Igor Weber

A role for myosin VII in dynamic cell adhesion

BACKGROUND The initial stages of phagocytosis and cell motility resemble each other. The extension of a pseudopod at the leading edge of a migratory cell and the formation of a phagocytic cup are actin dependent, and each rely on the plasma membrane adhering to a surface during dynamic extension. RESULTS A myosin VII null mutant exhibited a drastic loss of adhesion to particles, consistent with the extent of an observed decrease in particle uptake. Additionally, cell-cell adhesion and the adhesion of the leading edge to the substratum during cell migration were defective in the myosin VII null cells. GFP-myosin VII rescued the phagocytosis defect of the null mutant and was distributed in the cytosol and recruited to the cortical cytoskeleton, where it appeared to be enriched at the tips of filopods. It was also localized to phagocytic cups, but only during the initial stages of particle engulfment. During migration, GFP-myosin VII is found at the leading edge of the cell. CONCLUSIONS Myosin VII plays an important role in mediating the initial binding of cells to substrata, a novel role for an unconventional myosin.

Cytokinesis mediated through the recruitment of cortexillins into the cleavage furrow.

The fact that substrate‐anchored Dictyostelium cells undergo cytokinesis in the absence of myosin II underscores the importance of other proteins in enabling the cleavage furrow to constrict. Cortexillins, a pair of actin‐bundling proteins, are required for normal cleavage. They are targeted to the incipient furrow in wild‐type and, more prominently, in myosin II‐null cells. No other F‐actin bundling or cross‐linking protein tested is co‐localized. Green fluorescent protein fusions show that the N‐terminal actin‐binding domain of cortexillin I is dispensable and the C‐terminal region is sufficient for translocation to the furrow and the rescue of cytokinesis. Cortexillins are suggested to have a targeting signal for coupling to a myosin II‐independent system that directs transport of membrane proteins to the cleavage furrow.

Domain analysis of cortexillin I: actin‐bundling, PIP2‐binding and the rescue of cytokinesis

Cortexillins are actin‐bundling proteins that form a parallel two‐stranded coiled‐coil rod. Actin‐binding domains of the α‐actinin/spectrin type are located N‐terminal to the rod and unique sequence elements are found in the C‐terminal region. Domain analysis in vitro revealed that the N‐terminal domains are not responsible for the strong actin‐filament bundling activity of cortexillin I. The strongest activity resides in the C‐terminal region. Phosphatidylinositol 4,5‐bisphosphate (PIP2) suppresses this bundling activity by binding to a C‐terminal nonapeptide sequence. These data define a new PIP2‐regulated actin‐bundling site. In vivo the PIP2‐binding motif enhances localization of a C‐terminal cortexillin I fragment to the cell cortex and improves the rescue of cytokinesis. This motif is not required, however, for translocation to the cleavage furrow. A model is presented proposing that cortexillin translocation is based on a mitotic cycle of polar actin polymerization and midzone depolymerization.

Recruitment of cortexillin into the cleavage furrow is controlled by Rac1 and IQGAP‐related proteins

Cytokinesis in eukaryotic organisms is under the control of small GTP‐binding proteins, although the underlying molecular mechanisms are not fully understood. Cortexillins are actin‐binding pro teins whose activity is crucial for cytokinesis in Dictyostelium. Here we show that the IQGAP‐related and Rac1‐binding protein DGAP1 specifically interacts with the C‐terminal, actin‐bundling domain of cortexillin I. Like cortexillin I, DGAP1 is enriched in the cortex of interphase cells and translocates to the cleavage furrow during cytokinesis. The activated form of the small GTPase Rac1A recruits DGAP1 into a quaternary complex with cortexillin I and II. In DGAP1− mutants, a complex can still be formed with a second IQGAP‐related protein, GAPA. The simultaneous elimination of DGAP1 and GAPA, however, prevents complex formation and localization of the cortexillins to the cleavage furrow. This leads to a severe defect in cytokinesis, which is similar to that found in cortexillin I/II double‐null mutants. Our observations define a novel and functionally significant signaling pathway that is required for cytokinesis.

