|
|
The long-term goal of our research is to understand the signal-transduction mechanisms that control cellular growth and morphogenesis during the eukaryotic cell cycle and how these mechanisms impinge on other cellular processes, including virulence mechanisms. Cellular morphogenesis is the generation and maintenance of the three-dimensional organization of subcellular constituents that ultimately determines an organism's characteristic shape and growth patterns. Proper cellular morphogenesis is critical for many cellular and developmental processes, including the partitioning of cellular constituents during cell division, the generation of polarity and cell shape in early embryogenesis, axon migration and neurite outgrowth in early development, the intracellular movement of organelles and proteins in polarized epithelial cells, and the stimulated secretion of neurotransmitters.
An important component of morphogenesis is the establishment of cell polarity, which can be divided into several interdependent and heirarchial processes including the determination of an axis of polarization and the subsequent asymmetric distribution of cellular components along that axis. At the molecular level, cell polarity is best understood in the budding yeast Saccharomyces cerevisiae. Recent results strongly suggest that the molecular mechanisms controlling cell polarity in yeast are highly conserved in other eukaryotes. Therefore, the study of these fundamental signal-transduction mechanisms in yeast cells should shed light on similar mechanisms in other eukaryotes. Cytoskeletal elements, such as microtubules, actin-based microfilaments, and intermediate filaments are intimately associated with the generation of cellular polarity and disruption of these networks is correlated with high metastatic potential in fibrosarcoma, melanoma, and colon cancers, suggesting a possible causal relationship between the disruption of normal morphogenesis and cancer.
The mechanics of cell polarity during the S. cerevisiae mitotic cell cycle can be divided into these sequential events: (i) non-random bud-site selection; (ii) organization of proteins at the bud site leading to bud emergence; (iii) polarized growth, including asymmetric organellar inheritance, towards and within the enlarging bud; (iv) a switch from apical bud growth to isotropic growth; and (v) cessation of polarized growth with cytokinesis and septum formation at the mother-bud junction. Genetic and biochemical studies have identified numerous components of signal-transduction pathways (see Figure), including several GTPases and components of the actin cytoskeleton, that are involved in the cell polarity pathway, although the mechanistic links within this pathway are largely not understood and the regulatory mechanisms that control the synthesis, modification, and subcellular localization of these proteins have yet to be studied in depth.
We are interested in the signal-transduction mechanisms that control cellular polarity in yeast and have focused on the functions of the Cdc42p GTPase and its guanine-nucleotide exchange factor (GEF) Cdc24p. Cdc42p is a member of the Rac/Rho subfamily of the Ras superfamily of GTPases that act as molecular switches in the control of a variety of eukaryotic processes. Our analyses of the morphological phenotypes of S. cerevisiae cdc42 mutant alleles indicated that Cdc42p functions in multiple stages of the cell cycle, including the bud-site selection process, the organization of the presumptive bud site, the subsequent actin-based polarized cell growth, and in cytokinesis. For instance, the cdc42ts, cdc42D118A, cdc42C188S, and cdc42T35A mutants display G1/S arrests with a block of bud emergence while the cdc42V44A and cdc42K186R alleles show a cytokinesis defect. The cdc42D118A mutant exhibited a temperature-dependent, dominant-negative phenotype, suggesting that the Cdc42D118A mutant protein was inactive (GDP-bound) but could bind and sequester cellular factors necessary for the budding process. One strong candidate for this cellular factor was Cdc24p due to its ability to multicopy suppress the cdc42D118A mutant phenotype and because Cdc24p interacts with Cdc42D118Ap in two-hybrid studies.
Cdc24p has amino-acid sequence similarity with the Dbl onco-protein, which acts as a GEF for human Cdc42p. Subsequent biochemical evidence indicated that Cdc24p has GEF activity against yeast Cdc42p in vitro. This GEF activity is calcium-independent, which is interesting because certain cdc24 (cls4) mutants are sensitive to high levels of Ca2+ and Cdc24p has two putative calcium-binding domains. Ca2+ does affect the interaction between Cdc24p and Bem1p, but the functional consequences of this are unknown.
