Research this year was focused on: (1) the molecular basis of the localization of Acanthamoeba myosin IC (AMIC) to specific membranes in the cell, (2) the role of S-adenosylhomocysteine hydrolase (SAHH) in chemotaxis, (3) the properties of Tyr53-phosphorylated actin (pY53-actin), and (4) the functional importance of actin Tyr53.? ? (1) The 466-residue tail of AMIC has an N-terminal basic region followed by a Gly/Pro/Ala-rich (GPA-1) region, an SH3 domain, and a C-terminal GPA-2 region. We had shown previously (Hwang et al., 2007) that the basic region contains a putative plekstrin homology (PH) domain, and that the tail is folded back on itself at the junction between the basic region and GPA-1 (Ishikawa et al., 2004) with apparent interactions between the putative PH domain and the SH3 domain (Hwang et al. 2007). We speculated that, by analogy to some, but not all, proteins with a PH domain, the putative PH domain of AMIC might be involved in the high-affinity binding of AMIC to membranes and acidic phospholipid vesicles, which is known to be mediated through the basic region (Doberstein & Pollard, 1992). We further speculated that the folded tail might affect (regulate) binding of the basic region to acidic phospholipid vesicles. ? ? Last year, we had found that AMIC binds with higher affinity to unilamellar phosphatidylcholine vesicles containing phosphoinositol bisphosphate (PIP2)than to vesicles containing phosphatidylserine (PS). The tighter binding was fully attributable to the 4-fold higher negative charge of PIP2. For example, AMIC binds with equal affinity to vesicles containing 5% PIP2 or 25% PS. The folded tail did not affect binding in as much as full-length AMIC had the same binding affinity as AMIC truncated after the basic region. We found also that AMIC binds with 10-fold higher affinity to vesicles containing 5% PIP2+25% PS than to vesicles containing only 25% PS or 5% PIP2. This suggested that AMIC might be targeted in vivo to PIP2-enriched regions of membranes in the amoebae. Consistent with this hypothesis, we find that AMIC and PIP2 co-localize at endocytic cups, pseudopods, membrane ruffles and the leading edge of motile cells.? ? We have extended these studies this year showing that the putative PH domain probably is not a major contributor to binding of AMIC to acidic phospholipids. Mutation of an Arg residue to Ala that has been shown by others (Hokanson et al. (2006) Mol. Biol. Cell 103, 3118-3123) to block binding of mouse myo1c to acidic phospholipids had only minimal effect on the binding of AMIC. Myo1c is a mammalian class-I myosin with a basic region and putative PH domain similar to AMIC.? ? We then identified a 13-amino acid sequence of basic-hydrophobic-basic (BHB) residues that seems to be a major contributor to AMIC-phospholipid binding. A synthetic peptide with that sequence blocked binding of AMIC to phospholipid vesicles, and mutation of two or more of the hydrophobic residues to Ala inactivated the peptide. Furthermore, the corresponding peptide from Acanthamoeba myosin IB and Dictyostelium myosin ID, whose sequences are similar to the AMIC sequence, also block binding of AMIC to phospholipid vesicles, whereas the peptide from AMIA, which has a substantially different sequence, did not. Consistent with these in vitro results, AMIC, AMIB and DMID bind are known to bind the plasma membrane in vivo but AMIA is largely cytoplasmic.? ? (2) We had shown previously that S-adenosylhomocysteine hydrolase (SAHH), an essential component of the trans-methylation pathway in all eukaryotic cells, is diffuse in the cytoplasm of resting cells but localizes with polymerized actin at the front of both chemotaxing Dictyostelium amoebae and human neutrophils (Shu et al., 2006). Thus, SAHH is one of a list of proteins that relocate to either the front or rear of a chemotaxing cell, suggesting a previously unknown role for methylation in chemotaxis. Attempts to isolatte SAHH-knockout cells were unsuccessful. Dictyostelium cell lines were created that express only 25% of the normal amount of SAHH, but, not surpisingly given the multiple biochemical pathways involving methylation, it was difficult to relate the phenotype to any specific component of the chemotaxis process.? ? We attempted this year to identify a protein or proteins that might be associated with SAHH in chemotaxing cells, but not in non-chemotaxing cells. All efforts have been unsuccessful.? ? (3) Tyr53-phosphorylated actin (pY53-actin) forms late in the Dictyostelium developmental cycle, and also when cells in the amoeboid stage are subjected to stress. Tyr53-phosphorylation increases actin's critical concentration, reduces the rate of polymerization, substantially inhibits nucleation and elongation from the pointed-end of actin filaments, moderately reduces the rate of elongation from the barbed-end, and greatly reduces actin's affinity for DNase I (Liu et al., 2006). Under polymerization conditions, pY53-actin forms small oligomers that are converted to typical long filaments upon addition of myosin subfragment 1, which is activated normally by filamentous pY53-actin.? ? This year, in collaboration with Drs. Roberto Dominguez and Kyuwon Baek, University of Pennsylvania, we obtained and analyzed high resolution crystal structures of Dictyostelium actin and pY53-actin complexes with gelsolin segment 1, and of unphosphorylated actin with profilin/VASP (Baek et al. 2008). The phosphate on pY53-actin formed hydrogen bonds within the DNase I-binding loop (D-loop) stabilizing its conformation, which could explain all of the previously reported differences between pY53-actin and actin, and also our new observation that phosphorylation inhibits cleavage of the D-loop by subtilisin. Unexpectedly, we found that phosphorylation also inhibits the rate of exchange of actin-bound ATP. Comparison of the actin/profilin structure to the structure of other actin complexes showed that actin sub-domains 1 and 3 close around the profilin molecule which results in a moderate opening of the nucleotide cleft between subdomains 2 and 4, probably explaining the stimulation of nucleotide exchange by profilin.? ? (4) We have expressed FLAG-tagged actin mutants Y53F, Y53A, Y53D and Y53E, and N-FLAG-tagged-wild type (WT) actin in Dictyostelium. Each construct was expressed to a level of about 25% of the total actin. The WT and Y53F constructs had no effect on the phenotype, and none of the constructs affected cell growth or pinocytosis. The WT and Y53F constructs also had no effect on chemotaxis, cell streaming or development to mature fruiting bodies, showing that the FLAG-tag was not inhibitory. However, the Y53A, Y53D and Y53E all inhibited cell streaming, although they could still chemotax as individual cells, and development was blocked at the slug stage. ? ? When polymerized, purified FLAG-WT and FLAG-Y53F actins form filaments that are essentially indistinguishable from filaments formed by endogeous WT but the Y53D, Y53E and Y53A actin polymerize into short oligomers. Immunostaining experiments suggest that al of the Y53 mutants co-polymerize with endogenous actin in the amoebae and studies of the polymerization properties and filament characterization of mixtures of Y53A and endogenous actin are in progress.
Showing the most recent 10 out of 21 publications
