Planta
Planta (1983)159:347-356
9 Springer-Verlag 1983
Arrays of plasma-membrane "rosettes" involved in cellulose microfibril formation of Spirogyra Werner Herth Zellenlehre, Universitfit Heidelberg, Im Neuenheimer Feld 230, D-6900 Heidelberg, Federal Republic of Germany
Abstract. The cell-wall structure and plasma-membrane particle arrangement during cell wall formation of the filamentous chlorophycean alga Spirogyra sp. was investigated with the freeze-fracture technique. The cell wall consists of a thick outer slime layer and a multilayered inner wall with ribbon-like microfibrils. This inner wall shows three differing orientations of microfibrils: random orientation on its outside, followed by axial bundles of parallel microfibrils, and several internal layers of bands of mostly five to six parallel associated microfibrils with transverse to oblique orientation. The extraplasmatic fracture face of the plasma membrane shows microfibril imprints, relatively few particles, and "terminal complexes" arranged in a hexagonal package at the end of the imprint of a microfibril band. The plasmatic fracture face of the plasma membrane is rich in particles. In places, it reveals hexagonal arrays of "rosettes". These rosettes are best demonstrable with the double-replica technique. These findings on rosette arrays of the zygnematacean alga Spirogyra are compared in detail with the published data on the desmidiacean algae Micrasterias and Closterium. Key words: Cellulose microfibril formation - Plasma membrane rosettes - Spirogyra (freeze fracture).
Introduction
Algae have become important model systems in cell-wall research: Valonia has been a classic object for structural and X-ray studies of cellulose I, some algae have been demonstrated to use other structural polysaccharides like fl-l,3-xylan, fl-l,4-manAbbreviations: EF = Extra plasmaticfracture face; PF = Plasmatic fracture face
nan or chitin, and with the scale-forming Pleurochrysis it was shown that cellulose may be formed within the Golgi cisternae (compare reviews by: Preston/974; Brown et al./973; Herth et al. 1977; Herth 1979). For a long time, the cytological site of cellulose formation has been a point of controversal discussions. Several years ago, Preston formulated the hypothesis that cellulose microfibrils are synthesized by enzyme complexes in the plasma membrane (Preston 1974). Then, with the freezefracture technique, a variety of plasma-membraneparticle complexes were supposed to be identical with these postulated cellulose-synthetase enzyme complexes (for survey see Schnepf and Herth /978). The particle arrangements in the alga Oocystis (Brown and Montezinos /976), in Pelvetia embryos (Peng and Jaffe 1976) and in higher-plant protoplasts (e.g. Willison and Cocking /975; Robenek and Peveling 1975,/977) were rather different, so that their identification as cellulose synthetase was often questioned (e.g. Schnepf and Herth /978; Davey and Mathias 1979; Wilkinson and Northcote 1980). A breakthrough came with the freeze-fracture findings on the green alga Micrasterias (Kiermayer and Sleytr 1979; Giddings et al. /980): there, hexagonal arrays of rosettes, with six subunits each, are found on the plasmatic fracture face (PF) of the plasma membrane during the formation of the bands of secondary-wall microfibrils (freeze-fracture terminology according to Branton et al. /975). During primary-wall formation, only individual microfibrils are formed, and only single rosettes were found on the PF of the plasma membrane. The authors concluded that the rosettes are cellulose-synthetase complexes, each forming a 5-nm subunit microfibril, the number of cooperating rosettes in a row arrangement should determine the dimension of the microfibril produced. Mueller and Brown (/980) also demonstrated individual ro-
W. Herth : Microfibril formation of Spirogyra
348
settes on the PF of plasma membranes of corn, bean and pine seedlings, and terminal globules on the extra plasmatic fracture face (EF). These authors then said that the rosettes are only preserved when the tissue is not cryoprotected and extremely rapidly frozen. Meanwhile, further examples for rosettes have been reported: Wada and Staehelin (1981) found individual rosettes on the plasma-membrane PF in the growth region of the protonema of the fern Adiantum, Staehelin and Giddings (1982) found rows of rosettes during primary-wall formation of the green alga Closterium, and arrays of rosettes corresponding to those of Micrasterias during secondary-wall formation of Closteriurn. Rosettes now have also been observed after glutaraldehyde fixation and cryoprotection in higher plants (Mueller 1982) and in Micrasterias (Noguchi et al. 1981). Based on these collective findings, and on observations, mostly unpublished, of other higherplant systems, Mueller (1982) claimed that the terminal complex-rosette structure should be universally present in cellulose-forming higher-plant systems. In algae, however, different types of plasmamembrane particle arrangements involved in cellulose microfibrils seem to have evolved. It will be interesting from the phylogenetic point of view to find out in which organism the first rosettes may have arisen, and to examine the breadth of distribution of rosettes among different algal groups. As a start to such a survey I investigated the wellknown filamentous green alga Spirogyra, which belongs to the Zygnemataceae, a group related to the two documented algal examples with rosettes, Micrasterias and Closterium, which both belong to the Desmidiaceae. The Desmidiaceae and Zygnemataceae both belong to the old group of Conjugatae (Fott 1971; Oltmanns 1922; van den Hock 1978). Materials and methods Culture conditions. Spirogyra sp. was obtained from Professor W. Koch, Culture Collection of Algae, University of G6ttingen, FRG. The algae were cultivated in a medium according to Professor W, Koch consisting of: K N O 3 0.01% (w/v); (NH4)2HPO4 0.001% (w/v); M g S O 4 . 7 H 2 0 0.001% (w/v); i ml saturated CaSO 4 solution in 100 ml; 2 ml soil extract in 100 ml; 0.5 ml stock solution of trace elements in 100 ml; 0.5 ml stock solution of vitamin BI2 in I00 ml. They were kept at 2,000-3,000 lx, 14:10 h light/dark cycle, 19 ~ C in a thermostatically controlled tank (Ruharth, Hannover, FRG). Freeze-fracture. Three different experimental procedures were applied: a) With cryoprotection. The algae were fixed in 2.5% glutaraldehyde, 0.1 M cacodylate buffer, pH 7.2, for 30 rain at
20 ~ C; transferred for 2 h into 20% glycerol with the same buffer, 20 ~ C; then transferred to the specimen holder, and frozen via Freon 22 in liquid nitrogen. b) Without cryoprotection. Bundles of the filamentous algae were collected with forceps, cut into short parallel filaments with a razor blade, immediately transferred to the conventional specimen holder, and frozen via Freon 22 in liquid nitrogen. c) Without cryoprotection, double-replica technique (modified after Giddings et al. 1980). The algae were collected and cut into short filaments as in b), then suspended in water-diluted yeast paste, and a small drop of the algal suspension was then transferred to the double-replica holder. Again the material was frozen via Freon 22 in liquid nitrogen. The specimens then were freeze-fractured in a Balzers (Liechtenstein) BAF 400 apparatus at - 1 0 0 ~ C, and shadowed subsequently with platinum-carbon and carbon under control of the quartz monitor. The replicae were cleaned with 60% (v/v) chromosulfuric acid overnight, washed with distilled water and taken up with 200-mesh copper grids for electron microscopy. The replicae were examined with a model 400 electron microscope (Philips, Eindhoven, The Netherlands). The magnifications were calibrated with a grating replica.
