The assembly of amyloidogenic yeast sup35 as assessed by scanning (atomic) force microscopy: an analogy to linear colloidal aggregation?

Amyloidosis is a class of diseases caused by protein aggregation and deposition in various tissues and organs. In this paper, a yeast amyloid-forming protein Sup35 was used as a model for understanding amyloid fiber formation. The dynamics of amyloid formation by Sup35 were studied with scanning force microscopy. We found that: 1) the assembly of Sup35 fibers begins with individual NM peptides that aggregate to form large beads or nucleation units which, in turn, form dimers, trimers, tetramers and longer linear assemblies appearing as a string of beads; 2) the morphology of the linear assemblies differ; and 3) fiber assembly suggests an analogy to the aggregation of colloidal particles. A dipole assembly model is proposed based on this analogy that will allow further experimental testing.

Nucleated conformational conversion and the replication of conformational information by a prion determinant.

Prion proteins can serve as genetic elements by adopting distinct physical and functional states that are self-perpetuating and heritable. The critical region of one prion protein, Sup35, is initially unstructured in solution and then forms self-seeded amyloid fibers. We examined in vitro the mechanism by which this state is attained and replicated. Structurally fluid oligomeric complexes appear to be crucial intermediates in de novo amyloid nucleus formation. Rapid assembly ensues when these complexes conformationally convert upon association with nuclei. This model for replicating protein-based genetic information, nucleated conformational conversion, may be applicable to other protein assembly processes.

Scanning (atomic) force microscopy imaging of earthworm haemoglobin calibrated with spherical colloidal gold particles.

Scanning (atomic) force microscopy (SFM) permits high-resolution imaging of a biological specimen in physiological solutions. Untreated extracellular haemoglobin molecules of the common North American earthworm. Lumbricus terrestris, were imaged in NH4Ac solution using calibrated SFM. Individual molecules and their top and side views were clearly identified and were comparable with the images of the same molecule obtained by scanning transmission electron microscopy (STEM). A central depression, the presumed mouth of the hole, was detected. We analysed 75 individual molecules for their lateral dimensions. Compression varied for different molecules, presumably because of the variation of the interaction between the SFM tip and the protein molecule. Two effective heights which correspond to the heights of the points of the haemoglobin molecules first and last touched by the tip, h1 and h2, respectively, were measured for each protein and ranged between 1.58 and 16.2 nm for h1 and 1.23 and 13.6 nm for h2. The apparent diameter was measured and ranged from 44.9 to 86.6 nm (63.2 +/- 10.5 nm, n = 75), which is about twice the diameter of the molecule reported by STEM for the top view orientation. The higher the measured effective heights, the worse was the tip convolution effect. In order to determine the tip parameters (semivertical angle, curvature of radius and the cut-off height) and to calibrate images of earthworm haemoglobin molecules, spherical gold particles were scanned as standards. The tip sectional radii at distances of h1 and h2 above the tip apex were subtracted from the apparent diameter of the protein. The calibrated lateral dimension was 29.1 +/- 3.85 nm, which is close to the reported scanning transmission electron microscopy data 30.0 +/- 0.8 nm. The results presented here demonstrate that the calibration approach of imaging gold particles is practical and relatively accurate. Calibrated SFM imaging can be applied to the study of other biomacromolecules.

Flecainide and the electrophysiologic matrix: the effects of flecainide acetate on the determinants of cardiac excitability in sheep Purkinje fibers.

Flecainide was associated with excess mortality distributed virtually equally throughout the period of the Cardiac Arrhythmia Suppression Trial, suggesting the intersection of two events, drug effect and perhaps ischemia. Flecainide's effect on active properties has been studied extensively, but nothing is known of its effects on passive properties or on the balance among active and passive cellular properties that determines cardiac excitability. The multiple microelectrode method of intracellular current application and transmembrane voltage recording was used in sheep Purkinje fibers to determines strength- and charge-duration as well as constant current-voltage relationships and to estimate active properties, liminal length, and cable properties at a normal [K+]o and in a setting of hyperkalemia analogous to that of ischemia. A computer tracked in time the alterations in the active and passive properties relevant to excitability. Flecainide slightly decreased excitability at a normal [K+]o, primarily by depressing the sodium system with some contributory effect of passive properties. At high [K+]o, flecainide caused a frequency-dependent decrease in excitability and conduction, the latter best interpreted as a failure of the fiber to attain the liminal length requirements to produce a local action potential due primarily to an effect on sodium conductance. Together, the observations suggest that the action potential is the local phenomenon and that the propagated event is the sequential fulfillment of liminal length requirements. The data were interpreted in terms of the electrophysiologic matrix first proposed in detail in this Journal, which indicated that the electrophysiologic universe moved as a system in response to the drug and a change in [K+]o, the presumed antiarrhythmic and proarrhythmic electrophysiologic matrices for flecainide were quite similar, and the matrical configuration shared characteristics with the matrices of other drugs with known proarrhythmic potential.

