Similar endothelial glycocalyx structures in microvessels from a range of mammalian tissues: evidence for a common filtering mechanism?

The glycocalyx or endocapillary layer on the luminal surface of microvessels has a major role in the exclusion of macromolecules from the underlying endothelial cells. Current structural evidence in the capillaries of frog mesentery indicates a regularity in the structure of the glycocalyx, with a center-to-center fiber spacing of 20 nm and a fiber width of 12 nm, which might explain the observed macromolecular filtering properties. In this study, we used electron micrographs of tissues prepared using perfusion fixation and tannic acid treatment. The digitized images were analyzed using autocorrelation to find common spacings and to establish whether similar structures, hence mechanisms, are present in the microvessel glycocalyces of a variety of mammalian tissues. Continuous glycocalyx layers in mammalian microvessels of choroid, renal tubules, glomerulus, and psoas muscle all showed similar lateral spacings at ∼19.5 nm (possibly in a quasitetragonal lattice) and longer spacings above 100 nm. Individual glycocalyx tufts above fenestrations in the first three of these tissues and also in stomach fundus and jejunum showed evidence for similar short-range structural regularity, but with more disorder. The fiber diameter was estimated as 18.8 (± 0.2) nm, but we believe this is an overestimate because of the staining method used. The implications of these findings are discussed.

An investigation into the protonation states of the C1 domain of cardiac myosin-binding protein C.

Myosin-binding protein C (MyBP-C) is a myofibril-associated protein found in cardiac and skeletal muscle. The cardiac isoform (cMyBP-C) is subject to reversible phosphorylation and the surface-charge state of the protein is of keen interest with regard to understanding the inter-protein interactions that are implicated in its function. Diffraction data from the C1 domain of cMyBP-C were extended to 1.30 A resolution, where the of the diffraction data crosses 2.0, using intense synchrotron radiation. The protonation-state determinations were not above 2sigma (the best was 1.81sigma) and therefore an extrapolation is given, based on 100% data completeness and the average DPI, that a 3sigma determination could be possible if X-ray data could be measured to 1.02 A resolution. This might be possible via improved crystallization or multiple sample evaluation, e.g. using robotics or a yet more intense/collimated X-ray beam or combinations thereof. An alternative would be neutron protein crystallography at 2 A resolution, where it is estimated that for the unit-cell volume of the cMyBP-C C1 domain crystal a crystal volume of 0.10 mm3 would be needed with fully deuterated protein on LADI III. These efforts would optimally be combined in a joint X-ray and neutron model refinement.

Modelling muscle motor conformations using low-angle X-ray diffraction.

New results on myosin head organization using analysis of low-angle X-ray diffraction patterns from relaxed insect flight muscle (IFM) from a giant waterbug, building on previous studies of myosin filaments in bony fish skeletal muscle (BFM), show that the information content of such low-angle diffraction patterns is very high despite the 'crystallographically low' resolution limit (65 A) of the spacings of the Bragg diffraction peaks being used. This high information content and high structural sensitivity arises because: (i) the atomic structures of the domains of the myosin head are known from protein crystallography; and (ii) myosin head action appears to consist mainly of pivoting between domains which themselves stay rather constant in structure, thus (iii) the intensity distribution among diffraction peaks in even the low resolution diffraction pattern is highly determined by the high-resolution distribution of atomically modelled domain mass. A single model was selected among 5000+ computer-generated variations as giving the best fit for the 65 reflections recorded within the selected resolution limit of 65 A. Clear evidence for a change in shape of the insect flight muscle myosin motor between the resting (probably like the pre-powerstroke) state and the rigor state (considered to mimic the end-of-powerstroke conformation) has been obtained. This illustrates the power of the low-angle X-ray diffraction method. The implications of these new results about myosin motor action during muscle contraction are discussed.

Structure and nucleotide-dependent changes of thick filaments in relaxed and rigor plaice fin muscle.

