Richard H. Crepeau

Three-dimensional reconstruction of the 14-filament fibers of hemoglobin S.

Abstract The supramolocular structure of hemoglobin S has been studied by electron microscopy and computer-based image reconstruction. Negatively stained fibers prepared by the lysis of sickled cells or the stirring of hemoglobin S hemolysates have been observed to be almost exclusively of the 20-nm diameter form. These fibers have a periodic variation in diameter between the extremes of 18 nm and 23 nm. Computed Fourier transforms of the fibers show a, highly complex pattern of reciprocal space maxima, with 30 maxima on 20 layer-lines clearly resolved. The Bessel orders of the maxima were assigned with the aid of a newly developed technique, a combined real-space Fourier-space reconstruction method (REFORM). This method utilizes the filtered image produced by the inverse Fourier transform of the low-resolution maxima to calculate in real space the crosssection of a helical fiber. The REFORM analysis indicated that the fibers have an elliptical cross-section and are composed of 14 hexagonally packed filaments with 10 outer filaments surrounding four inner filaments. On the basis of this cross-section, the Bessel orders of all the maxima were assigned, permitting the calculation of three-dimensional reconstructions by Fourier Bessel synthesis. From these reconstructions details of the location of hemoglobin S molecules of each filament were obtained. Hemoglobin S molecules are staggered in adjacent filaments to produce a closely packed helical structure. Reconstructions of fibers at various stages of disassembly revealed a stable intermediate containing 10 filaments which could be characterized in terms of the loss of two pairs of specific outer filaments. A partially disassembled fiber with only six filaments at positions corresponding to three inner and three outer filaments of the parent structure was also identified. The six-filament structure appears to be produced from the 10-filament structure by the loss of two specific pairs of filaments. Thus pairs of filaments are evidently significant structural units in the stabilization of the complete fibers and the orientation of the molecules in these pairs may be related to the filament pairs known to occur in crystals of hemoglobin S.

Structural studies on porcine brain tubulin in extended sheets

Abstract Structural studies have been conducted on porcine brain tubulin, assembled into microtubule-related structures, by electron microscopy in conjunction with optical diffraction and image reconstruction techniques. By minimizing background noise and sample damage, we have improved the resolution on negatively stained samples, extending the data from the previous limit of a 42 A layer line to an additional layer line at 21 A. The new reflections confirm the basic surface lattice proposed from the earlier studies and extend the structural features that can be assigned to individual tubulin molecules. Data are obtained for microtubules in standard buffers for both the helical form and flat sheets of up to 13 protofilaments. When zinc is added to the preparations, sheets with more than 13 protofilaments are formed and the extended lattices provide more reflections on both the 42 A and 21 A layer lines, as well as the equator. The lattice in the presence of zinc differs considerably from the normal lattice, with adjacent protofilaments staggered by 21 A, compared to the staggering of adjacent filaments of about 10 A in the absence of zinc. There is also a distinct pairing of adjacent protofilaments in the zinc-induced sheets. Initial studies with the Unwin-Henderson method on unstained zinc-tubulin sheets suggest that the adjacent protofilaments may be related by a dyad axis, either perpendicular or parallel to the protofilament axes.

Three-dimensional reconstruction of tubulin in zinc-induced sheets.

The three-dimensional structure of porcine brain tubulin in planar sheets formed in the presence of zinc has been determined to a resolution of approximately 20 A by electron microscopy and image reconstruction on negatively stained samples. The samples were prepared with a mica floatation technique, which yields tubulin sheets with 36 reciprocal space maxima on lattice lines at 21, 28, 42 and 84 A−1 in Fourier transforms of digitized images. In order to obtain three-dimensional data, sheets were tilted with the goniometer stage of the electron microscope to provide images at various angles between 0 ° and ± 60 °. Transforms of 33 tilted images plus the transform of untilted sheets based on an average of nine untilted images were combined to give the third dimension of reciprocal space (z∗). These data, were expressed in terms of the phases and amplitudes along the z∗ lattice line for each of the 36 maxima observed in untilted samples, as well as five additional lattice lines which have zero-amplitudes in the non-tilted central section of the three-dimensional transform. Home of these zero-amplitudes arise from systematic absences which are due to a 2-fold screw axis relating adjacent protofilaments of tubulin in the zinc-induced sheets. Thus in the three-dimensional reconstructions of the sheets a polarity of the protofilaments is apparent, with adjacent protofilaments aligned in opposite directions to give an antiparallel pattern, in contrast to normal microtubules composed of protofilaments in parallel alignment. Two classes of morphological units, each with a mass corresponding to a molecular weight of about 55,000, are found to alternate along the protofilaments. These distinct morphological units are identified as the α and β subunits of tubulin, confirming the representation of tubulin as an αβ heterodimer. Furthermore, the extensive internal contact between subunits within a dimer can readily be distinguished from the less extensive contact between dimer units. Such differences in contacts were not apparent in the earlier two-dimensional reconstructions. In addition, areas of excluded stain joining one class of subunits to the subunits of the other class in adjacent protofilaments have been resolved for tubulin polymerized in zinc-induced sheets. Of the two classes of subunits one is distinguished by a prominent cleft. Identification of which class of subunits is α and which is β is not yet possible.