Atomic (scanning) force microscopy in cardiovascular research.

The promise of atomic (scanning) force microscopy (AFM) for cardiovascular research is enormous. The AFM images by using a sharp cantilever tip to sense the repulsive and attractive forces between the tip and the sample surface. The force of interaction is kept constant while raster scanning, resulting in images of the surface contours with molecular and, on hard inorganic surfaces, even atomic resolution. Movement of the cantilever in the Z plane is detected by a laser beam reflected off the cantilever to a photodiode system, a piezotube allows an X and Y raster, and a three-dimensional image results. Its capabilities include: (1) the three-dimensional imaging of membranes and biomolecules with molecular and submolecular resolution; (2) such imaging not only of dry specimens but of specimens in a physiologic solution, thereby allowing the investigation of dynamic processes in both viable biomolecules and living cells; (3) the sensing of charge and intermolecular interaction forces; (4) the chemical or biochemical modification of the cantilever tip, which allows the identification of specific structures and the measurement of specific interactions (e.g., a ligand-receptor interaction); (5) nanometer control of the position and force of the cantilever, which, in turn, allows the physical manipulation of biomolecules, the dissection of biological structures (e.g., the separation of one gap junctional hemichannel from its neighbor, thereby revealing normally inaccessible surfaces), the delivery of ligands, drugs, or other materials to specific locations, and the precise measurement of interacting forces at specific sites; and (6) the modification of the apparatus by adding complementary methodologies (e.g., magnetic resonance imaging, fluorescence microscopy, confocal microscopy, and perhaps electrophysiology). AFM, however, is only now being applied to biological research, many technical and methodologic problems exist, and a number of them are considered in this review. Little work has been done in cardiovascular research and the purpose of this review is to introduce this new and exciting approach to investigation.

Electrostatic force microscope for probing surface charges in aqueous solutions.

A scanning force microscope was converted to an electrostatic force microscope by charging the usually neutral cantilever with phospholipids. The electrostatic force microscope was used to study surface electrostatic charges of samples in aqueous solutions. Lysozymes, DEAE-Sephadex beads, 3-propyltriethoxysilane-treated glass and mica were imaged in water or phosphate buffer with electrostatic force microscopy. The adhesion force measured when a charged probe and oppositely charged specimen interacted was up to 500 times greater than when a bare probe was used. This dramatic increase in measured adhesion force can be attributed to the energy required to break the salt bridges formed between the charged probe and the specimen. The use of phospholipids to functionalize the cantilever tip allows the incorporation of other biomolecules and ligands that can be used as biologically specific tips (e.g., receptors, drugs) for the study of intermolecular interactions.

Heart gap junction preparations reveal hemiplaques by atomic force microscopy.

Current structural models of gap junctions indicate two apposed plasma membranes with hexagonally packed hemichannels in each membrane aligning end to end. These channels connect the cytoplasms of contacting cells. Images of isolated rat heart gap junctions have been made with the atomic force microscope in aqueous media. We show that native cardiac gap junctions have a thickness of 25 +/- 0.6 nm. This decreases to 17 nm when they are treated with trypsin, which is known to remove some cytoplasmic components of connexin 43. Imaging shows subunits with a center to center spacing of approximately 9-10 nm and long range hexagonal packing, measurements in agreement with studies using freeze-fracture and negative-stain electron microscopy. In addition to gap junctions, we imaged structures that had all the characteristics of native gap junctions except their thickness was limited to 9-11 nm. They also show long range hexagonal packing and center to center spacing of 9-10 nm. These structures decrease in thickness, to 6-9 nm, when treated with trypsin. We have called these structures hemiplaques. They appear to be present endogenously in the preparation, as we have ruled out their being an artifact of imaging by AFM. However, it remains to be determined if they are a consequence of the procedure used in isolating gap junctions or a possible intermediary in gap junction formation.