The myosin crossbridge array, positions of non-crossbridge densities on the backbone, and the A-band "end filaments" have been compared in chemically skinned, unfixed, uncryoprotected relaxed, and rigor plaice fin muscles using the freeze-fracture, deep-etch, rotary-shadowing technique. The images provide a direct demonstration of the helical packing of the myosin heads in situ in relaxed muscle and show rearrangements of the myosin heads, and possibly of other myosin filament proteins, when the heads lose ATP on going into rigor. In the H-zone these changes are consistent with crossbridge changes previously shown by others using freeze-substitution. In addition, new evidence is presented of protein rearrangements in the M-region (bare zone), associated with the transition from the relaxed to the rigor state, including a 27-nm increase in the apparent width of the M-region. This is interpreted as being mostly due to loss or rearrangement of a nonmyosin (M9) protein component at the M-region edge. The structure and titin periodicity of the end-filaments are described, as are suggestions of titin structure on the myosin filament backbone.

A new twist in the collagen story--the type VI segmented supercoil.

Collagen occurs in two major forms: fibrillar and non-fibrillar. Non-fibrillar collagens are structurally more variable and relatively ill-understood. In this work we analysed the amino acid sequence of type VI collagen, a non-fibrillar collagen that forms antiparallel dimers. A sequence motif was discovered that gives rise to systematic molecular coiling. There is a common periodicity ( approximately 23 or 2 x 23 residues) in the charged amino acids, in the prolines and in the discontinuities in the Gly-X-Y triplets. In addition, there is a different periodicity ( approximately 21 amino acids) in the apolar groups. The two repeats mean that the only way to simultaneously maximize both the hydrophobic and polar interactions during dimer formation is with the molecules antiparallel, overlapped by 75 nm as observed, and supercoiled. The alternating proline-rich and charge-rich patches, often together with discontinuities in the Gly-X-Y sequences, coincide with each half-turn of the supercoil, thus breaking it into segments. We have termed this structure the collagen segmented supercoil. The segmented supercoil and variants may be common aggregation motifs for the non-fibrillar collagens.

Quasi-periodic substructure in the microvessel endothelial glycocalyx: a possible explanation for molecular filtering?

The luminal surface of endothelial cells is lined with the glycocalyx, a network structure of glycoproteins probably 50 to 100 nm thick. It has been suggested that a relatively regular fibre-matrix structure may be responsible for the ultrafiltration properties of microvascular walls, both when the endothelium is continuous and when it is fenestrated. Positive structural evidence demonstrating an underlying periodicity in the glycocalyx has been hard to obtain. Here we present structural analysis of glycocalyx samples prepared in a variety of ways for electron microscopy. Using computed autocorrelation functions and Fourier transforms of representative areas of the electron micrograph images, we show that there is an underlying three-dimensional fibrous meshwork within the glycocalyx with characteristic spacings of about 20 nm. Together with a fibre diameter consistent with our observations of about 10-12 nm, the 20-nm spacing provides just the size regime to account satisfactorily for the observed molecular filtering; the observations are consistent with the fibre matrix model. We also show that the fibrous elements may occur in clusters with a common intercluster spacing of about 100 nm and speculate that this may reveal organisation of the glycocalyx by a quasi-regular submembranous cytoskeletal scaffold.

Structure of abnormal molecular assemblies (collagen VI) associated with human full thickness macular holes.

Transversely banded deposits with an approximately 100-nm periodicity have been seen in association with a number of eye pathologies (e.g., age-related macular degeneration). Recently such aggregates have also been discovered in the cortical vitreous of a patient suffering from full thickness macular holes. The aggregates in the vitreous were of sufficient size and regularity for us to attempt 3D ultrastructural studies in the electron microscope. The molecules forming this aggregate pack in a centered tetragonal unit cell of dimensions approximately 26 x 26 x 180 nm. A real-space (r-weighted back projection) 3D reconstruction was computed. The aggregate is discussed in terms of its possible protein constituents. Collagen VI has been singled out as the most likely protein to form the aggregate. Two alternative models for the molecular packing are proposed, comprising aggregates of molecular tetramers or octamers. Understanding the structure of these abnormal banded deposits in the eye should help to throw light on the pathophysiological mechanisms of the diseases, including age-related macular degeneration, in which they occur.

A-band architecture in vertebrate skeletal muscle: polarity of the myosin head array.

