gelation factor in the mutant cells. We conclude that the actin assemblies that are formed and/or stabilized by both F-actin crosslinking proteins have a protective function during osmotic stress and are essential for proper cell shape and motility.

Dictyostelium cell shape generation requires myosin II

Myosin II function has been implicated in non-muscle cell behaviors including movement, cytokinesis, and multicellular morphogenesis. Surprisingly, two-dimensional morphology and behavior of Dictyostelium amoebae lacking myosin II (mhcA-) is not dramatically altered when cells are examined using conventional imaging techniques. We have observed amoebae from the side (side-view microscopy or SVM) and find that wild-type but not mhcA- cells undergo extensive three-dimensional (3D) shape changes. These shape changes often occur above the plane of the substrate and are morphologically distinct from pseudopods. For example, unlike pseudopods, vertically extended cell regions are not enriched in F-actin and do not exclude organelles. In contrast, mhcA- cells generate F-actin-filled pseudopods and spread laterally but do not undergo vertical extension. When wild-type cells are removed from the substrate and shaken in suspension, they retain their irregular three-dimensional shape. Suspended mhcA- cells, however, rapidly become spherical. Thus, cells lacking myosin appear to be incapable of generating or maintaining 3D shape independent of a substrate. When a substrate is available, these cells can attach and spread on the surface in a myosin-independent manner. These previously undescribed defects of mhcA- cells reveal that 3D cell shape generation requires myosin II and is mechanistically distinct from pseudopod formation.

Targeted disruption of the Dictyostelium myosin essential light chain gene produces cells defective in cytokinesis and morphogenesis

We have previously demonstrated that the myosin essential light chain (ELC) is required for myosin function in a Dictyostelium cell line, 7–11, in which the expression of ELC was inhibited by antisense RNA overexpression. We have now disrupted the gene encoding the ELC (mlcE) in Dictyostelium by gene targeting. The mlcE- mutants provide a clean genetic background for phenotypic analysis and biochemical characterization by removing complications arising from the residual ELC present in 7–11 cells, as well as the possibility of mutations due to insertion of the antisense construct at multiple sites in the genome. The mlcE- mutants, when grown in suspension, exhibited the typical multinucleate phenotype observed in both myosin heavy chain mutants and 7–11 cells. This phenotype was rescued by introducing a construct that expressed the wild-type Dictyostelium ELC cDNA. Myosin purified from the mlcE- cells exhibited significant calcium ATPase activity, but the actin-activated ATPase activity was greatly reduced. The results obtained from the mlcE- mutants strengthen our previous conclusion based on the antisense cell line 7–11 that ELC is critical for myosin function. The proper localization of myosin in mlcE- cells suggests that its phenotypic defects primarily arise from defective contractile function of myosin rather than its mislocalization. The enzymatic defect of myosin in mlcE- cells also suggests a possible mechanism for the observed chemotactic defect of mlcE- cells. We have shown that while mlcE- cells were able to respond to chemoattractant with proper directionality, their rate of movement was reduced. During chemotaxis, proper directionality toward chemoattractant may depend primarily on proper localization of myosin, while efficient motility requires contractile function. In addition, we have analyzed the morphogenetic events during the development of mlcE- cells using lacZ reporter constructs expressed from cell type specific promoters. By analyzing the morphogenetic patterns of the two major cell types arising during Dictyostelium development, prespore and prestalk cells, we have shown that the localization of prespore cells is more susceptible to the loss of ELC than prestalk cells, although localization of both cell types is abnormal when developed in chimeras formed by mixing equal numbers of wild-type and mutant cells. These results suggest that the morphogenetic events during Dictyostelium development have different requirements for myosin.

Mutants lacking myosin II cannot resist forces generated during multicellular morphogenesis

We have used fluorescent labeling, confocal microscopy and computer-assisted motion analysis to observe and quantify individual wild-type and myosin II mutant cell behavior during early multicellular development in Dictyostelium discoideum. When cultured with an excess of unlabeled wild-type cells, labeled control cells are randomly distributed within aggregation streams, while myosin II mutant cells are found primarily at the lateral edges of streams. Wild-type cells move at average rates of 8.5 +/- 4.9 microns/min within aggregation streams and can exhibit regular periodic movement at 3.5 minute intervals; half as long as the 7 minute period reported previously for isolated cells. Myosin II mutants under the same conditions move at 5.0 +/- 4.8 microns/min, twice as fast as reported previously for isolated myosin II mutant cells, and fail to display regular periodic movement. When removed from aggregation streams myosin II mutant cells move at only 2.5 +/- 2.0 microns/min, while wild-type cells under these conditions move at 5.9 +/- 4.5 microns/min. Analysis of cell morphology further reveals that myosin II mutant cells are grossly and dynamically deformed within wild-type aggregation streams but not when removed from streams and examined in isolation. These data reveal that the loss of myosin II has dramatic consequences for cells undergoing multicellular development. The segregation of mutant cells to aggregation stream edges demonstrates that myosin II mutants are unable to penetrate a multicellular mass of wild-type cells, while the observed distortion of myosin II mutant cells suggests that the cortex of such cells is too flacid to resist forces generated during movement. The increased rate of mutant cell movement and distortion of mutant cell morphology seen within wild-type aggregation streams further argues both that movement of wild-type cells within a multicellular mass can generate traction forces on neighboring cells and that mutant cell morphology and behavior can be altered by these forces. In addition, the distortion of myosin II mutant cells within wild-type aggregation streams indicates that myosin is not required for the formation of cell-cell contacts. Finally, the consequences of the loss of myosin II for cells during multicellular development are much more severe than has been previously revealed for isolated cells. The techniques used here to analyze the behavior of individual cells within multicellular aggregates provide a more sensitive assay of mutant cell phenotype than has been previously available and will be generally applicable to the study of motility and cytoskeletal mutants in Dictyostelium.

Differentiation and patterning in Dictyostelium

Dictyostelium development is orchestrated by diffusible signals. Progress has been made in understanding how cAMP signaling triggers post-aggregative development and in defining the number of cell types that eventually differentiate. Ammonia is an unusual signal that may act by alkalinizing acidic vesicles. A chlorinated signal, differentiation-inducing factor (DIF), may be universal amongst the slime moulds. The first genes have been cloned using restriction enzyme mediated integration (REMI) insertional mutagenesis; one encodes a novel cytosolic protein essential for activation of adenylyl cyclase.

Stage-specific requirement for myosin II during Dictyostelium development

Dictyostelium cells that lack a functional myosin II heavy chain are motile and are capable of aggregation, but fail to undergo further multicellular development. We have used a Dictyostelium mutant expressing a cold-sensitive myosin heavy chain to examine the requirement for myosin throughout the course of development. The loss of myosin function upon cooling is rapid and reversible. Temperature-shift experiments reveal that myosin is essential during two different stages of development. During aggregation, myosin function appears to be necessary for cells to sort correctly in a way that allows further development to occur. During the final stage of development, it is required for the formation of a complete stalk and the raising of the spore head. Development between those stages, however, proceeds normally in the absence of myosin function. Aggregates at non-permissive temperature undergo an aberrant form of development resulting in a ball of cells. Calcofluor staining and reporter gene fusions reveal that these structures contain defective spores and a miniature stalk.