The application of atomic force microscopy for the detection of microcrystals in synovial fluid from patients with recurrent synovitis.

Synovial fluid from 33 patients with inflammatory arthritis was examined with a polarized light microscope (PLM) and an atomic force microscope (AFM). Two samples were imaged with a transmission electron microscope (TEM) to determine calcium/phosphate ratios and identify microcrystals of calcium pyrophosphate dihydrate and octacalcium phosphate. Additional correlative x-ray diffraction studies were performed on several samples including purified hydroxyapatite and sodium chloride crystals. Monosodium urate, calcium pyrophosphate dihydrate, hydroxyapatite, octacalcium phosphate, and cholesterol crystals were identified with AFM. AFM images of these microcrystals revealed detailed surface topology, including lattice parameters and structural irregularities at the crystals' surface. These features were consistent with those obtained by TEM and x-ray diffraction studies. In addition, AFM images revealed that some specimens contained microcrystals that were undetected by PLM and/or TEM. These results suggest that AFM may provide a simple yet powerful technique for the detection of microcrystals in synovial fluid taken from patients with crystal-induced arthritis.

Calibration of the scanning (atomic) force microscope with gold particles.

Scanning force microscopy (SFM) holds great promise for biological research. Two major problems that have confronted imaging with the scanning force microscope have been the distortion of the image and overestimation in measurements of lateral size due to the varying geometry and characteristics of the scanning tip. In this study, spherical colloidal gold particles (10, 20 and 40 nm in diameter) were used to determine (1) tip parameters (size, shape and semi-vertical angle); (2) the distortion of the image caused by the tip; and (3) the overestimation or broadening of lateral dimensions. These gold particles deviate little in size, are rigid and have a size similar to biological macromolecules. Images of the colloidal gold particles by SFM were compared with those obtained by electron microscopy (EM). The height of the gold particles as measured by SFM and EM was comparable and was little affected by the tip geometry. The measurements of the lateral dimensions of colloidal gold, however, showed substantial differences between SFM and EM in that SFM resulted in an overestimate of the lateral dimensions. Moreover, the distortion of images and broadening of lateral dimensions were specific to the SFM tip used. The calibration of the SFM tip with mica provided little clue as to the type of distortion and the amount of lateral broadening observed when the larger gold particles were scanned. The SFM image also depended on the orientation of the tip with respect to the specimen. Our results suggest that quantitative SFM imaging requires calibration to identify and account for both the distortions and the magnitude of lateral broadening caused by the cantilever tip. Calibration with gold particles is fast and nondestructive to the tip. The raw imaging data of the specimen can be corrected for the tip effect and true structural information can be derived. In summary, we present a simple and practical method for the calibration of the SFM tip using gold particles with a size in the range of biomacromolecules that allows: (1) selection of a cantilever tip that produces an image with minimal distortion; (2) quantitative determination of tip parameters; (3) reconstruction of the shape of the tip at different heights from the tip apex; (4) appreciation of the type of distortion that may be introduced by a specific tip and quantification of the overestimation of the lateral dimensions; and (5) calculation of the true structure of the specimen from the image data. The significance is that such calibration will permit quantitative and accurate imaging with SFM.

Imaging of reconstituted biological channels at molecular resolution by atomic force microscopy.

Using atomic force microscopy (AFM), we obtained high-resolution surface images of the bacterial outer membrane channels Escherichia coli OmpF porin and Bordetella pertussis porin that were reconstituted in artificial bilayer membranes as two-dimensional crystalline arrays. These porins were chosen because they are among the most extensively studied proteins of this type and are known for their well-defined crystalline nature in the native membrane. Such reconstituted membrane proteins are ideal specimens to assess the suitability and resolution of AFM for imaging biomembranes and associated proteins. Although OmpF porin often showed a mixed pattern of rectangular and hexagonal arrays with approximately 8.4 x 9.8- and approximately 7.2-nm-spacings, respectively, B. pertussis porin showed mostly a rectangular pattern with an approximately 7.9 x 13.8-nm spacing. The packing patterns of the E. coli OmpF porin in the membrane are very close to those found in electron-microscopic studies. When B. pertussis porin was imaged in a buffer solution, its trimeric subunits were apparently resolved, and the surface of each monomer revealed beadlike structures. This is the first report of such a high-resolution structural analysis of B. pertussis porin by any imaging method. We also imaged the lipid bilayer itself as an internal control for imaging and to further ascertain the resolution. Individual polar head groups of bilayer lipid molecules were resolved, suggesting the intrinsic resolution of AFM for bioimaging.