Despite extensive knowledge of many muscle A-band proteins (myosin molecules, titin, C-protein (MyBP-C)), details of the organization of these molecules to form myosin filaments remain unclear. Recently the myosin head (crossbridge) configuration in a relaxed vertebrate muscle was determined from low-angle X-ray diffraction (Hudson et al. (1997), J Mol Biol 273: 440-455). This showed that, even without C-protein, the myosin head array displays a characteristic polar pattern with every third 143 A-spaced crossbridge level particularly prominent. However, X-ray diffraction cannot determine the polarity of the crossbridge array relative to the neighbouring actin filaments; information crucial to a proper understanding of the contractile event. Here, electron micrographs of negatively-stained goldfish A-segments and of fast-frozen, freeze-fractured plaice A-bands have been used to determine the resting myosin head polarity relative to the M-band. In agreement with the X-ray data, the prominent 429 A-spaced striations are seen outside the C-zone, where no non-myosin proteins apart from titin are thought to be located. The head orientation is with the concave side of the curved myosin heads (containing the entrance to the ATP-binding site) facing towards the M-band and the convex surface (containing the actin-binding region at one end) facing away from the M-band.

Partially systematic molecular packing in the hexagonal columnar phase of dogfish egg case collagen.

The collagen that forms the egg case of the dogfish Scyliorhinus canicula is stored in bulk in the female nidamental glands. Here the collagen molecules are thought to undergo a series of distinct pH-dependent liquid crystalline aggregation phase changes before assembling into the final arrangement encountered in the mature egg case. One liquid crystalline phase is hexagonal with the centres of two adjacent hexagons about 36 nm apart. We have collected tilt series of the hexagonal phase from plastic sections of the nidamental gland and have produced a three-dimensional reconstruction of the collagen arrangement of this phase. The reconstruction features axial columns of protein density lying regularly on the vertices of hexagonal cells of edge length 21 nm. Each column is connected to three nearest neighbours by irregular sheets of protein, but there appear to be preferred molecular directions at about 40 degrees to 50 degrees to the columns. The reconstruction has been interpreted in terms of known interactions of this collagen in other assemblies.

A new look at thin filament regulation in vertebrate skeletal muscle.

It is 30 years since Ebashi and colleagues showed that Ca2+ ions directly affect regulation of the myosin-actin interaction in muscle through the action of tropomyosin and troponin on muscle thin filaments. It is more than 20 years since the idea was put forward that tropomyosin might act, at least in part, by changing its position on actin, thus uncovering or modifying the myosin binding site on actin when troponin molecules take up Ca2+. Since that time, a great deal of evidence for and against this steric blocking mechanism has been published: a structure for actin filaments at close to atomic resolution has been proposed, and the whole regulation story has become both more complicated and more subtle. Here we review structural and biochemical aspects of regulation in vertebrate skeletal muscle. We show that some basic ideas of the steric blocking mechanism remain valid. We also show that additional factors, such as troponin movements and structural changes within the actin monomers themselves, may be crucial. A number of the resulting regulation scenarios need to be distinguished.

Myosin head configuration in relaxed fish muscle: resting state myosin heads must swing axially by up to 150 A or turn upside down to reach rigor.

The arrangement and shape of myosin heads in relaxed muscle have been determined by analysis of low-angle X-ray diffraction data from a very highly ordered vertebrate muscle in bony fish. This reveals the arrangement and interactions between the two heads of the same myosin molecule, the shape of the resting myosin head (M.ADP.Pi) assuming a putative hinge between the myosin catalytic domain and the light chain binding-domain, and the way that the actin-binding sites on myosin are arrayed around the actin filaments in the bony fish muscle A-band cell unit. The results are discussed in terms of possible force-generating mechanisms. Changes in myosin head shape or tilt have been implicated in the mechanism of force generation. The myosin head arrangement, including perturbations from perfect helical symmetry, has all heads oriented roughly the same way up (there is only a small range of rotations around the head long axis). X-ray data do not define the absolute polarity of the myosin head array. The resting head rotation is either similar to (65 degrees difference) or opposite to (115 degrees difference) the rotation in the rigor state. If the rotations are similar, probably the more likely possibility, then the average relative axial displacement of the inner and outer ends of the heads from the resting state to rigor is about 140 to 150 A. If (less likely) the resting head rotation is opposite to rigor, then the heads would need to turn over (i.e. rotate about 115 degrees around their own long axes) and the mean relative axial displacement from relaxed to rigor would only be 20 to 30 A.