Results
Of the three technical variants applied, the first one, with cryoprotection by glycerol after glutaraldehyde fixation, proved absolutely unsatisfactory because of tremendous vesiculations of the plasma membrane. The two other methods without cryoprotection yielded somewhat differing results as follows.
No cryoprotection, single replica after fracture with the blade. Cross-fractures of the wall show two distinct layers. An outer layer of radial elements showing tree-like branching corresponds to the slime layer which yields the well-known slipperiness of Spirogyra (between black arrows in Fig. 1). The inner layer (between white arrows in Fig. 1) is relatively thin in comparison to the slime layer and contains the cellulosic wall component. In fractures tangential to the cellulosic wall layer, sublayers become evident. The outer layers show randomly oriented microfibrils (Fig. 2 a), followed internally by parallel associated micro fibrils (Fig. 2a, arrows). The innermost layers are composed of bands of laterally associated microfibrils (Fig. 2b). Each band consists of 5-6 parallel ribbon-like microfibrils (between arrows in Fig. 2b) with widths of approx. 20 nm. The various bands show a mostly transverse to oblique orientation, but are rather irregular, whereas the parallel microfibrils of the outer sublayer are oriented parallel to the longitudinal axis of the elongate cell. After freeze-fracture, the EF of the plasma membrane of Spirogyra shows imprints of the microfibrils (Fig. 3 a), and both the sublayer of the parallel microfibrils and a few bands of microfi-
W. Herth: Microfibril formation of Spirogyra
349
Fig. 1. Cross-fracture of Spirogyra sp., no cryoprotection, fractured with the blade. Slime layer (between black arrows) and microfibrillar layers (between white arrows). B a r = 1 gm; x 10,000
brils are distinguishable. Fibrillar ends are not frequent, but at closer examination of such EFs, the fibrillar bands are seen to end in globular "terminal complexes" (terminology according to Mueller et al. 1976; Mueller and Brown 1980) which are closely packed in a hexagonal arrangement (Fig. 3 b, arrows). These terminal complexes are arranged in rows corresponding to the number of microfibrils associated in the band ending there; they consist of three to five globules per row in this figure. The smallest array of globules encountered was three rows of 2, 3 and 3 globules; the biggest is shown in Fig. 4a and consists of five rows of up to four globules per row. This high magnification shows that each globule has a diameter of approx. 24 nm, and a center-to-center spacing of approx. 30 nm. The EF shows relatively few particles (approx. 1,000 gm -2) (Fig. 4b). In contrast to the EF, the PF is rich in particles of various sizes (approx. 2,000-3,000 Bm-2). These particles appear individually or associated with one or more other particles. It is rather difficult to detect the structures corresponding to the terminal globules in this mass of particles. However, once found, the corresponding structures clearly are rosettes of six subunits each which are associated into a hexagonal array (Fig. 4 b, rosettes encircled). As with the terminal globules, they appear in rows of up to five rosettes, and again there are five rows corresponding to the number of microfibrils associated in the band. The rosette diameter is approx. 24 nm, each subunit particle has a diameter of approx. 8 nm.
Double-replica technique. With
this technique, the yield of well-fractured cells was much higher, and the cells showed better preservation and fewer ice crystals, probably due to the faster rate of cooling for the specimens sandwiched between the two metal surfaces (compare corresponding remarks of Mueller 1982). N o t only the plasma-membrane EF, which looked very similar to the figures described above, but also the PF showed clear imprints of microfibrils (Fig. 5a). In one case, the imprint of a band of microfibrils (arrow in Fig. 5 a) clearly is seen to end in an array of rosettes. Several rosette arrays were found in close proximity (Fig. 5a, in circles), whereas large areas of the PF appeared without rosette arrays. The number and arrangement of the rosettes varied from a few linear rows of few rosettes to arrays with six rows of rosettes (Fig. 5b). The rosettes appeared more distinct, as the center-to-center spacing was greater in these preparations, and the space between the rosettes appeared smooth. Under high magnification (Fig. 5 c), the individual rosettes show six subunits, have a diameter of 22.5 nm, and a center-tocenter spacing of 35 nm. In this largest rosette array found yet for Spirogyra, six rows of rosettes are associated. The microfibril imprints leading to this array (not to be seen at this high magnification) come from the top of this figure, indicating that the microfibril band arises from a sequence of 2, 4, 4, 5, 6 and 5 rosettes. In the cell wall, the highest number of microfibrils associated in a band was eight, but most bands had five to six microfibrillar subunits.
