An intact SH3 domain is required for myosin I-induced actin polymerization.

The yeast type I myosins (MYO3 and MYO5) are involved in endocytosis and in the polarization of the actin cytoskeleton. The tail of these proteins contains a Tail Homology 2 (TH2) domain that constitutes a putative actin-binding site. Because of the important mechanistic implications of a second ATP-independent actin-binding site, we analyzed its functional relevance in vivo. Even though the myosin tail interacts with actin, and this interaction seems functionally important, deletion of a major portion of the TH2 domain did not abolish interaction. In contrast, we found that the SH3 domain of Myo5p significantly contributes to this interaction, implicating other proteins. We found that Vrp1p, the yeast homolog of WIP [Wiskott-Aldrich syndrome protein (WASP)-interacting protein], seems necessary to sustain the Myo5p tail-F-actin interaction. Consistent with recent results implicating the yeast type I myosins in regulating actin polymerization in vivo, we demonstrate that the C-terminal domain of Myo5p is able to induce cytosol-dependent actin polymerization in vitro, and that this activity requires both an intact Myo5p SH3 domain and Vrp1p.

Endocytic internalization in yeast and animal cells: similar and different.

The internalization step of endocytosis has been the focus of several laboratories during the last forty years. Unlike some other budding events in the cell, many fundamental questions regarding the molecular machinery involved in the mechanism of budding itself still remain unsolved. Over the last few years the general picture of the field has quickly evolved from the originally simplistic view which postulated that clathrin polymerization is the major force driving budding at the plasma membrane. Refinement of the assays and molecular markers to measure endocytosis in animal cells has shown that other factors in addition to the clathrin coat are required and that endocytosis can also take place through clathrin-independent mechanisms. At the same time, recent introduction of genetic approaches to study endocytosis has accelerated the identification of molecules required for this process. The isolation of endocytosis mutants in budding yeast has been especially fruitful in this respect. Preliminary comparison of the results obtained in yeast and animal cells did not seem to coincide, but further progress in both systems now suggests that part of the divergence originally seen may be due to the particular experimental approaches used rather than fundamental differences in endocytic mechanisms. In this review we present a short historical overview on the advances made in yeast and animal cells regarding the study of endocytosis, underlining both emerging similarities and still interesting differences.

Lysine-rich gamma-zeins are secreted in transgenic Arabidopsis plants.

We have previously shown that the maize (Zea mays L.) storage prolamine gamma-zein, accumulates in endoplasmic reticulum-derived protein bodies in transgenic plants of Arabidopsis thaliana (L.) ecotype R + P. The retention of gamma-zein in the endoplasmic reticulum was found to be mediated by structural features contained in the polypeptide, an N-terminal proline-rich and a C-terminal cysteine-rich domain which were necessary for the correct retention and assembly of gamma-zein within protein bodies (M.I. Geli et al., 1994, Plant Cell 6: 1911-1922). In the present work we incorporated in the gamma-zein gene lysine-rich coding sequences which were positioned after the N-terminal proline-rich domain and at five amino-acid residues from the C-terminus. The targeting of lysine-rich gamma-zeins was analyzed by expression of chimeric genes regulated by the cauliflower mosaic virus (CaMV) 35S promoter in transgenic Arabidopsis plants. The lysine-rich gamma-zeins were detected by immunoblotting and we found that these proteins were modified posttranslationally to reach their mature form. Subcellular fractionation and immunocytochemical studies demonstrated that glycosylated lysine-rich gamma-zeins were secreted to the cell wall of transgenic Arabidopsis leaf cells.

Distinct functions of calmodulin are required for the uptake step of receptor-mediated endocytosis in yeast: the type I myosin Myo5p is one of the calmodulin targets.

The uptake step of receptor-mediated endocytosis in yeast is dependent on the calcium binding protein calmodulin (Cmd1p). In order to understand the role that Cmd1p plays, a search was carried out for possible targets among the genes required for the internalization process. Co-immunoprecipitation, two-hybrid and overlay assays demonstrated that Cmd1p interacts with Myo5p, a type I unconventional myosin. Analysis of the endocytic phenotype and the Cmd1p-Myo5p interaction in thermosensitive cmd1 mutants indicated that the Cmd1p-Myo5p interaction is required for endocytosis in vivo. However, the Cmd1p-Myo5p interaction requirement was partially overcome by deleting the calmodulin binding sites (IQ motifs) from Myo5p, suggesting that these motifs inhibit Myo5p function. Additionally, genetic and biochemical evidence obtained with a collection of cmd1 mutant alleles strongly suggests that Cmd1p plays an additional role in the internalization step of receptor-mediated endocytosis in yeast.

Lysine-rich modified gamma-zeins accumulate in protein bodies of transiently transformed maize endosperms.

During maize seed development, endosperm cells synthesize large amounts of storage proteins, alpha-, beta-, and gamma-zeins, which accumulate within endoplasmic reticulum (ER)-derived protein bodies. The absence of lysine in all zein polypeptides results in an imbalance in the amino acid composition of maize seeds. We modified the maize gamma-zein gene through the introduction of lysine-rich (Pro-Lys)n coding sequences at different sites of the gamma-zein coding sequence. Maize endosperms were transiently transformed by biolistic bombardment with Lys-rich gamma-zein constructs under the control of the 1.7 kb gamma-zein seed-specific promoter and the cauliflower mosaic virus (CaMV) 35S promoter. When (Pro-Lys)n sequences were inserted contiguous to or in substitution of the Pro-Xaa region of the gamma-zein, high levels of protein were observed. In contrast, when (Pro-Lys)n sequences were inserted five residues from the C-terminal, the transcript was present but modified protein was not detected. These results suggest that only an appropriate positioning of Lys-rich inserts leads to the modified molecule displaying correct folding and stability. Subcellular localization analyses and immunoelectron microscopy studies on isolated protein bodies demonstrated that modified gamma-zeins accumulate within these organelles and co-localized with endogenous alpha- and gamma-zeins. The studies reported here show the feasibility of manipulating the gamma-zein gene in order to obtain stable and correctly targeted Lys-rich zeins in maize seeds.