Architecture and function in the muscle sarcomere.

Striated muscle sarcomeres in vertebrates comprise ordered arrays of actin and myosin filaments, organized by an elaborate protein scaffold. Recent innovative work in a number of laboratories has greatly improved our knowledge of these structures, their organization and their interactions. Structural details have been reported on myosin filaments, actin filaments, Z-bands, M-bands, titin, and nebulin. Time-resolved X-ray diffraction and electron microscopy are revealing the molecular movements involved in force production and regulation.

Evolution of myosin filament arrangements in vertebrate skeletal muscle.

A survey of skeletal muscles throughout craniates shows basic kinds of myosin filament arrangement, simple-lattice and superlattice, within the A-band of each sarcomere. Distribution of simple- and superlattice arrangements across a phylogeny of craniates suggests that the superlattice arrangement is primitive and that Amia and teleosts are derived in showing simple-lattice arrangements. Two taxa examined (Scyliorhinus and Acipenser) show both lattice types within the same organism implying that there is not a simple evolutionary transformation of one to the other fiber arrangement. We discuss the possible functional significance of the different lattice types. We believe that the crossbridges may have greater competition for actin binding sites in simple-lattice muscles compared to the superlattice types.

Packing of alpha-helical coiled-coil myosin rods in vertebrate muscle thick filaments.

Muscle myosin filament backbones are aggregates of long coiled-coil alpha-helical myosin rods, with the myosin heads arranged approximately helically on the filament surface, but the details of the rod packing are not known. Computed Fourier transforms of plausible molecular packing models for the vertebrate striated muscle myosin filament have been compared with observed high-angle X-ray diffraction patterns from plaice fin muscle. Models considered include those in which the coiled-coil rod parts of myosin are packed into various kinds of subfilaments or into a curved molecular crystalline layer. A general conclusion is that if the myosin rods are tilted by less than about 1 degree or more than about 3 degrees from the filament long axis, very poor agreement is obtained between the computed and observed high-angle diffraction patterns. Qualitative comparison of calculated Fourier transforms, taken together with electron micrograph information, shows that the curved molecular crystal model and a model with hexagonally close-packed 4-nm subfilaments appear to explain the whole set of observations more satisfactorily than the alternatives. It is argued on other grounds that of these two possibilities the curved molecular crystal model is the more plausible.

Structural changes in actin-tropomyosin during muscle regulation: computer modelling of low-angle X-ray diffraction data.

The crystal structure of G-actin monomer has been used together with tropomyosin in a filament model to explain the low-angle X-ray diffraction data from relaxed and activated actin filaments. The four-subdomain actin monomer can be approximated quite well by a four-sphere unit. Orienting this unit and tropomyosin into a filament by searching for the best fit between the computed Fourier transform and the observed vertebrate skeletal muscle low-angle actin layer-lines from muscles at non-overlap sarcomere lengths produced models for the structural changes within the thin filaments (actin plus tropomyosin) between the resting state and the active states, which occur as a result of calcium-activation and independent of myosin interaction with actin. The models are very sensitive to changes in the positions of the centres of mass of the subdomains, but not to the exact shape of the objects used to represent them (e.g. spheres, ellipsoids etc.), as long as the volume is fixed, at the resolution here considered. It is concluded that, even with a four-subdomain structure for the actin molecules, the observed low-angle X-ray diffraction patterns cannot be explained without a substantial azimuthal swing of the tropomyosin strands when resting filaments are calcium-activated. The direction of this swing upon calcium-activation is away from a position close to the proposed major binding site of the myosin head on actin; a result consistent with the original "steric blocking model" of thin filament-based regulation in which the tropomyosin position on actin is crucial for regulation of the myosin crossbridge cycle on actin. Tropomyosin sterically hindering myosin attachment in the "off" state remains a possibility. However, even in the "on" state, the tropomyosin position is close enough to the myosin-binding site to have an effect, where it could regulate the transition of the head from a weak to a strong state. In addition to this tropomyosin movement there are small, but plausible, actin subdomain movements. A tropomyosin shift on its own will not explain the data. Allowance for possible movement of actin subdomain 2 along with the tropomyosin shift still does not explain the data. An additional small movement of subdomain 1; the main myosin-binding subdomain, is postulated.

