Helicoidal microfibril deposition in a tip-growing cell and microtubule alignment during tip morphogenesis: a dry-cleaving and freeze-substitution study

Cell wall microfibril alignment in the tubular portion of Equisetum hyemale root hairs is helicoidal. Lamellae of helicoidal texture are deposited from tip to base; thus, different microfibril orientations are aligned with the plasma membrane successively. Zones with constant mean microfibril orientation are about 300 μm long. In any such zone of dry-cleaned, shadowed preparations, the frequency of microfibrils at the proximal end is 5 to 7 microfibrils per micrometre, which decreases to 0 at the distal end. The orientation of microfibrils of the underlying lamella, the microfibril frequency of which is 5 to 7/μm throughout, is the same as the microfibril orientation of the neighbouring distal lamella. Microfibrils of the cell wall are randomly oriented in the hair dome. Microtubule alignment in these root hairs was examined by means of freeze substitution. In the extreme tip of the root hair, microtubules run parallel to the plasma membrane and transverse to the long axis of the hair; the hemisphere of the hair contains randomly oriented microtubules. From extreme tip to base of the hair dome, microtubules become more and more axially aligned, and remain axially oriented in the hair tube. Further down the hair, where microfibril alignment is transverse and microfibrils are actively being deposited, microtubules still run in the axial direction. The observations emphasize the involvement of microtubles in root hair tip morphogenesis, but not in determining the alignment of the microfibrils in the hair tube.

Application of Low-Voltage SEM in the Analysis of Complex Intracellular Structures

In 1986 the Madison IMR acquired a Hitachi S-900 LVSEM and I became interested in exploring whether it could help in the analysis of threedimensional structures where thin and thick sections had failed. The question was not whether cytoplasmic structures known from TEM were visible by SEM, but whether LVSEM could provide unique information that was not supplied by other means. In a study of intracellular transport of pigment granules in xanthophores of the goldfish, HVEM pictures of the thin edge of cells had shown that the carotenoid granules were attached to intermediate filaments (IF). But it was impossible to see this relationship in the interior of cells either in whole mounts or in thick sections. To see internal structure with SEM, the cell has to be cracked open without causing distortions. We found two simple methods that gave useful results: 1.Dry-cleaving.

Structure and molecular organization of higher plant coated vesicles

Suspension-cultured cells of carrot contain three populations of coated vesicles, associated with the plasma membrane (84–91 nm diameter), Golgi dictyosomes and the partially coated reticulum (61–73 nm diameter). These were observed by thin sectioning, dry-cleaving and rapid-freeze deep-etching of cells. Dissociation of clathrin coats with Tris, released triskelions that were morphologically identical with those from mammalian tissue. The triskelion arm length of carrot clathrin was greater (61nm versus 44–50 nm), but packaging results in clathrin cages of pentagons and hexagons of similar size to those from mammalian cells. SDS-PAGE of Tris-released triskelion preparations revealed a complex of three polypeptides of 190, 60 and 57(x103)Mr. The 190x103Mr protein is the plant clathrin heavy chain, slightly larger than the mammalian heavy chain. The 60 and 57(x103)Mr bands showed the same sensitivities to protease treatment as mammalian light chains. Triskelion preparations containing these three proteins reassembled into polyhedral cages. These results are discussed in relation to the structural organization of coated vesicles and clathrin cages in other systems.

Localization of cell surface glycoproteins in membrane domains associated with the underlying filament network.

To visualize the localization of cell surface constituents in relation to the plasma membrane-associated filament network, we developed a method based on a combination of immunogold labeling and dry-cleaving. For labeling we used trinitrophenyl-derivatized ligand, anti-TNP antibodies, and protein A-coated colloidal gold. Dry-cleaving (Mesland, D. A. M., H. Spiele, and E. Roos, 1981, Exp. Cell Res., 132: 169-184) involves cleavage of lightly fixed critical point-dried cells by means of adhesive tape. Since cells cleave close to the cell surface, the remaining layer is thin enough to be examined in transmission electron microscopy. Using this method, we studied concanavalin A-binding constituents on the medium-facing surface of H35 hepatoma cells. The distribution of the gold particles, which was partly dispersed and partly patchy, coincided strikingly with membrane-associated filaments, and label was virtually absent from areas overlying openings in the filament network. In stereo pairs we observed the label to be localized to areas of somewhat enhanced electron density at the plane of the membrane. These areas were interconnected in a pattern congruent with the filament network. Preliminary observations on wheat germ agglutinin receptors on the hepatoma cells as well as concanavalin A receptors on isolated hepatocytes yielded comparable results. It thus appears that surface glycoproteins, although seemingly randomly distributed as observed in thin sections, may actually be localized to particular membrane domains associated with underlying filaments.

Brief extraction with detergent induces the appearance of many plasma membrane-associated microtubules in hepatocytic cells

In cultured H35 hepatoma cells membrane-associated cortical networks have a microtrabecular appearance as revealed by dry-cleaving. Filaments having diameters of 15 nm can be readily distinguished within these networks and have not been described previously. Microtubules are seldom observed to be part of this structure. Extraction of cells with 0.1% Saponin in microtubule-stabilizing buffer produces holes in the membrane and reorganization of the networks resulting in the loss of microtrabecular structure, the loss of 15 nm filaments and the appearance of abundant membrane-associated microtubules (about 1.25 micron per micron2 substrate-adherent membrane). These observations were confirmed by immunolabelling experiments with affinity-purified anti-tubulin immunoglobulin G. By both fluorescence microscopy and electron microscopy it was shown that labelled tubulin in the cortical networks became organized into microtubules upon treatment with detergent. By determination of the microtubule density, expressed as micron microtubule per micron2 membrane, the effects of various conditions on microtubule occurrence were determined. The Saponin-induced appearance of microtubules in the membrane-associated network could be inhibited by: 1% and 2% glutaraldehyde, 0 degrees C, millimolar Ca2+, absence of Mg2+ (subsequent reversal of inhibition by addition of Mg2+ was shown), and 20 microM-nocodazole (but not 20 microM-colchicine). In addition to Saponin, extraction with 0.1% Nonidet P-40 or 0.1% Triton X-100 also resulted in microtubule-containing cortical networks. However, 0.1% Triton N-101 was not effective, although holes were produced in the plasma membrane. These data provide evidence suggesting rapid polymerization of membrane-associated microtubule protein rather than detergent-induced displacement or collapse of existing microtubules. The arguments for this hypothesis and its implications are discussed.

Plasma membrane-associate filament systems in cultured cells visualized by dry-cleaving

Substrate-attached critical-point-dried cells cleave along the level of the substrate-adherent membrane if removed by means of adhesive tape. The remaining membrane fragments on grids can be visualized three-dimensionally by means of stereo transmission electron microscopy. Attachment of cells may be achieved by active spreading of the cell, or artificially by poly-L-lysine adherence of prefixed cells. In 11 different cell types a filamentous network appears to remain associated with the cytoplasmic face of the membrane. In one hepatoma cell type virtually no filamentous network could be detected. Two general network morphologies are described: the hepatocytic network and the lymphoid network. Since no correspondence could be found between cytoplasmic structure and the structure of the membrane-associated network, and since cells generally cleave along the level of this network, excluding cell organelles, we conclude that it comprises a distinct structural system, analogous to the membrane skeleton of the red cell membrane.