350
W. Herth: Microfibril formation of Spirogyra
Fig. 2a, b. Tangential fractures of the cell wall, method as Fig. 1. a A r a n d o m outermost layer of microfibrils (lower part) is followed by bundles of axially oriented microfibrils (arrows). b The internal part of the cell wall consists of bands of five to six parallel oriented microfibrils, with main orientation transverse to oblique to the cell axis. Note one band between the arrows. Shadow angle=encircled arrows; bars = i gm; x 40,000 (a); x 60,000 (b)
W. Herth: Microfibril formation of
Spirogyra
351
Fig. 3 a, b. Plasma membrane of Spirogyra sp., extraplasmatic fracture face (EF). Method as Fig. 1. a Imprints of the parallel bundles of microfibrils (arrowheads) and of a few bands of microfibrils (e.g. at the arrows) are readily discerned on the EF. b At higher magnification, imprints of microfibril bands are seen to terminate in "terminal complexes" (TC; arrows). Shadow angle = encircled arrows; rnfi= microfibril imprint; b a r s - 1 ~m; x 30,000 (a); x 60,000 (b)
352
W. Herth: Microfibril formation of Spirogyra
Fig. 4a, b. High magnifications of the two fracture faces of the plasma membrane of Spirogyra sp., method as Fig. 1. a EF with imprint of microfibrils (follow arrows) ending in a terminal complex consisting of globules arranged in rows. b Plasmatic fracture face (PF) with very dense particle population and a hexagonal array of rosettes (encircled) consisting each of six subunit particles. Shadow angle=encircled arrows; bars=0.2 gm; x 180,000 (a); x 240,000 (b)
W. Herth: Microfibril formation of Spirogyra
353
Fig. 5a-c. PF of the Spirogyra sp. plasma membrane, double-replica technique, a Survey of a region with several rosette arrays (in circles). In one case, a microfibril band imprint ends in a rosette array (arrow). b Higher magnification from a, with two rosette arrays showing different numbers of rosettes arranged in rows (arrows). e High magnification of a big rosette array with almost triangular arrangement of the rosettes (between arrows). Microfibril imprints come from top of the figure. Note the six subunits of the rosettes and the relatively big center-to-center spacing between the rosettes. Shadow angles=encircled arrows; b a r s = 1 gm (a), 0.1 gm (b, e); x 40,000 (a); x 150,000 (b); x 280,000 (e)
354
W. Herth: Microfibrilformation of Spirogyra
Discussion
Spirogyra, the microfibril dimensions in one band
The cell-wall architecture of Spirogyra after freezefracture confirms the earlier publications based on light microscopy and thin-section analysis (Jordan 1970; Dawes 1965; Buer 1964). The slime layer seems to be secreted across the microfibrillar layers; there are no distinct pores as in Micrasterias (for review see Neuhaus and Kiermayer 1981). The microfibrillar layers show a similar arrangement to those of Micrasterias, with randomly arranged individual microfibrils in the primarywall layer and bands of parallel associated microfibrils in the secondary-wall layer (Giddings et al. 1980). In addition, Spirogyra has a layer of axial parallel microfibrils, which seems to be the first layer of secondary-wall formation, which then is followed by the more transversely microfibrillar bands. These parallel microfibrils might also indicate the end of primary-wall formation, since in Closterium the primary wall consists of parallel microfibrils (Staehelin and Giddings 1982). The chemical nature of the wall microfibrils of Spirogyra has been characterized by X-ray diffraction (Kreger 1957; Dawes 1965). They consist of a cellulose modification which is remarkable in that it shows a uniplanar orientation with the crystal plane (101) or (002), according to the species, parallel to the wall surface, instead of (101) parallel to the surface as in Valonia cellulose I. This indicates a certain configurational difference from the typical cellulose I. Comparable data are not yet available for Micrasterias, as this alga has not yet been chemically analysed for cell-wall composition and the X-ray pattern of the microfibrils. The cellulosic nature of the Micrasterias fibrils has only been deduced from staining reactions. Comparison of the freeze-fracture results shows that the terminal globules in Spirogyra plasma-membrane EF are more distinct than in Micrasterias, the PF of the Spirogyra plasma membrane contains more particles, and the rosettes are arranged in smaller arrays (five to six rows of up to five rosettes instead of up to 16 rows with up to 15 rosettes). As in Micrasterias, the rosettes have six subunits, and have almost the same dimensions and the same center-to-center spacing. In Micrasterias, the micro fibril width of the central microfibril of a band is greater than that of the peripheral microfibril, and is assumed to be formed by more rosettes (Giddings et al. 1980; Staehelin and Giddings 1982). Each rosette is believed to form one 5-nm subunit microfibril, the products of the rosettes in one row then laterally associate and crystallize (Giddings et al. 1980). In
are more uniform, in the range of 20 nm. The number of rosettes in one row engaged in microfibril formation is five to six in the central part of the array. This means that one rosette here should synthesize an approx. 4-nm microfibrillar subunit. This value is close to the postulated "elementary fibrils" (Mfihlethaler I967) of 3.5 nm diameter, if the thickness of the metal shadow coat is subtracted from these 4 nm (Ohad and Danon 1964). This would probably indicate that each of the six subunits of a rosette might synthesize six glucan chains which then co-crystallize into a 36-glucanchain "elementary microfibril". The sub-elementary microfibrils of smaller dimensions (e.g. Franke and Ermen 1969; Hanna and C6t~ 1974; Herth et al. 1974, 1975) then might be the result of less than six, or even individual, enzyme particles. It seems that the limitation of the fibrillar width is given by the distance of the rosettes, and not so much by the number of rosettes. If the calculations above are correct, then the Spirogyra band microfibrils should consist only of a single layer of laterally associated 4-nm subunits, whereas in Micrasterias more layers are depicted in the model of Giddings et al. (1980). In this model, Giddings et al. do not draw a flat ribbon for the microfibril made by five to six rosettes, but give a rather isodiametric cross section, whereas in their diagram the broader microfibrils, made by the central ten rosettes, consist of three layers of three- to four-subunit microfibrils. A microfibril made by 16 rosettes should then contain four layers each of four subunits, representing a cross-section of 20 x 20 nm. However, as judged from the equal shadows in their Fig. 3, the micro fibrils of Micrasterias have equal thickness. As in Spirogyra a row of five to six rosettes is already capable of forming a 20-rim microfibril, this seems to indicate that not all the rosettes of a 16-rosette row in Micrasterias seem to be active at the same time. Moreover, in Closterium (Staehelin and Giddings 1982), five to six rosettes in one row make a ribbon of 16.5 nm, a value rather close to Spirogyra, supporting the conclusions discussed above. The random microfibrils of the primary-wall layer of Spirogyra are probably made by individual rosettes as in Micrasterias (Giddings et al. 1980) and in ferns and higher plants (Wada and Staehelin 1981; Mueller 1982), but these rosettes are difficult to detect with the high background of particles on the Spirogyra plasma-membrane PF. The parallel layer of microfibrils may be formed by rows of rosettes, as in Closterium (Staehelin and Giddings 1982), but these stages have not been de-
W. Herth: Microflbril formation of Spirogyra
tected in my freeze-fracture replicas. It is possible that individual particles, indistinguishable from other intramembrane particles, are involved in microfibril formation in these stages. The Spirogyra rosette arrays are intermediate between the rows of rosettes in Closterium primary-wall formation and the larger arrays of rosettes found in desmid secondary-wall formation (Staehelin and Giddings 1982). This might indicate that Spirogyra exhibits a lower status of evolution of the rosette arrays than the desmids. The concept of the rosette as a plasma-membrane-bound multienzyme complex involved in cellulose formation (see Introduction) is supported by the findings with Spirogyra. However, the number of well-documented cases is small and the generality of the rosette concept of cellulose microfibril formation requires further investigation (for critical discussion see e.g. Willison and Klein 1982). The author thanks Professor E. Schnepf and Dr. H.-D. Reiss for valuable discussions, and B. Heck for technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft.