We and others have characterized highly conserved (80-85% identical) functional homologs of S. cerevisiae Cdc42p in a variety of other eukaryotes including C. albicans, S. pombe, C. elegans, D. melanogaster , and H. sapiens, suggesting that yeast cell polarity genes may have conserved functional homologs in these other eukaryotes. The Cdc42p homolog in C. albicans plays a critical role in polarized budded cell growth, hyphal elongation, and potentially virulence (see C. albicans research page). Cdc42p homologs in C. elegans and D. melanogaster function in the polarized events of axon migration and neurite outgrowth. In mammalian cells, Cdc42p mediates both rapid cytoskeletal changes and slower responses, including the induction of filopodia in response to bradykinin, activation of the JNK and p38 MAP kinase cascades that regulate the transcription factor c-jun, activation of p70s6k kinase that phosphorylates ribosomal protein S6 leading to translational control of selective mRNA transcripts during the G1/S transition, G-protein coupled Na+-K+ exchange, and progression through the cell cycle. Recent evidence using specific Cdc42 effector-domain mutations has indicated that the induction of filopodia, activation of p70s6k kinase, and the Pak-dependent activation of the JNK MAP kinase cascade are mediated independently, suggesting a divergence of signalling pathways beyond Cdc42p (see Figure).
 |
How does Cdc42p regulate multiple signal-transduction pathways? It is clear that Cdc42p can interact with multiple protein kinases, all of which contain a conserved Cdc42p-binding motif, termed PBD or CRIB. The first kinase identified was a rat brain protein kinase termed p65pak which specifically interacted with GTP-bound Cdc42p. As a result of this association, the p65pak kinase autophosphorylated on serine and threonine residues and became activated towards exogenous substrates. The interaction between these kinases and Cdc42p is eliminated by the cdc42T35A effector-domain mutation, suggesting that these kinases are downstream effectors of Cdc42p function. In S. cerevisiae, there are three p65pak homologs, termed Ste20p, Cla4p, and Skm1. Ste20p is required for mating and filamentous growth of diploids, which is another example of polarized growth, and Cla4p plays a role in cytokinesis. Like p65pak, Ste20p and Cla4p bind to GTP-bound Cdc42p in vitro and interaction with Cdc42p in vivo is required for correct localization of a GFP-Ste20 fusion protein to sites of polarized cell growth. Thus, these kinases may be among the first kinases activated in parallel signaling complexes that diverge from Cdc42p (see Figure). In addition, Cdc42p interacts with a yeast formin protein, Bni1p, which also interacts with profilin and actin thereby providing a possible link between the Cdc42p signaling cascade and actin reorganization. For numerous downstream effectors to interact with Cdc42p, they must either interact through different Cdc42p domains or at different times in the cell cycle or compete for the same Cdc42p effector domain. Recent evidence from our lab suggests that there are differential interactions between downstream kinases and different Cdc42p effector-domain residues.
How does Cdc42p transmit cell polarity signals to downstream kinase cascades? The answer to this question is unknown in both yeast and mammalian systems. However, a key component of this signal transmission must be the subcellular localization of Cdc42p and its downstream effectors. One of the major roles of Ras is to target and anchor its downstream effector kinase Raf to the plasma membrane. The subcellular localization of the Cdc42 effector Ste20p to sites of polarized growth is dependent on binding to Cdc42p. We have localized Cdc42p to internal membranes, including the nuclear and vacuolar membranes, and the periphery of cells, where it clusters at sites of polarized growth. Cdc42p contains the C-terminal 183Lys-Lys-Ser-Lys-Lys-Cys-Thr-Ile-Leu sequence that is modified by geranylgeranylation at the Cys residue. The CTIL sequence is sufficient for localization to internal membranes and the KKSKK sequences are necessary and sufficient for localization to the periphery of the cell at polarized growth sites, but neither is sufficient for clustering to occur. The Cdc24p GEF localizes to the nucleus in haploid cells and co-localizes with clustered Cdc42p. Cdc24p localization is a function of specific targeting to these sites and anchoring within a cytoskeletal complex at these sites.
Experimentally, we use classical and molecular genetic techniques along with immunological and biochemical protocols to study cellular polarity in S. cerevisiae. The types of experiments we will be performing over the next five years include: (1) characterizing the interactions between Cdc42p, Cdc24p, and Rdi1p, using bimolecular fluorescence complementation (BiFC), biochemical assays and site-specific mutagenesis protocols; (2) studying the differential membrane localization of Cdc42p, Cdc24p, and downstream effectors, using fusion proteins to the Green Fluorescent Protein (GFP) and specific antibodies with cell fractionation and immunolocalization techniques; and (3) identifying and isolating new genes whose products interact with Cdc42p and Cdc24p, using classical and molecular genetic techniques.
It is becoming increasing clear that the signal-transduction mechanisms controlling cell polarity are highly conserved among eukaryotic organisms. In order to decipher these mechanisms, one must bring to bear the full arsenal of experimental approaches, including genetic, molecular genetic, biochemical, immunological, and cell biological approaches. The study of cell polarity in yeast allows us to bring this full arsenal to bear and should allow us to develop useful and instructive mechanistic paradigms that may be applicable to multiple signal-transduction pathways in many, if not all, eukaryotic organisms.