M-band structure, M-bridge interactions and contraction speed in vertebrate cardiac muscles.

Cardiac muscle M-band structures in several mammals (guinea pig, rabbit, rat and cow) and also from three teleosts (plaice, carp and roach), have been studied using electron microscopy and image processing. Axial structure seen in negatively stained isolated myofibrils or negatively stained cryo-sections shows the presence of five strong M-bridge lines (M6, M4, M1, M4' and M6') except in the case of the teleost M-bands in which the central M-line (M1) is absent, giving a four-line M-band. The M4 (M4') lines are consistently strong in all muscles, supporting the suggestion that bridges at this position are important for the structural integrity of the A-band myosin filament lattice. Across the vertebrate kingdom, cardiac M-band ultrastructure appears to correlate roughly with heartbeat frequency, just as in skeletal muscles it correlates with contraction speed, reinforcing the suggestion that some M-band components may have a significant physiological role. Apart from rat heart, which is relatively fast and has a conventional five-line M-band with M1 and M4 approximately equal, the rabbit, guinea pig and beef heart M-bands from a new 1 + 4 class; M1 is relatively very much stronger than M4. Transverse sections of the teleost (roach) cardiac A-band show a simple lattice arrangement of myosin filaments, just as teleost skeletal muscles. Almost all other vertebrate striated muscles, including mammalian heart muscles, have a statistical superlattice structure. The high degree of filament lattice order in teleost cardiac muscles indicates their potential usefulness for ultrastructural studies. It is shown that, in four-line M-bands in which the central (M1) M-bridges are missing, interactions at M4 (M4') are sufficient to define the different myosin filament orientations in simple lattice and superlattice A-bands. However the presence of M1 bridges may improve the axial order of the A-band.

Evaluation of high-resolution shadowing applied to freeze-fractured, deep-etched particles: 3-D helical reconstruction of shadowed actin filaments.

Images of shadowed F-actin filaments on mica surfaces obtained using a quick-freeze, freeze-fracture, deep-etch technique were subjected to conventional 3-D helical reconstruction methods. Although the shadowing must vary systematically from subunit to subunit, the computed transforms of isolated filaments were characteristic of the helical actin transform. Helical reconstruction was therefore judged to be valid. The theoretical basis for such reconstruction is outlined. The reconstructions showed an average thin (about 3 nm) layer of shadow on the filament surface and both the outer and the inner surfaces of the shadow layer could be visualized. By comparison with the F-actin structure postulated by Holmes et al. (1990) on the basis of the known structure of the actin monomer, it is shown that, at the resolution considered, the inner surface of the shadow provides a reasonably faithful outline of the molecular surface. This, in turn, confirms that the original 3-D structure of the protein molecules has been well preserved throughout the whole preparation procedure up to the final replica. The "shadowed" filaments can thus be correlated axially and azimuthally with known actin structures and, in principle, features such as myosin head location on decorated filaments can be determined. The result emphasizes the amount of detail present in good quality images of shadowed particles and, in this case, shows that detailed evaluation of molecules labeling actin can be made.

Molecular movements in contracting muscle: towards "muscle--the movie".

The recent publication of the crystal structures of G-actin and of myosin subfragment-1, together with analysis of a time-resolved series of well sampled low-angle 2D X-ray diffraction patterns from bony fish muscle permits the study of the molecular movements in muscle that are associated with generation and regulation of contractile force. Here it is shown that even though low-angle (i.e. low resolution) X-ray diffraction patterns are being used, these patterns are sensitive, for example, to sub-domain movements of as little as 3 A or 4 degrees within the actin monomers of actin filaments. Actin filament diffraction patterns from whole muscle are being used to define actin domain and tropomyosin movements involved in regulation. Myosin and actin filament diffraction patterns are being used together to start to show how the complete "quasi-crystalline" unit cell in the bony fish muscle A-band can be modelled as a series of time-slices through a typical tetanic contraction of the muscle. In this way, the time sequence of images can be used to create "muscle--the movie".