Cytokinesis without myosin II

The ability of substrate-anchored Dictyostelium cells to divide without myosin II has opened the possibility of analysing the formation of cleavage furrows in the absence of a contractile ring made of filamentous myosin and actin. Similar possibilities exist in mutants of budding yeast and, less strictly, also in drug-treated mammalian cells. Myosin-II-independent activities in Dictyostelium include the microtubule-induced programming of the cell surface into ruffling areas and regions that are converted into a concave furrow, as well as the translocation of cortexillins and cross-linked membrane proteins towards the cleavage furrow. A centripetal flow of actin filaments followed by their disassembly in the cleavage furrow is proposed to underlie the translocation.

Dynamic organization of the actin system in the motile cells of Dictyostelium

The actin system forms a supramolecular, membrane-associated network that serves multiple functions in Dictyostelium cells, including cell motility controlled by chemoattractant, phagocytosis, macropinocytosis, and cytokinesis. In executing these functions the monomeric G-actin polymerizes reversibly, and the actin filaments are assembled into membrane-anchored networks together with other proteins involved in shaping the networks and controlling their dynamics. Most impressive is the speed at which actin-based structures are built, reorganized, or disassembled. We used GFP-tagged coronin and Arp3, an intrinsic constituent of the Arp2/3 complex, as examples of proteins that are recruited to highly dynamic actin-filament networks. By fluorescence recovery after photobleaching (FRAP), average exchange rates of cell-cortex bound coronin were estimated. A nominal value of 5 s for half-maximal incorporation of coronin into the cortex, and a value of 7 s for half-maximal dissociation from cortical binding sites has been obtained. Actin dynamics implies also flow of F-actin from sites of polymerization to sites of depolymerization, i.e. to the tail of a migrating cell, the base of a phagocytic cup, and the cleavage furrow in a mitotic cell. To monitor this flow, we expressed in Dictyostelium cells a GFP-tagged actin-binding fragment of talin. This fragment (GFP-TalC63) translocates from the front to the tail during cell migration and from the polar regions to the cleavage furrow during mitotic cell division. The intrinsic dynamics of the actin system can be manipulated in vivo by drugs or other probes that act either as inhibitors of actin polymerization or as stabilizers of filamentous actin. In order to investigate structure–function relationships in the actin system, a technique of reliably arresting transient network structures is in demand. We discuss the potential of electron tomography of vitrified cells to visualize actin networks in their native association with membranes.

Differential localization of the Dictyostelium kinase DPAKa during cytokinesis and cell migration.

The Dictyostelium kinase DPAKa is a member of the p21-activated kinase (PAK) family, consisting of an N-terminal domain characterized by a coiled-coil region and proline-rich motifs, a Rac-binding CRIB-domain, and a highly conserved C-terminal kinase domain. In this study we show that cells overexpressing a C-terminal DPAKa fragment comprising the kinase domain are significantly impaired in motility and phagocytosis, whereas DPAKa-null cells display no obvious phenotypic change. We analyzed the in vivo localization of full-length and truncated DPAKa tagged with green fluorescent protein (GFP). The N-terminal fragments show a highly dynamic cortical localization without a permanent polarized enrichment, whereas the C-terminal fragment is homogenously distributed throughout the cell. The localization of full-length DPAKa is similar to that of myosin II at the rear end of locomoting cells and at the base of phagocytic cups. During mitosis DPAKa is gradually recruited to the cell cortex starting at metaphase, which also parallels the dynamics of myosin II cortical recruitment. However, in contrast to myosin II, DPAKa does not accumulate in the cleavage furrow but stays uniformly distributed throughout the cell cortex. This finding contrasts with previous work claiming accumulation of DPAKa in the cleavage furrow of dividing cells. Our results suggest that the N-terminus directs DPAKa to the cortex, and the C-terminus is necessary for restricting its localization to the rear of moving cells during chemotaxis. Therefore, DPAKa may play distinct roles in myosin II regulation during cell movement and cell division.