Reconstruction of imperfectly ordered zinc-induced tubulin sheets using cross-correlation and real space averaging

Large negatively-stained two-dimensional tubulin arrays induced by incubation with zinc were examined using electron microscopy. When Fourier methods are applied for the reconstruction, the resolution depends critically on the order of these arrays. The real space cross-correlation method as described in this paper, however, can be satisfactorily applied to disordered arrays. The method involves selection of a small area which is used as a reference to locate correlation peaks in other regions. In this way the locations of various unit cells are determined, often at points which deviate from the exact lattice. The unit cells at their determined locations are averaged, and the average is used as an improved reference in another search for correlation peaks. The process can be iterated to refine the reference and improve the correlation mapping. If necessary, poorly correlating unit cells can be ignored in the averaging, and rotational as well as translational corrections can be considered; but neither of these steps were found to be necessary in the studies on the zinc-induced tubulin sheets reported here. When both correlation and Fourier methods were compared for the same areas of selected sheets, the two methods gave comparable results for relatively well ordered regions, but the correlation images were superior for the more poorly ordered regions. In addition, even for the relatively well ordered regions the correlation images revealed some details that were lacking in the simple Fourier reconstructions but did appear in more extensive studies involving three-dimensional reconstruction. A direct comparison of the efficacy of real space and Fourier space reconstruction methods is presented using disordered model data.

A theoretical model of phospholipid dynamics in membranes

Static and dynamic properties related to the internal configurational motions have been calculated for the alkyl chains of phospholipid molecules in a membrane environment in the liquid crystal phase. The calculations have been performed for the chain 1 of 1,2‐dipalmitoyl 3‐sn‐phosphatidylcholine (DPPC), a typical constituent of phospholipid membranes. Under the assumption of fixed bond lengths and bond angles, the internal dynamics of the chain is described in terms of 15 dihedral angles. The time evolution of the angular variables is assumed to be diffusional in character, and a master equation for transitions among the stable conformers is constructed from the energetics and hydrodynamics of the chain. This method is an extension to the time domain of the rotational isomeric state (RIS) approximation, which has been widely used to compute static properties of the chains. After calculation of the suitable correlation functions, effective rate constants relevant for spectroscopic and kinetic observables ...

Two-dimensional Fourier transform ESR in the slow-motional and rigid limits: 2D-ELDOR☆

Abstract The two-dimensional Fourier transform ESP techniques of stimulated SECSY and 2D-ELDOR are shown to be powerful methods for the study of slow motions for nitroxides. Stimulated SECSY provides magnetization transfer rates, whereas 2D-ELDOR displays how the rotational motions spread the spins out from their initial spectral positions to new spectral positions, as a function of mixing time. The role of nuclear modulation in studies of structure and dynamics is also considered.

Studies of spin relaxation and molecular dynamics in liquid crystals by two‐dimensional Fourier transform electron spin resonance. II. Perdeuterated‐tempone in butoxy benzylidene octylaniline and dynamic cage effects

Two-dimensional Fourier transform~2D-FT!-electron spin resonance ~ESR! studies on the small globular spin probe perdeuterated tempone ~PDT! in the liquid crystal solvent 4O,8~butoxy benzylidene octylaniline! are reported. These experiments, over the temperature range of 95 °C to 24 °C, cover the isotropic ( I ), nematic (N), smecticA (SA), smecticB (SB), and crystal ( C) phases. The 2D-ELDOR~two-dimensional electron–electron double resonance ! sp ctra confirm the anomalously rapid reorientation of PDT, especially in the lower temperature phases. The model of a slowly relaxing local structure~SRLS! leads to generally very good non-linear least squares ~NLLS! global fits to the sets of 2D-ELDOR spectra obtained at each temperature. These fits are significantly better than those achieved by the standard model of Brownian reorientation in a macroscopic orienting potential. The SRLS model is able to account for anomalies first observed in an earlier 2D-ELDOR study on PDT in a different liquid crystal in its smectic phases. Although it is instructional to extract the various spectral densities from the COSY ~correlation spectroscopy! and 2D-ELDOR spectra, the use of NLLS global fitting to a full set of 2D-ELDOR spectra is shown to be more reliable and convenient for obtaining optimum model parameters, especially in view of possible~incipient! slow motional effects from the SRLS or dynamic cage. The cage potential is found to remain fairly constant at about kBT over the various phases ~with only a small drop in the SB phase!, but its asymmetry increases with decreasing temperature T. This value is significantly larger than the weak macroscopic orienting potential which increases from 0.1 to 0.3k BT with decreasingT. The cage relaxation rate, given by R is about 3 10 s in the I phase, but increases to about 10 8 s in theSA , SB , andC phases. The rotational diffusion tensor for PDT shows only a small T-independent asymmetry, and its mean rotational diffusion coefficient is of order 10 s, with however, a small increase in the SB phase. These results are consistent with a model previously proposed for PDT in benzylidene liquid crystal solvents, that as T is reduced the PDT molecules are partially expelled from the hard core ~dipolar! region of the liquid crystalline molecules toward the more flexible aliphatic chain region as a result of increased core packing from smectic layer formation, and it thus experiences a more fluid ~for a given temperature! local cage structure. ©1996 American Institute of Physics. @S0021-9606~96!51333-4#