References Branton, D., Bultivant, S., Gilula, N.B., Karnovski, M.J., Moor, H., Mfihlethaler, K., Northcote, D.H., Packer, L., Satir, B., Satir, P., Speth, V., Staehelin, L.A., Steere, R.L., Weinstein, R.S. (1975) Freeze-etching nomenclature. Science 190, 54-56 Brown, R.M., Herth, W., Franke, W.W., Romanovicz, D. (1973) The role of the Golgi apparatus in the biosynthesis and secretion of a cellulosic glycoprotein in Pleurochrysis: A model system for the synthesis and secretion of structural polysaccharides. In: Biogenesis of plant cell wall polysaccharides, pp. 207 257, Loewus, F., ed. Academic Press, New York London Brown, R.M., Montezinos, D. (1976) Cellulose microfibrits: Visualization of biosynthetic and orienting complexes in association with the plasma membrane. Proc. Natl. Acad. Sci. USA 73, 143-147 Buer, F. (1964) Licht- und elektronenmikroskopische Untersuchungen an Zellw/inden nnd Gallerten einiger Zygnemataceen. Flora 154, 34%375 Davey, M.R., Mathias, R.J. (1979) Close-packing of plasma membrane particles during wall regeneration by isolated higher plant protoplasts - fact or artefact. Protoplasma 100, 85-99 Dawes, C.J. (1965) An ultrastructure study of Spirogyra. J. Phycol. 1,121 127 Fott, B. (1971) Algenkunde. Gustav Fischer, Stuttgart Franke, W.W., Ermen, B. (1969) Negative staining of plant slime cellulose: An examination of the elementary fibril concept. Z. Naturforsch. 246, 918-922 Giddings, T.H., Brower, D.L., Staehelin, L.A. (1980) Visualization of particle complexes in the plasma membrane of Micrasterias denticulata associated with the formation of cellulose fibrils in primary and secondary walls. J. Cell Biol. 84, 327-339
355 Hanna, R.B., C6t4, W.A. (1974) The sub-elementary fibril of plant cell wall cellulose. Cytobiologie 10, 102-116 Herth, W. (1979) The site of,&chitin fibril formation in centric diatoms. II. The chitin forming cytoplasmic structures. J. Ultrastruct. Res. 68, 16-27 Herth, W., Franke, W.W., Bittiger, H., Kuppel, A., Keilich, G. (1974) Alkali-resistant fibrils oft-1,3- and/3-1,4-glucans : structural polysaccharides in the pollen tube wall of Lilium longiflorurn. Cytobiologie 9, 344-357 Herth, W., Kuppel, A., Franke, W.W., Brown, R.M. (1975) The ultrastructure of the scale cellulose from Pleurochrysis scherflfelii under various experimental conditions. Cytobiologie 10, 268-284 Herth, W., Kuppel, A., Schnepf, E. (1977) Chitinous fibrils in the lorica of the flagellate chrysophyte Poteriochromonas stipitata (syn. Ochromonas malhamensis). J. Cell Biol. 73, 311-321 Jordan, E.G. (1970) Ultrastructural aspects of cell wall synthesis in Spirogyra. Protoplasma 69, 40~416 Kiermayer, O., Sleytr, U.B. (1979) Hexagonally ordered "rosettes" of particles in the plasma membrane of Micrasterias denticulata Br~b. and their significance for microfibril formation and orientation. Protoplasma 101, 133-138 Kreger, A.R. (1957) New orientations of cellulose in Spirogyra cell walls. Nature (London) 180, 914 Mueller, S.C. (1982) Cellulose microfibril assembly and orientation in higher plant cells with particular reference to seedlings of Zea mays. In: Cellulose and other natural polymer systems. Biogenesis, structure and degradation, pp. 82104, Brown, R.M., ed. Plenum Press, New York London Mueller, S.C., Brown, R.M. (1980) Evidence for an intramembrane component associated with a cellulose microfibril synthesizing complex in higher plants. J. Cell Biol. 84, 315 326 Mueller, S.C., Brown, R.M., Scott, T.K. (1976) Cellulosic microfibrils : nascent stages of synthesis in a higher plant cell. Science 194, 949-951 Mtihlethaler, K. (1967) Ultrastructure and formation of plant cell walls. Annu. Rev. Plant Physiol. 18, 1-23 Neuhaus, G., Kiermayer, O. (1981) Formation and distribution of cell wall pores in desmids. In: Cytomorphogenesis in plants, pp. 215-228, Kiermayer, O., ed. Springer, Wien New York Noguchi, T., Tanaka, K., Ueda, K (1981) Membrane structure of dictyosomes, large vesicles and plasma membranes in a green alga, Micrasterias crux-melitensis. Cell Struct. Funct. 6, 217-229 Ohad, I., Danon, D. (1964) On the dimensions of cellulose microfibrils. J. Cell Biol. 22, 302-305 Oltmanns, F. (1922) Morphologie und Biologie der Algen. Gustav Fischer, Jena Peng, H.B., Jaffe, L.F. (1976) Cell wall formation in Pelvetia embryos. A freeze-fracture study. Planta 133, 57-71 Preston, R.D. (1974) The physical biology of plant cell walls. Chapman and Hall, London Robenek, H., Peveling, E. (1975) VerS.nderungen des Plasmalemmas w/ihrend der Zellwandregeneration an isolierten Protoplasten aus dem SproBkallus von Skimmia japonica Thunb. Planta 127, 281 284 Robenek, H., Peveling, E. (1977) Ultrastructure of the cell wall regeneration of isolated protoplasts of Skimmia japonica Thunb. Planta 136, 135 145 Schnepf, E., Herth, W. (1978) General and molecular cytology. Prog. Bot. 40, 1-11 Staehelin, L.A., Giddings, T.H. (1982) Membrane-mediated control of cell wall microfibrillar order. Developmental order: its origin and regulation, pp. 133-147. Alan R. Liss, Inc., New York
356 Van den Hoek, C. (1978) Algen. Einfiihrung in die Phykologie. Georg Thieme, Stuttgart Wada, M., Staehelin, L.A. (1981) Freeze-fracture observations on the plasma membrane, the cell wall and the cuticle of growing protonemata of Adiantum capillus-veneris L. Planta 151,462-468 Wilkinson, M.J., Northcote, D.H. (1980) Plasma membrane ultrastructure during plant protoplast plasmolysis, isolation and wall regeneration: a freeze-fracture study. J. Cell Sci. 42, 401-415
W. Herth: Microfibril formation of Spirogyra Willison, J.H.M., Cocking, E.C. (1975) Microfibril synthesis at the surface of tobacco mesophyll protoplasts, a freezeetch study. Protoplasma 84, 147-159 Willison, J.H.M., Klein, A.S. (1982) Cell wall regeneration by protoplasts isolated from higher plants. In: Cellulose and other natural polymer systems. Biogenesis, structure and degradation, pp. 61-86, Brown, R.M., ed. Plenum Press, New York London Received 3 May; accepted 13 June 1983