Voltage-dependent gating and single-channel conductance of adult mammalian atrial gap junctions.

In the heart, the rapid propagation and synchronization of action potentials necessary for a normal heart rhythm and an effective cardiac output are mediated by specialized ionic channels that link adjacent cells and are known collectively as gap junctions. Cardiac gap junctions are gated by various physiological and pharmacological agents, but the role of voltage in their gating is unclear. Whereas embryonic or neonatal ventricular cells have voltage-gated gap junctions, adult cells are reported to have only voltage-independent gap junctions. We studied the voltage dependence of adult rat atrial gap junctions by individually voltage clamping each cell of a connected cell pair and controlling the transjunctional voltage (Vj), measuring transjunctional current (Ij), and calculating junctional conductance (gj). Two distinct populations of cell pairs were observed: highly coupled pairs with the peak gjs ranging from 3.4 to 40 nS and weakly coupled pairs with the peak gjs ranging from 0.3 to 2.0 nS. gj was dependent on Vj, and Ij decayed exponentially, with the time constants being voltage dependent. Voltage dependence was most apparent when cells were poorly coupled. The gj did not decrease to zero. The normalized conductance--Vj plot was fit with a two-state Boltzmann model as a first approximation, resulting in a half-inactivation potential and gating charge of 42.5 mV and 1.14 eV, respectively, for the weakly coupled cell pairs. For highly coupled cell pairs, the half-inactivation potential shifted to 53.3 mV. Single gap junctional channels had a gj of 36.2 +/- 7.6 pS (range, 27-49 pS), which was Vj independent.(ABSTRACT TRUNCATED AT 250 WORDS)

Atomic force microscopy and dissection of gap junctions.

An atomic force microscope (AFM) was used to study the structure of isolated hepatic gap junctions in phosphate-buffered saline (PBS). The thickness of these gap junctions appears to be 14.4 nanometers, close to the dimensions reported by electron microscopy (EM). When an increasing force is applied to the microscope tip, the top membrane of the gap junction can be "dissected" away, leaving the extracellular domains of the bottom membrane exposed. When such "force dissection" is performed on samples both trypsinized and fixed with glutaraldehyde, the hexagonal array of gap junction hemichannels is revealed, with a center-to-center spacing of 9.1 nanometers.

Cardiac excitability and antiarrhythmic drugs: a different perspective.

A matrix of active and passive cellular properties determines net cardiac excitability. The hypothesis of altered excitability suggests that for cardiac arrhythmias to arise, the normal matrix must be perturbed by arrhythmogenic influences to produce a proarrhythmic matrical configuration to permit rhythm disturbances caused by abnormalities of propagation, abnormal automaticity, or altered excitability. Antiarrhythmic drugs may act with one or more components of the normal or proarrhythmic matrix to normalize or to create new antiarrhythmic or, perhaps, proarrhythmic matrices. Traditionally, antiarrhythmic drug classifications have been based on predominant drug actions. These classifications have clinical and some experimental utility but fail to consider the complicated effects that pathophysiologic influences and pharmacologic actions may have on active and passive cellular properties. Cluster analysis may allow the development of new classifications of arrhythmogenesis and antiarrhythmic drugs. The matrical concept has important clinical implications and suggest strategies for treating patients with cardiac rhythm disturbances.

Effects of ethmozin on excitability in sheep Purkinje fibers: the balance among active and passive cellular properties which comprise the electrophysiologic matrix.

