Structure of the AAA ATPase p97.

p97, an abundant hexameric ATPase of the AAA family, is involved in homotypic membrane fusion. It is thought to disassemble SNARE complexes formed during the process of membrane fusion. Here, we report two structures: a crystal structure of the N-terminal and D1 ATPase domains of murine p97 at 2.9 A resolution, and a cryoelectron microscopy structure of full-length rat p97 at 18 A resolution. Together, these structures show that the D1 and D2 hexamers pack in a tail-to-tail arrangement, and that the N domain is flexible. A comparison with NSF D2 (ATP complex) reveals possible conformational changes induced by ATP hydrolysis. Given the D1 and D2 packing arrangement, we propose a ratchet mechanism for p97 during its ATP hydrolysis cycle.

Oligomeric ring structure of the Bloom's syndrome helicase.

Bloom's syndrome is a recessive human genetic disorder associated with an elevated incidence of many types of cancer. The Bloom's syndrome gene product, BLM, belongs to the RecQ subfamily of DNA helicases and is required for the maintenance of genomic stability in human cells - in particular, the suppression of reciprocal exchanges between sister chromatids. We have investigated the quaternary structure of BLM using a combination of size-exclusion chromatography and electron microscopy with reference-free image processing. We found that BLM forms hexameric ring structures with an overall diameter of approximately 13 nm surrounding a central hole of approximately 3.5 nm diameter. A fourfold symmetric square form with approximately 11 nm sides and a hole of approximately 4 nm diameter was also detected, which might represent a distinct oligomeric species or a side view of the hexameric form. Chromatography studies indicated that the majority of enzymatically active BLM has an apparent molecular mass of > 700 kDa, which is consistent with an oligomeric structure for BLM. This provides the first structural analysis of an oligomeric ring helicase of eukaryotic cellular origin. These results have implications for the mechanism of action of BLM and suggest that other RecQ family helicases, including the WRN protein associated with Werner's syndrome, might also adopt ring structures.

Estimation of the lateral dimensions of cellulose crystallites using 13C NMR signal strengths.

Differences in proton rotating-frame spin relaxation rates were exploited to edit the 13C NMR spectra of solid lignocellulosics, separating signals assigned to cellulose crystallites from signals assigned to amorphous material. Clusters of signals at 89 and 85 ppm were assigned to C-4 in the interiors and on the surfaces of cellulose crystallites, respectively. Relative signal areas were used to estimate the weight-averaged lateral dimensions of crystallites, using a model in which crystallites have approximately square cross sections. The same 10 samples were also characterized by wide-angle X-ray scattering (WAXS). There was a strong correlation (r2 = 0.988) between the two sets of lateral dimensions, but those estimated by WAXS were typically 10% lower than those estimated by NMR. The deviations were attributed to differences in molecular conformations between interior and surface chains, causing broadening of the WAXS peaks. In the case of an eleventh sample containing well-ordered xylan, the NMR and WAXS methods were in good agreement only after exclusion of a xylan signal at 82.6 ppm from the NMR data.

Crystallization of an intact GST-estrogen receptor hormone binding domain fusion protein.

Crystals of an intact GST-estrogen receptor hormone binding domain fusion protein have been grown from solutions of MPD. The crystals grew as clusters of thin plates and needles of maximum dimensions 100 x 20 x 1 micrometer but were unsuitable for X-ray diffraction analysis. However, examination by electron microscopy shows an ordered lattice in which the protein molecules are clearly visible. Image analysis of electron micrographs of the protein crystals revealed electron stain-excluding density which showed a two-domain trimeric structure in projection, with each molecule of dimensions 12.0 x 5.0 nm diameter. The use of GST-fusion proteins in crystallisation are discussed.

Solid-State 13C Nuclear Magnetic Resonance Characterization of Cellulose in the Cell Walls of Arabidopsis thaliana Leaves.

Solid-state 13C nuclear magnetic resonance was used to characterize the molecular ordering of cellulose in a cell-wall preparation containing mostly primary walls obtained from the leaves of Arabidopsis thaliana. Proton and 13C spin relaxation time constants showed that the cellulose was in a crystalline rather than a paracrystalline state or amorphous state. Cellulose chains were distributed between the interiors (40%) and surfaces (60%) of crystallites, which is consistent with crystallite cross-sectional dimensions of about 3 nm. Digital resolution enhancement revealed signals indicative of triclinic and monoclinic crystalline forms of cellulose mixed in similar proportions. Of the five nuclear spin relaxation processes used, proton rotating-frame relaxation provided the clearest distinction between cellulose and other cell-wall components for purposes of editing solid-state 13C nuclear magnetic resonance spectra.

20S human proteasomes bind with a specific orientation to lipid monolayers in vitro.