Note added in proof. Meanwhile a preliminary report describing similar rosette arrays for Spirogyra sp. Utex strain #918 is
in press (Hotchkiss AT, Roberts EM, Itoh T, Brown RM Jr (1983) Microfibril assembly among selected algae of the Zygnematales. J Cell Biol, ASCB abstract)
Planta (1983)159:347-356
9 Springer-Verlag 1983
Arrays of plasma-membrane "rosettes" involved in cellulose microfibril formation of Spirogyra Werner Herth Zellenlehre, Universitfit Heidelberg, Im Neuenheimer Feld 230, D-6900 Heidelberg, Federal Republic of Germany
Abstract. The cell-wall structure and plasma-membrane particle arrangement during cell wall formation of the filamentous chlorophycean alga Spirogyra sp. was investigated with the freeze-fracture technique. The cell wall consists of a thick outer slime layer and a multilayered inner wall with ribbon-like microfibrils. This inner wall shows three differing orientations of microfibrils: random orientation on its outside, followed by axial bundles of parallel microfibrils, and several internal layers of bands of mostly five to six parallel associated microfibrils with transverse to oblique orientation. The extraplasmatic fracture face of the plasma membrane shows microfibril imprints, relatively few particles, and "terminal complexes" arranged in a hexagonal package at the end of the imprint of a microfibril band. The plasmatic fracture face of the plasma membrane is rich in particles. In places, it reveals hexagonal arrays of "rosettes". These rosettes are best demonstrable with the double-replica technique. These findings on rosette arrays of the zygnematacean alga Spirogyra are compared in detail with the published data on the desmidiacean algae Micrasterias and Closterium. Key words: Cellulose microfibril formation - Plasma membrane rosettes - Spirogyra (freeze fracture).
Introduction
Algae have become important model systems in cell-wall research: Valonia has been a classic object for structural and X-ray studies of cellulose I, some algae have been demonstrated to use other structural polysaccharides like fl-l,3-xylan, fl-l,4-manAbbreviations: EF = Extra plasmaticfracture face; PF = Plasmatic fracture face
nan or chitin, and with the scale-forming Pleurochrysis it was shown that cellulose may be formed within the Golgi cisternae (compare reviews by: Preston/974; Brown et al./973; Herth et al. 1977; Herth 1979). For a long time, the cytological site of cellulose formation has been a point of controversal discussions. Several years ago, Preston formulated the hypothesis that cellulose microfibrils are synthesized by enzyme complexes in the plasma membrane (Preston 1974). Then, with the freezefracture technique, a variety of plasma-membraneparticle complexes were supposed to be identical with these postulated cellulose-synthetase enzyme complexes (for survey see Schnepf and Herth /978). The particle arrangements in the alga Oocystis (Brown and Montezinos /976), in Pelvetia embryos (Peng and Jaffe 1976) and in higher-plant protoplasts (e.g. Willison and Cocking /975; Robenek and Peveling 1975,/977) were rather different, so that their identification as cellulose synthetase was often questioned (e.g. Schnepf and Herth /978; Davey and Mathias 1979; Wilkinson and Northcote 1980). A breakthrough came with the freeze-fracture findings on the green alga Micrasterias (Kiermayer and Sleytr 1979; Giddings et al. /980): there, hexagonal arrays of rosettes, with six subunits each, are found on the plasmatic fracture face (PF) of the plasma membrane during the formation of the bands of secondary-wall microfibrils (freeze-fracture terminology according to Branton et al. /975). During primary-wall formation, only individual microfibrils are formed, and only single rosettes were found on the PF of the plasma membrane. The authors concluded that the rosettes are cellulose-synthetase complexes, each forming a 5-nm subunit microfibril, the number of cooperating rosettes in a row arrangement should determine the dimension of the microfibril produced. Mueller and Brown (/980) also demonstrated individual ro-
W. Herth : Microfibril formation of Spirogyra
348
settes on the PF of plasma membranes of corn, bean and pine seedlings, and terminal globules on the extra plasmatic fracture face (EF). These authors then said that the rosettes are only preserved when the tissue is not cryoprotected and extremely rapidly frozen. Meanwhile, further examples for rosettes have been reported: Wada and Staehelin (1981) found individual rosettes on the plasma-membrane PF in the growth region of the protonema of the fern Adiantum, Staehelin and Giddings (1982) found rows of rosettes during primary-wall formation of the green alga Closterium, and arrays of rosettes corresponding to those of Micrasterias during secondary-wall formation of Closteriurn. Rosettes now have also been observed after glutaraldehyde fixation and cryoprotection in higher plants (Mueller 1982) and in Micrasterias (Noguchi et al. 1981). Based on these collective findings, and on observations, mostly unpublished, of other higherplant systems, Mueller (1982) claimed that the terminal complex-rosette structure should be universally present in cellulose-forming higher-plant systems. In algae, however, different types of plasmamembrane particle arrangements involved in cellulose microfibrils seem to have evolved. It will be interesting from the phylogenetic point of view to find out in which organism the first rosettes may have arisen, and to examine the breadth of distribution of rosettes among different algal groups. As a start to such a survey I investigated the wellknown filamentous green alga Spirogyra, which belongs to the Zygnemataceae, a group related to the two documented algal examples with rosettes, Micrasterias and Closterium, which both belong to the Desmidiaceae. The Desmidiaceae and Zygnemataceae both belong to the old group of Conjugatae (Fott 1971; Oltmanns 1922; van den Hock 1978). Materials and methods Culture conditions. Spirogyra sp. was obtained from Professor W. Koch, Culture Collection of Algae, University of G6ttingen, FRG. The algae were cultivated in a medium according to Professor W, Koch consisting of: K N O 3 0.01% (w/v); (NH4)2HPO4 0.001% (w/v); M g S O 4 . 