Ethmozin is a phenothiazine derivative that is effective against supraventricular and ventricular arrhythmias. Studies to date have examined ethmozin's effects on active cellular properties and automaticity, but nothing is known of its effects on passive properties or on the interrelationships among the several active and passive properties that are of particular relevance to cardiac excitability. The hypothesis tested in this study was that ethmozin, in concentrations equivalent to clinically effective antiarrhythmic levels, would simultaneously affect passive and active cellular properties so as to produce a net decrease in cardiac excitability. The multiple micro-electrode method of intracellular constant current application and trans-membrane voltage recording was used in sheep Purkinje fibers to determine strength-duration and constant current-voltage relationships as well as cable properties. A rapid, on-line computerized data analysis system tracked in time the alterations in the active and passive properties relevant to excitability. Ethmozin, at concentrations of 1.1 and 2.2 microM (0.5 and 1.0 mg/l), decreased cardiac excitability as manifested by an increase in the current required to attain threshold and/or an upward shift in strength- and charge-duration relationships, by depressing the sodium system (decreased maximal rate of rise of phase 0 of the action potential, voltage threshold and overshoot), by decreasing slope resistance and altering nonlinearities of the current-voltage relationships in the subthreshold potential range, by decreasing membrane resistance and by affecting other properties dependent on membrane resistance which would depress excitability. The data for ethmozin and other antiarrhythmic drugs are interpreted in terms of the recently proposed electrophysiologic matrix which we believe has important advantages over traditional hierarchical classifications.

Management of arrhythmias in heart failure.

The literature for coronary artery disease as well as ischemic and dilated cardiomyopathy suggests that ventricular arrhythmias and left ventricular dysfunction are independent risk factors for sudden death, but that the presence of organic heart disease provides the substrate for potentially lethal arrhythmias. Patients with a cardiomyopathy and ventricular tachycardia are at a high risk for sudden death as a group. The general risk, then, is high for the group with CHF and arrhythmias. The prognostic indices for hypertrophic cardiomyopathy are imprecise, but the risk for sudden death for the group is high in the young and remains high even among the adult survivors. Many conditions associated with CHF and its treatment may lead to arrhythmias and are potentially reversible. Most studies suggest that EPS and exercise provocation have limited power in predicting the risk to the individual patient. Therapeutically, reversible causes of arrhythmias should be sought and corrected. In general, antiarrhythmic drug therapy has been disappointing with adequate control being achieved in only about 30 per cent of patients and uncertainties about the effectiveness of such therapy in altering long-term prognosis. This is due to various causes including the inability to find an effective drug, problems with patient compliance, the failure of physicians to properly monitor drug levels, and changes in the anatomical and physiologic substrate due to disease and therapy. Surgical ablation or resection of arrhythmogenic foci is effective in selected patients. The AICD will become first-line therapy in patients at high risk for sudden death due to ventricular arrhythmias, with antiarrhythmic drugs and other approaches being used to minimize the frequency of the arrhythmias.

Effects of quinidine sulfate on the balance among active and passive cellular properties that comprise the electrophysiologic matrix and determine excitability in sheep Purkinje fibers.

Quinidine is the most commonly used drug for the chronic treatment of ventricular arrhythmias, but it may be arrhythmogenic. Much information exists concerning quinidine's effects on active properties in cardiac tissues, but virtually nothing is known of its effects on passive properties. We studied the effects of quinidine, in a clinically relevant concentration, on the balance among active and passive cellular properties that comprise the electrophysiologic matrix that determines cardiac excitability. The multiple microelectrode method of intracellular-current application and transmembrane voltage recording was used in sheep Purkinje fibers to determine strength-duration and constant current-voltage relations as well as cable properties. A rapid, on-line computerized data analysis system tracked in time the alterations in the active and passive properties relevant to excitability. In normal fibers at [K+]o = 5.4 mM, quinidine increased cardiac excitability as manifested by a decrease in the current required to attain threshold and/or a downward shift in strength- and charge-duration relations by altering passive properties despite a depressed sodium system and a slowed conduction velocity. During washout, excitability and passive properties remained altered despite a return of descriptive action potential parameters such as the resting potential, the maximum rate of rise of phase 0, overshoot, and the action potential duration to or nearly to normal. At [K+]o = 8.0 mM, quinidine could either increase or decrease excitability; net excitability depends on the balance between altered passive properties and the depressed sodium system. The results explain, in part, the antiarrhythmic actions and arrhythmogenic potential of quinidine. The data for quinidine and other antiarrhythmic drugs are interpreted in terms of the electrophysiologic matrix, which we believe has important advantages over traditional hierarchical classifications.