20S Proteasomes are non-lysosomal, high molecular weight proteinases implicated in the degradation of misfolded proteins and several short-lived regulatory proteins. They have a well established role, as the core of the 26S proteasome complex, in the ubiquitin-dependent proteolytic pathway and in antigen processing. While correctly folded proteins are not degraded by the 20S proteasome, unfolding, for example by oxidation, may render them degradable. The 20S proteasome is a 700-kDa cylindrical particle, composed of 14 subunits of molecular masses 20-35 kDa. There is evidence that 20S proteasomes are in close proximity to or associate with the endoplasmic reticulum and nuclear and plasma membranes in vivo. To better understand the lipid association of 20S proteasomes in vitro, we used a lipid monolayer system as a simple model system for biological membranes. The structure and orientation of the monolayer lipid bound 20S proteasomes has been determined by electron microscopy. 20S proteasomes associated in an "end-on' configuration specifically on PI lipid monolayers forming large arrays, with their channels opposite the lipid headgroups. On ER and Golgi lipid films 20S proteasomes were oriented in the same way as on the PI lipid film but were monodisperse. Protein molecules were randomly oriented in the presence of PA, PG, PS, PC and mitochondrial lipid monolayers. We show that 20S proteasomes bind to phospholipids in vitro in a preferred orientation which places the proteasome channel perpendicular to the membrane.

Separating subspectra from cross-polarization magic-angle spinning nuclear magnetic resonance spectra by proton spin relaxation editing.

Differences in proton spin relaxation time constants can be exploited to edit cross-polarization magic-angle spinning nuclear magnetic resonance (CP-MAS NMR) spectra of heterogeneous mixtures of different types of organic matter. This paper describes an extension of the editing procedure from two-component to three-component mixtures. Clean separation of 13C NMR subspectra was achieved for three synthetic polymers mixed as powders. Applying the procedure to both 13C and 31P CP-MAS NMR spectra of solid dairy pond sludge provided clues to the location of the phosphorus relative to different types of organic matter, and provided estimates of the proportions of organic matter in categories labeled "plant fragments", "partly degraded residues" and "recalcitrant structures". The editing procedure increased noise levels by factors between 2 and 11 in these worked examples, depending on the degree of difficulty involved in distinguishing differences in proton spin relaxation time constants.

Substance P-related antagonists inhibit vasopressin and bombesin but not 5'-3-O-(thio)triphosphate-stimulated inositol phosphate production in Swiss 3T3 cells.

The substance P (SP) analogues [DArg1, DPhe5, DTrp7,9, Leu11] SP (AntD) and [Arg6, DTrp7,9, MePhe8] SP (6-11) (AntG) inhibit the action of many different neuropeptides including SP. These analogues might be useful in the treatment of small cell lung cancer but their mechanism of action is unclear. Here, we analyzed the effect of AntD and AntG on neuropeptide vs. guanosine 5'-3-O-(thio) triphosphate (GTP gamma S)-stimulated inositol phosphate generation in permeabilized Swiss 3T3 cells. AntD inhibited vasopressin and bombesin stimulated inositol phosphate formation (IC50 of 0.75 microM and 2 microM, respectively). Similarly, AntG inhibited vasopressin-stimulated inositol phosphate generation with an IC50 of 1 microM. Strikingly, neither AntD up to 10 microM nor AntG up to 20 microM was able to inhibit GTP gamma S-stimulated inositol phosphate generation. Dose-response curves of neuropeptide-induced inositol phosphate generation were dramatically displaced to the right by either 10 microM AntD or 20 microM AntG. However, neither antagonist affected the dose response of GTP gamma S-stimulated inositol phosphate generation. Furthermore, 20 microM AntD had no effect on AIF-4-induced inositol phosphates in COS-1 cells transfected with G alpha q. AntD inhibited [3H]vasopressin binding competitively in intact Swiss 3T3 cells and both AntD and AntG inhibited [3H]vasopressin binding in Swiss 3T3 and rat liver membranes. Scatchard analysis revealed that AntD inhibited vasopressin binding by reducing receptor affinity without affecting receptor number in both intact and membrane preparations of Swiss 3T3 cells. The results strongly suggest that SP analogues AntD and AntG block the action of the Ca2+ mobilizing neuropeptides at the receptor level, rather than inhibiting G protein-stimulated inositol phosphate production.

Microcrystals of the beta 1 isoenzyme of protein kinase C: an electron microscopic study.

Microcrystals of protein kinase C beta 1 have been grown from solutions of poly(ethylene glycol). Image analysis of electron micrographs of the protein crystals, which diffracted to 5.0 nm, revealed p3 symmetry with a unit cell of about 10.3 nm x 10.3 nm. The electron stain-excluding densities showed a three-domain ring structure in projection, giving kidney bean-shaped molecules of about 7.0 nm x 4.5 nm diameter, packed as trimers. The implications of these observations for the function of the enzyme are discussed.