7 H 2 0 0.001% (w/v); i ml saturated CaSO 4 solution in 100 ml; 2 ml soil extract in 100 ml; 0.5 ml stock solution of trace elements in 100 ml; 0.5 ml stock solution of vitamin BI2 in I00 ml. They were kept at 2,000-3,000 lx, 14:10 h light/dark cycle, 19 ~ C in a thermostatically controlled tank (Ruharth, Hannover, FRG). Freeze-fracture. Three different experimental procedures were applied: a) With cryoprotection. The algae were fixed in 2.5% glutaraldehyde, 0.1 M cacodylate buffer, pH 7.2, for 30 rain at
20 ~ C; transferred for 2 h into 20% glycerol with the same buffer, 20 ~ C; then transferred to the specimen holder, and frozen via Freon 22 in liquid nitrogen. b) Without cryoprotection. Bundles of the filamentous algae were collected with forceps, cut into short parallel filaments with a razor blade, immediately transferred to the conventional specimen holder, and frozen via Freon 22 in liquid nitrogen. c) Without cryoprotection, double-replica technique (modified after Giddings et al. 1980). The algae were collected and cut into short filaments as in b), then suspended in water-diluted yeast paste, and a small drop of the algal suspension was then transferred to the double-replica holder. Again the material was frozen via Freon 22 in liquid nitrogen. The specimens then were freeze-fractured in a Balzers (Liechtenstein) BAF 400 apparatus at - 1 0 0 ~ C, and shadowed subsequently with platinum-carbon and carbon under control of the quartz monitor. The replicae were cleaned with 60% (v/v) chromosulfuric acid overnight, washed with distilled water and taken up with 200-mesh copper grids for electron microscopy. The replicae were examined with a model 400 electron microscope (Philips, Eindhoven, The Netherlands). The magnifications were calibrated with a grating replica.
Results
Of the three technical variants applied, the first one, with cryoprotection by glycerol after glutaraldehyde fixation, proved absolutely unsatisfactory because of tremendous vesiculations of the plasma membrane. The two other methods without cryoprotection yielded somewhat differing results as follows.
No cryoprotection, single replica after fracture with the blade. Cross-fractures of the wall show two distinct layers. An outer layer of radial elements showing tree-like branching corresponds to the slime layer which yields the well-known slipperiness of Spirogyra (between black arrows in Fig. 1). The inner layer (between white arrows in Fig. 1) is relatively thin in comparison to the slime layer and contains the cellulosic wall component. In fractures tangential to the cellulosic wall layer, sublayers become evident. The outer layers show randomly oriented microfibrils (Fig. 2 a), followed internally by parallel associated micro fibrils (Fig. 2a, arrows). The innermost layers are composed of bands of laterally associated microfibrils (Fig. 2b). Each band consists of 5-6 parallel ribbon-like microfibrils (between arrows in Fig. 2b) with widths of approx. 20 nm. The various bands show a mostly transverse to oblique orientation, but are rather irregular, whereas the parallel microfibrils of the outer sublayer are oriented parallel to the longitudinal axis of the elongate cell. After freeze-fracture, the EF of the plasma membrane of Spirogyra shows imprints of the microfibrils (Fig. 3 a), and both the sublayer of the parallel microfibrils and a few bands of microfi-
W. Herth: Microfibril formation of Spirogyra
349
Fig. 1. Cross-fracture of Spirogyra sp., no cryoprotection, fractured with the blade. Slime layer (between black arrows) and microfibrillar layers (between white arrows). B a r = 1 gm; x 10,000
brils are distinguishable. Fibrillar ends are not frequent, but at closer examination of such EFs, the fibrillar bands are seen to end in globular "terminal complexes" (terminology according to Mueller et al. 1976; Mueller and Brown 1980) which are closely packed in a hexagonal arrangement (Fig. 3 b, arrows). These terminal complexes are arranged in rows corresponding to the number of microfibrils associated in the band ending there; they consist of three to five globules per row in this figure. The smallest array of globules encountered was three rows of 2, 3 and 3 globules; the biggest is shown in Fig. 4a and consists of five rows of up to four globules per row. This high magnification shows that each globule has a diameter of approx. 24 nm, and a center-to-center spacing of approx. 30 nm. The EF shows relatively few particles (approx. 1,000 gm -2) (Fig. 4b). In contrast to the EF, the PF is rich in particles of various sizes (approx. 2,000-3,000 Bm-2). These particles appear individually or associated with one or more other particles. It is rather difficult to detect the structures corresponding to the terminal globules in this mass of particles. However, once found, the corresponding structures clearly are rosettes of six subunits each which are associated into a hexagonal array (Fig. 4 b, rosettes encircled). As with the terminal globules, they appear in rows of up to five rosettes, and again there are five rows corresponding to the number of microfibrils associated in the band. The rosette diameter is approx. 24 nm, each subunit particle has a diameter of approx. 8 nm.