Tyrosine phosphorylation of PYK2 mediates heregulin-induced glioma invasion: Novel heregulin/HER3-stimulated signaling pathway in glioma

Receptor tyrosine kinases of the EGFR family transmit extracellular signals that control diverse cellular functions such as proliferation, differentiation and survival. Signaling function of a member of this family, HER3, is believed to be impaired due to deviations in its kinase consensus motifs. Here we address the functional role and signaling mechanisms of HER3. HER3 preferentially forms heterodimers with HER2 inducing the most potent mitogenic signal among EGFR family members. Our data show that in a glioma‐derived cell line the cytoplasmic tyrosine kinase PYK2 is constitutively associated with HER3 and that stimulation with Heregulin results in PYK2 tyrosine phosphorylation. HER3, but not HER2, mediates the phosphorylation of the C‐terminal region of PYK2 to promote a mitogenic response through activation of the MAPK pathway. A central role of PYK2 in signaling downstream of HER3 is substantiated by the demonstration that expression of a dominant‐negative PYK2‐KM construct abrogates the Heregulin‐induced MAPK activity and inhibits the invasive potential of glioma cells. These results suggest a novel Heregulin/HER3‐stimulated signaling pathway in glioblastoma‐derived cell lines that involves phosphorylation of PYK2 and mediates invasiveness of glioma cells.

Two-step positioning of a cleavage furrow by cortexillin and myosin II

BACKGROUND Myosin II, a conventional myosin, is dispensable for mitotic division in Dictyostelium if the cells are attached to a substrate, but is required when the cells are growing in suspension. Only a small fraction of myosin II-null cells fail to divide when attached to a substrate. Cortexillins are actin-bundling proteins that translocate to the midzone of mitotic cells and are important for the formation of a cleavage furrow, even in attached cells. Here, we investigated how myosin II and cortexillin I cooperate to determine the position of a cleavage furrow. RESULTS Using a green fluorescent protein (GFP)-cortexillin I fusion protein as a marker for priming of a cleavage furrow, we found that positioning of a cleavage furrow occurred in two steps. In the first step, which was independent of myosin II and substrate, cortexillin I delineated a zone around the equatorial region of the cell. Myosin II then focused the cleavage furrow to the middle of this cortexillin I zone. If asymmetric cleavage in the absence of myosin II partitioned a cell into a binucleate and an anucleate portion, cell-surface ruffles were induced along the cleavage furrow, which led to movement of the anucleate portion along the connecting strand towards the binucleate one. CONCLUSIONS In myosin II-null cells, cleavage furrow positioning occurs in two steps: priming of the furrow region and actual cleavage, which may proceed in the middle or at one border of the cortexillin ring. A control mechanism acting at late cytokinesis prevents cell division into an anucleate and a binucleate portion, causing a displaced furrow to regress if it becomes aberrantly located on top of polar microtubule asters.

A dual role for Rac1 GTPases in the regulation of cell motility

Rac proteins are the only canonical Rho family GTPases in Dictyostelium, where they act as key regulators of the actin cytoskeleton. To monitor the dynamics of activated Rac1 in Dictyostelium cells, a fluorescent probe was developed that specifically binds to the GTP-bound form of Rac1. The probe is based on the GTPase-binding domain (GBD) from PAK1 kinase, and was selected on the basis of yeast two-hybrid, GST pull-down and fluorescence resonance energy transfer assays. The PAK1 GBD localizes to leading edges of migrating cells and to endocytotic cups. Similarly to its role in vertebrates, activated Rac1 therefore appears to control de novo actin polymerization at protruding regions of the Dictyostelium cell. Additionally, we found that the IQGAP-related protein DGAP1, which sequesters active Rac1 into a quaternary complex with actin-binding proteins cortexillin I and cortexillin II, localizes to the trailing regions of migrating cells. Notably, PAK1 GBD and DGAP1, which both bind to Rac1-GTP, display mutually exclusive localizations in cell migration, phagocytosis and cytokinesis, and opposite dynamics of recruitment to the cell cortex upon stimulation with chemoattractants. Moreover, cortical localization of the PAK1 GBD depends on the integrity of the actin cytoskeleton, whereas cortical localization of DGAP1 does not. Taken together, these results imply that Rac1 GTPases play a dual role in regulation of cell motility and polarity in Dictyostelium.