A model of the structure of human annexin VI bound to lipid monolayers.

Annexin VI is an eight repeat member of the annexin family of proteins which are both water soluble and bind to negatively charged phospholipids in a calcium-dependent manner. Here we present a model for annexin VI based on fitting the three-dimensional structure of two annexin V molecules (Huber (1990) EMBO J. 9, 3867-3874) to the two-dimensional stain-excluding density of lipid-bound annexin VI (Newman (1989) J. Mol. Biol. 206, 213-219). Both annexin VI lobes could only be fitted with their convex faces closest to the lipid monolayer. This supports the hypothesis that annexin-lipid binding is mediated by the interaction between calcium bound to the loops protruding from the convex protein surface and phospholipid headgroups.

Amino acid sequence analysis of the annexin super-gene family of proteins.

The annexins are a widespread family of calcium-dependent membrane-binding proteins. No common function has been identified for the family and, until recently, no crystallographic data existed for an annexin. In this paper we draw together 22 available annexin sequences consisting of 88 similar repeat units, and apply the techniques of multiple sequence alignment, pattern matching, secondary structure prediction and conservation analysis to the characterisation of the molecules. The analysis clearly shows that the repeats cluster into four distinct families and that greatest variation occurs within the repeat 3 units. Multiple alignment of the 88 repeats shows amino acids with conserved physicochemical properties at 22 positions, with only Gly at position 23 being absolutely conserved in all repeats. Secondary structure prediction techniques identify five conserved helices in each repeat unit and patterns of conserved hydrophobic amino acids are consistent with one face of a helix packing against the protein core in predicted helices a, c, d, e. Helix b is generally hydrophobic in all repeats, but contains a striking pattern of repeat-specific residue conservation at position 31, with Arg in repeats 4 and Glu in repeats 2, but unconserved amino acids in repeats 1 and 3. This suggests repeats 2 and 4 may interact via a buried saltbridge. The loop between predicted helices a and b of repeat 3 shows features distinct from the equivalent loop in repeats 1, 2 and 4, suggesting an important structural and/or functional role for this region. No compelling evidence emerges from this study for uteroglobin and the annexins sharing similar tertiary structures, or for uteroglobin representing a derivative of a primordial one-repeat structure that underwent duplication to give the present day annexins. The analyses performed in this paper are re-evaluated in the Appendix, in the light of the recently published X-ray structure for human annexin V. The structure confirms most of the predictions and shows the power of techniques for the determination of tertiary structural information from the amino acid sequences of an aligned protein family.

The virulence of British isolates of bovid herpesvirus 1 in relationship to viral genotype.

Six strains of bovid herpesvirus 1 isolated from British cases of infectious bovine rhinotracheitis (IBR) were inoculated intranasally into calves. The clinical, virological and serological responses were measured and comparisons made between virus strains. Calves infected with viruses of subtype 1 developed more severe clinical signs and excreted higher titres of virus than calves infected with subtype 2b strains. The findings could be related to the history of IBR in Great Britain.

2D crystal forms of annexin IV on lipid monolayers.

Two-dimensional crystalline arrays of annexin IV were generated by interaction of the purified protein with a phospholipid monolayer. Image analysis of electron micrographs of the protein crystals, which diffracted to 3.5 nm respectively, revealed p6 and p3 symmetry. Annexin IV gave two crystal forms with unit cells of 18 x 18 nm and 28 x 28 nm. The former unit cell was similar to a previously described form of annexin VI. The implications of these observations are discussed.

Two-dimensional crystallization of proteins on lipid monolayers.

This review includes details of recent macromolecular crystallizations using lipid monolayers. Crystallization conditions are discussed together with suggestions for improving resolution.

Bedside blood sugar determinations in the critically ill.

Critically ill patients often have dramatic fluctuations in blood glucose level, with dire physiologic consequences. Although urine glucose testing can be used to assess hypoglycemia and hyperglycemia, this technique is often inaccurate. Therefore, a study was conducted to determine the accuracy and cost-effectiveness of various glucose monitoring systems. The Glucoscan 2000 was compared with the standard laboratory method in 41 patients, by use of 110 blood glucose determinations. There were no significant differences between the two methods in accuracy or time required to run the test. The laboratory charges for 110 determinations were $990, whereas the Glucoscan 2000 cost for the same number of tests was $55. The Glucoscan 2000 provides an accurate, timely, and cost-effective method of monitoring blood glucose.

Detection of antibodies to bovid herpesvirus 4 by ELISA.

An enzyme linked immunosorbent assay for antibodies to bovid herpesvirus 4 was developed using antigen prepared by detergent lysis of infected cell cultures. The assay was used to study the immune responses of experimentally-immunised calves. The results correlated well with the indirect fluorescent antibody method. A viral neutralizing antibody response could not be demonstrated in the calves.