Double-replica technique. With
this technique, the yield of well-fractured cells was much higher, and the cells showed better preservation and fewer ice crystals, probably due to the faster rate of cooling for the specimens sandwiched between the two metal surfaces (compare corresponding remarks of Mueller 1982). N o t only the plasma-membrane EF, which looked very similar to the figures described above, but also the PF showed clear imprints of microfibrils (Fig. 5a). In one case, the imprint of a band of microfibrils (arrow in Fig. 5 a) clearly is seen to end in an array of rosettes. Several rosette arrays were found in close proximity (Fig. 5a, in circles), whereas large areas of the PF appeared without rosette arrays. The number and arrangement of the rosettes varied from a few linear rows of few rosettes to arrays with six rows of rosettes (Fig. 5b). The rosettes appeared more distinct, as the center-to-center spacing was greater in these preparations, and the space between the rosettes appeared smooth. Under high magnification (Fig. 5 c), the individual rosettes show six subunits, have a diameter of 22.5 nm, and a center-tocenter spacing of 35 nm. In this largest rosette array found yet for Spirogyra, six rows of rosettes are associated. The microfibril imprints leading to this array (not to be seen at this high magnification) come from the top of this figure, indicating that the microfibril band arises from a sequence of 2, 4, 4, 5, 6 and 5 rosettes. In the cell wall, the highest number of microfibrils associated in a band was eight, but most bands had five to six microfibrillar subunits.
350
W. Herth: Microfibril formation of Spirogyra
Fig. 2a, b. Tangential fractures of the cell wall, method as Fig. 1. a A r a n d o m outermost layer of microfibrils (lower part) is followed by bundles of axially oriented microfibrils (arrows). b The internal part of the cell wall consists of bands of five to six parallel oriented microfibrils, with main orientation transverse to oblique to the cell axis. Note one band between the arrows. Shadow angle=encircled arrows; bars = i gm; x 40,000 (a); x 60,000 (b)
W. Herth: Microfibril formation of
Spirogyra
351
Fig. 3 a, b. Plasma membrane of Spirogyra sp., extraplasmatic fracture face (EF). Method as Fig. 1. a Imprints of the parallel bundles of microfibrils (arrowheads) and of a few bands of microfibrils (e.g. at the arrows) are readily discerned on the EF. b At higher magnification, imprints of microfibril bands are seen to terminate in "terminal complexes" (TC; arrows). Shadow angle = encircled arrows; rnfi= microfibril imprint; b a r s - 1 ~m; x 30,000 (a); x 60,000 (b)
352
W. Herth: Microfibril formation of Spirogyra
Fig. 4a, b. High magnifications of the two fracture faces of the plasma membrane of Spirogyra sp., method as Fig. 1. a EF with imprint of microfibrils (follow arrows) ending in a terminal complex consisting of globules arranged in rows. b Plasmatic fracture face (PF) with very dense particle population and a hexagonal array of rosettes (encircled) consisting each of six subunit particles. Shadow angle=encircled arrows; bars=0.2 gm; x 180,000 (a); x 240,000 (b)
W. Herth: Microfibril formation of Spirogyra
353
Fig. 5a-c. PF of the Spirogyra sp. plasma membrane, double-replica technique, a Survey of a region with several rosette arrays (in circles). In one case, a microfibril band imprint ends in a rosette array (arrow). b Higher magnification from a, with two rosette arrays showing different numbers of rosettes arranged in rows (arrows). e High magnification of a big rosette array with almost triangular arrangement of the rosettes (between arrows). Microfibril imprints come from top of the figure. Note the six subunits of the rosettes and the relatively big center-to-center spacing between the rosettes. Shadow angles=encircled arrows; b a r s = 1 gm (a), 0.1 gm (b, e); x 40,000 (a); x 150,000 (b); x 280,000 (e)
354
W. Herth: Microfibrilformation of Spirogyra
Discussion
Spirogyra, the microfibril dimensions in one band
The cell-wall architecture of Spirogyra after freezefracture confirms the earlier publications based on light microscopy and thin-section analysis (Jordan 1970; Dawes 1965; Buer 1964). The slime layer seems to be secreted across the microfibrillar layers; there are no distinct pores as in Micrasterias (for review see Neuhaus and Kiermayer 1981). The microfibrillar layers show a similar arrangement to those of Micrasterias, with randomly arranged individual microfibrils in the primarywall layer and bands of parallel associated microfibrils in the secondary-wall layer (Giddings et al. 1980). In addition, Spirogyra has a layer of axial parallel microfibrils, which seems to be the first layer of secondary-wall formation, which then is followed by the more transversely microfibrillar bands. These parallel microfibrils might also indicate the end of primary-wall formation, since in Closterium the primary wall consists of parallel microfibrils (Staehelin and Giddings 1982). The chemical nature of the wall microfibrils of Spirogyra has been characterized by X-ray diffraction (Kreger 1957; Dawes 1965). They consist of a cellulose modification which is remarkable in that it shows a uniplanar orientation with the crystal plane (101) or (002), according to the species, parallel to the wall surface, instead of (101) parallel to the surface as in Valonia cellulose I. This indicates a certain configurational difference from the typical cellulose I. Comparable data are not yet available for Micrasterias, as this alga has not yet been chemically analysed for cell-wall composition and the X-ray pattern of the microfibrils. The cellulosic nature of the Micrasterias fibrils has only been deduced from staining reactions. Comparison of the freeze-fracture results shows that the terminal globules in Spirogyra plasma-membrane EF are more distinct than in Micrasterias, the PF of the Spirogyra plasma membrane contains more particles, and the rosettes are arranged in smaller arrays (five to six rows of up to five rosettes instead of up to 16 rows with up to 15 rosettes). As in Micrasterias, the rosettes have six subunits, and have almost the same dimensions and the same center-to-center spacing. In Micrasterias, the micro fibril width of the central microfibril of a band is greater than that of the peripheral microfibril, and is assumed to be formed by more rosettes (Giddings et al. 1980; Staehelin and Giddings 1982). Each rosette is believed to form one 5-nm subunit microfibril, the products of the rosettes in one row then laterally associate and crystallize (Giddings et al. 1980). In