Actin-, myosin- and ubiquitin-dependent endocytosis.

Endocytosis is a general term that is used to describe the internalization of external and plasma membrane molecules into the cell interior. In fact, several different mechanisms exist for the internalization step of this process. In this review we emphasize the work on the actin-dependent pathways, in particular in the yeast Saccharomyces cerevisiae, because several components of the molecular machinery are identified. In this yeast, the analysis of endocytosis in various mutants reveals a requirement for actin, calmodulin, a type I myosin, as well as a number of other proteins that affect actin dynamics. Some of these proteins have homology to proteins in animal cells that are believed to be involved in endocytosis. In addition, the demonstration that ubiquitination of some cell surface molecules is required for their efficient internalization is described. We compare the actin, myosin and ubiquitin requirements for endocytosis with recent results found studying these processes using Dictyostelium discoideum and animal cells.

Role of type I myosins in receptor-mediated endocytosis in yeast.

Type I myosins are thought to drive actin-dependent membrane motility, but the direct demonstration in vivo of their involvement in specific cellular processes has been difficult. Deletion of the genes MYO3 and MYO5, which encode the yeast type I myosins, almost abolished growth. A double-deleted mutant complemented with a MYO5 temperature-sensitive allele (myo5-1) showed a strong defect in the internalization step of receptor-mediated endocytosis, whereas the secretory pathway remained apparently unaffected. Thus, myosin I activity is required for a budding event in endocytosis but not for several other aspects of membrane traffic.

end5, end6, and end7: mutations that cause actin delocalization and block the internalization step of endocytosis in Saccharomyces cerevisiae.

Four mutants defective in endocytosis were isolated by screening a collection of temperature-sensitive yeast mutants. Three mutations define new END genes: end5-1, end6-1, and end7-1. The fourth mutation is in END4, a gene identified previously. The end5-1, end6-1, and end7-1 mutations do not affect vacuolar protein localization, indicating that the defect in each mutant is specific for internalization at the plasma membrane. Interestingly, localization of actin patches on the plasma membrane is affected in each of the mutants. end5-1, end6-1, and end7-1 are allelic to VRP1, RVS161, and ACT1, respectively. VRP1 and RVS161 are required for correct actin localization and ACT1 encodes actin. To our surprise, the end6-1 mutation fails to complement the act1-1 mutation. Disruption of the RVS167 gene, which is homologous to END6/RVS161 and which is also required for correct actin localization, also blocks endocytosis. The end7-1 mutant allele has a glycine 48 to aspartic acid substitution in the DNase I-binding loop of actin. We propose that Vrp1p, Rvs161p, and Rvs167p are components of a cytoskeletal structure that contains actin and fimbrin and that is required for formation of endocytic vesicles at the plasma membrane.

Two Structural Domains Mediate Two Sequential Events in [gamma]-Zein Targeting: Protein Endoplasmic Reticulum Retention and Protein Body Formation.

[gamma]-Zein is a maize storage protein synthesized by endosperm cells and stored together with [alpha]- and [beta]-zeins in specialized organelles called protein bodies. Previous studies have shown that in maize there is only one type of protein body and it is derived directly from the endoplasmic reticulum (ER). In this article, we describe the domains of [gamma]-zein involved in ER retention and the domains involved in protein body formation. To identify the signal responsible for [gamma]-zein retention in ER-derived protein bodies, DNAs encoding various deletion mutants of [gamma]-zein were constructed and introduced into Arabidopsis as a heterologous system. By using pulse-chase experiments and immunoelectron microscopy, we demonstrated that the deletion of a proline-rich domain at the N terminus of [gamma]-zein puts an end to its retention in the ER; this resulted in the secretion of the mutated protein. The amino acid sequence of [gamma]-zein necessary for ER retention is the repeat domain composed of eight units of the hexapeptide PPPVHL. In addition, we observed that only those [gamma]-zein mutants that contained both the proline-rich repeat domain and the C-terminal cysteine-rich domain were able to form ER-derived protein bodies. We suggest that the retention of [gamma]-zein in the ER could be a result of a protein-protein association or a transient interaction of the repeat domain with ER membranes.

Role of structural domains for maize gamma-zein retention in Xenopus oocytes.

In order to examine the role of cysteine (Cys)-rich domains in the accumulation of maize (Zea mays L.) gamma-zein within the endoplasmic-reticulum-derived protein bodies, we studied the localization of gamma-zein and of two truncated forms of gamma-zein in Xenopus laevis oocytes. The two derivatives were constructed from a DNA encoding the gamma-zein: one by deletion of the Pro-X linker region (21 amino acids) and the other by deletion of the Cys-rich domain (94 amino acids). In-vitro-synthesized transcripts were injected into oocytes and the distribution of the translation products was then analyzed. The entire gamma-zein and both truncated forms of the gamma-zein had accumulated efficiently in microsomes and no traces of secretion were observed. We suggest that neither C-terminal Cys-rich nor Pro-X domains are essential for gamma-zein retention in oocyte vesicles. Therefore, structural features derived from disulphide bonds are not necessary for gamma-zein targeting on the endoplasmic reticulum.