The intermediate filament protein consensus motif of helix 2B: its atomic structure and contribution to assembly.

Nearly all intermediate filament proteins exhibit a highly conserved amino acid motif (YRKLLEGEE) at the C-terminal end of their central alpha-helical rod domain. We have analyzed its contribution to the various stages of assembly by using truncated forms of Xenopus vimentin and mouse desmin, VimIAT and DesIAT, which terminate exactly before this motif, by comparing them with the wild-type and tailless proteins. It is surprising that in buffers of low ionic strength and high pH where the full-length proteins form tetramers, both VimIAT and DesIAT associated into various high molecular weight complexes. After initiation of assembly, both VimIAT and DesIAT aggregated into unit-length-type filaments, which rapidly longitudinally annealed to yield filaments of around 20 nm in diameter. Mass measurements by scanning transmission electron microscopy revealed that both VimIAT and DesIAT filaments contained considerably more subunits per cross-section than standard intermediate filaments. This indicated that the YRKLLEGEE-motif is crucial for the formation of authentic tetrameric complexes and also for the control of filament width, rather than elongation, during assembly. To determine the structure of the YRKLLEGEE domain, we grew crystals of peptides containing the last 28 amino acid residues of coil 2B, chimerically fused at its amino-terminal end to the 31 amino acid-long leucine zipper domain of the yeast transcription factor GCN4 to facilitate appropriate coiled-coil formation. The atomic structure shows that starting from Tyr400 the two helices gradually separate and that the coiled coil terminates with residue Glu405 while the downstream residues fold away from the coiled-coil axis.

Calcium-mediated structural changes of native nuclear pore complexes monitored by time-lapse atomic force microscopy.

Nuclear pore complexes (NPCs) are large macromolecular assemblies embedded in the double membrane nuclear envelope. They are the major gateways mediating transport of ions, small molecules, proteins, RNAs, and ribonucleoprotein particles in and out of the nucleus in interphase cells. Understanding structural changes at the level of individual pores will be a prerequisite to eventually correlate the molecular architecture of the NPC with its distinct functional states during nucleocytoplasmic transport. Toward this goal, we have employed time-lapse atomic force microscopy of native NPCs kept in buffer, and recorded calcium-mediated structural changes such as the opening (i.e. +Ca2+) and closing (i.e. -Ca2+) of individual nuclear baskets. Most likely, this structural change of the nuclear basket involves its distal ring which may act as an iris-like diaphragm. In order to directly correlate distinct structural features with corresponding functional states and dynamic aspects, we also addressed the question of whether the "central plug" or "transporter" actually represents a calcium-sensitive component of the NPC involved in mediating nucleocytoplasmic transport. Our data indicate that in the absence of ATP, cytoplasmic plugging/unplugging of the NPC is insensitive to calcium.

Determination of the inelastic mean free path of electrons in vitrified ice layers for on-line thickness measurements by zero-loss imaging.

The inelastic mean free path of 120 keV electrons in vitrified ice layers has been determined in an energy-filtering TEM. From the ratio of the unfiltered and zero-loss-filtered image intensities recorded with a slow-scan CCD camera, the relative sample thickness t/Lambda can be calculated. For calibration, the geometric ice thickness was measured by imaging a tilted view of a cylindrical hole which had been burnt into the ice layer. The total inelastic mean free path was found to be 161 nm, and the partial inelastic mean free path for an acceptance angle of 4.2 mrad was 232 nm. These results were built into a standard protocol for use in cryo-electron microscopy allowing on-line measurements of local ice-layer thicknesses by zero-loss-filtered/unfiltered imaging.