Analytical ultracentrifugation with absorption optics and a scanner-computer system.

Abstract A computer system has been designed which measures directly individual photomultiplier pulses of the Beckman model E analytical ultracentrifuge and converts digitized values for the pulses into an optical density (OD) as a function of radius. The system consists of a Nova computer with 8K memory, a 12 bit analog/digital converter, a Sykes cassette tape recorder for long-term storage, and a logic control and multiplexing unit which triggers the A/D converter and provides cell and sector identification for use with multihole rotors. The system can, in two revolutions of the rotor, measure the sample and reference pulses of up to five cells as well as the baseline or dard value of the photomultiplier. The radial position of the collected data is determined by counting the number of revolutions of the rotor to interpolate the distance between the outside and inside counterbalance holes. In actual operation, data are collected for 50 or 100 revolutions, sorted in terms of rotor hole and sample or reference sector, and converted into an average optical density for each cell. This single data point for each cell is then stored and the procedure is repeated for the next 100 revolutions. Calculation of OD within the computer eliminates the possibility of systematic or random error as may be present in the scanner log amplifier, holding circuits, and associated electronics. The routine for averaging the data from 100 revolutions also calculates the standard deviation for each final OD value which may be used in any analysis program for weighting the data. After the scan, the data in the computer can be stored on magnetic tape, typed out for examination, or analyzed with various fitting routines. Present developments with the system include application to high-speed and low-speed sedimentation equilibrium. The results of a sedimentation equilibrium run with myoglobin are presented to illustrate the operation and performance of the system. The useful range of OD measurements currently available with the computer system is from 0.005 to 1.6 OD.

Fourier transform electron spin resonance imaging

Abstract Modern Fourier transform (FT) ESR methods have been combined with fast, high power pulsed magnetic field gradients to enable FT-ESR imaging. Spectral—spatial imaging by frequency and phase encoded FT methods are compared with cw methods. The initial phase encoded results are comparable in quality to those from the well-developed cw methods and further improvements which would enhance FT-ESR imaging are noted.

Composite pulses in time-domain ESR

Recently, new and important Fourier transform (1-4) and two-dimensional Fourier transform ESR (I, 3,5) techniques have been developed which have the promise of revolutionizing ESR as they have NMR. The techniques have been described in detail and have been successfully applied to the class of nitroxide ESR spectra (I, 3, 6). A fundamental problem is the competing requirements of ( 1) delivering a large enough rotating microwave B, field at the sample to irradiate the whole spectrum (i.e., 2y,B1 > Aw,, where Aw, is the spectral extent) and at the same time (2) having a resonator with a low enough QL (loaded Q) that the short pulse of width tP required for a 7r/2 (or r) rotation ofthe spins (i.e., yeBltp = 7r/2) is not significantly distorted. That is, we require that the half-power full bandwidth of the resonator, Aw = w / QL, obey Aw 3 Aw,. The problem, of course, is that for a given power, PO incident on the resonator, B, and QL are intimately related ( 7) by PO = p. V,B: u/ 2QL, where v is the resonator frequency and V, is the effective volume. From the above comments we can then state that the power required to irradiate the spectrum obeys PO a Awi (3). Gorcester and Freed (GF) (3) found for their 2D FT ESR spectrometer with a cavity resonator and TWT output of 1 kW that the optimum bandwidth was achieved with a B, of 6 G ( tP = 14 ns for a ?r/2 pulse) and aQL100 for which Aw/27r = 95 MHz, corresponding to uniform spectral rotation over a bandwidth of about 34 MHz ( =yeB, / ?r ). Nevertheless, they succeeded in obtaining wideband excitation for a motionally narrowed nitroxide with spectral extent, Aw, = 90 MHz. However, the single-pulse FID leads to an FT spectrum with outer hyperIine lines of reduced intensity (i.e., 50-60% of the central line). In a threepulse 2D FT spectrum (i.e., 2D ELDOR), this attenuation of the outer peaks goes approximately as the cube of the attenuation of the simple FT spectral line. While appropriate corrections may be made for this effect in the analysis of the experiment, it does lead to reduced sensitivity. An even more serious problem arises in attempting to perform IT experiments on slow-motional nitroxide spectra. These have a spectral extent of Aw,/2* x 220 MHz