are more uniform, in the range of 20 nm. The number of rosettes in one row engaged in microfibril formation is five to six in the central part of the array. This means that one rosette here should synthesize an approx. 4-nm microfibrillar subunit. This value is close to the postulated "elementary fibrils" (Mfihlethaler I967) of 3.5 nm diameter, if the thickness of the metal shadow coat is subtracted from these 4 nm (Ohad and Danon 1964). This would probably indicate that each of the six subunits of a rosette might synthesize six glucan chains which then co-crystallize into a 36-glucanchain "elementary microfibril". The sub-elementary microfibrils of smaller dimensions (e.g. Franke and Ermen 1969; Hanna and C6t~ 1974; Herth et al. 1974, 1975) then might be the result of less than six, or even individual, enzyme particles. It seems that the limitation of the fibrillar width is given by the distance of the rosettes, and not so much by the number of rosettes. If the calculations above are correct, then the Spirogyra band microfibrils should consist only of a single layer of laterally associated 4-nm subunits, whereas in Micrasterias more layers are depicted in the model of Giddings et al. (1980). In this model, Giddings et al. do not draw a flat ribbon for the microfibril made by five to six rosettes, but give a rather isodiametric cross section, whereas in their diagram the broader microfibrils, made by the central ten rosettes, consist of three layers of three- to four-subunit microfibrils. A microfibril made by 16 rosettes should then contain four layers each of four subunits, representing a cross-section of 20 x 20 nm. However, as judged from the equal shadows in their Fig. 3, the micro fibrils of Micrasterias have equal thickness. As in Spirogyra a row of five to six rosettes is already capable of forming a 20-rim microfibril, this seems to indicate that not all the rosettes of a 16-rosette row in Micrasterias seem to be active at the same time. Moreover, in Closterium (Staehelin and Giddings 1982), five to six rosettes in one row make a ribbon of 16.5 nm, a value rather close to Spirogyra, supporting the conclusions discussed above. The random microfibrils of the primary-wall layer of Spirogyra are probably made by individual rosettes as in Micrasterias (Giddings et al. 1980) and in ferns and higher plants (Wada and Staehelin 1981; Mueller 1982), but these rosettes are difficult to detect with the high background of particles on the Spirogyra plasma-membrane PF. The parallel layer of microfibrils may be formed by rows of rosettes, as in Closterium (Staehelin and Giddings 1982), but these stages have not been de-
W. Herth: Microflbril formation of Spirogyra
tected in my freeze-fracture replicas. It is possible that individual particles, indistinguishable from other intramembrane particles, are involved in microfibril formation in these stages. The Spirogyra rosette arrays are intermediate between the rows of rosettes in Closterium primary-wall formation and the larger arrays of rosettes found in desmid secondary-wall formation (Staehelin and Giddings 1982). This might indicate that Spirogyra exhibits a lower status of evolution of the rosette arrays than the desmids. The concept of the rosette as a plasma-membrane-bound multienzyme complex involved in cellulose formation (see Introduction) is supported by the findings with Spirogyra. However, the number of well-documented cases is small and the generality of the rosette concept of cellulose microfibril formation requires further investigation (for critical discussion see e.g. Willison and Klein 1982). The author thanks Professor E. Schnepf and Dr. H.-D. Reiss for valuable discussions, and B. Heck for technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft.
References Branton, D., Bultivant, S., Gilula, N.B., Karnovski, M.J., Moor, H., Mfihlethaler, K., Northcote, D.H., Packer, L., Satir, B., Satir, P., Speth, V., Staehelin, L.A., Steere, R.L., Weinstein, R.S. (1975) Freeze-etching nomenclature. Science 190, 54-56 Brown, R.M., Herth, W., Franke, W.W., Romanovicz, D. (1973) The role of the Golgi apparatus in the biosynthesis and secretion of a cellulosic glycoprotein in Pleurochrysis: A model system for the synthesis and secretion of structural polysaccharides. In: Biogenesis of plant cell wall polysaccharides, pp. 207 257, Loewus, F., ed. Academic Press, New York London Brown, R.M., Montezinos, D. (1976) Cellulose microfibrits: Visualization of biosynthetic and orienting complexes in association with the plasma membrane. Proc. Natl. Acad. Sci. USA 73, 143-147 Buer, F. (1964) Licht- und elektronenmikroskopische Untersuchungen an Zellw/inden nnd Gallerten einiger Zygnemataceen. Flora 154, 34%375 Davey, M.R., Mathias, R.J. (1979) Close-packing of plasma membrane particles during wall regeneration by isolated higher plant protoplasts - fact or artefact. Protoplasma 100, 85-99 Dawes, C.J. (1965) An ultrastructure study of Spirogyra. J. Phycol. 1,121 127 Fott, B. (1971) Algenkunde. Gustav Fischer, Stuttgart Franke, W.W., Ermen, B. (1969) Negative staining of plant slime cellulose: An examination of the elementary fibril concept. Z. Naturforsch. 246, 918-922 Giddings, T.H., Brower, D.L., Staehelin, L.A. (1980) Visualization of particle complexes in the plasma membrane of Micrasterias denticulata associated with the formation of cellulose fibrils in primary and secondary walls. J. Cell Biol. 84, 327-339
355 Hanna, R.B., C6t4, W.A. (1974) The sub-elementary fibril of plant cell wall cellulose. Cytobiologie 10, 102-116 Herth, W. (1979) The site of,&chitin fibril formation in centric diatoms. II. The chitin forming cytoplasmic structures. J. Ultrastruct. Res. 68, 16-27 Herth, W., Franke, W.W., Bittiger, H., Kuppel, A., Keilich, G. (1974) Alkali-resistant fibrils oft-1,3- and/3-1,4-glucans : structural polysaccharides in the pollen tube wall of Lilium longiflorurn. Cytobiologie 9, 344-357 Herth, W., Kuppel, A., Franke, W.W., Brown, R.M. (1975) The ultrastructure of the scale cellulose from Pleurochrysis scherflfelii under various experimental conditions. Cytobiologie 10, 268-284 Herth, W., Kuppel, A., Schnepf, E. (1977) Chitinous fibrils in the lorica of the flagellate chrysophyte Poteriochromonas stipitata (syn. Ochromonas malhamensis). J. Cell Biol. 73, 311-321 Jordan, E.G. (1970) Ultrastructural aspects of cell wall synthesis in Spirogyra. Protoplasma 69, 40~416 Kiermayer, O., Sleytr, U.B. (1979) Hexagonally ordered "rosettes" of particles in the plasma membrane of Micrasterias denticulata Br~b. and their significance for microfibril formation and orientation. Protoplasma 101, 133-138 Kreger, A.R. (1957) New orientations of cellulose in Spirogyra cell walls. Nature (London) 180, 914 Mueller, S.C. (1982) Cellulose microfibril assembly and orientation in higher plant cells with particular reference to seedlings of Zea mays. In: Cellulose and other natural polymer systems. Biogenesis, structure and degradation, pp. 82104, Brown, R.M., ed. Plenum Press, New York London Mueller, S.C., Brown, R.M. (1980) Evidence for an intramembrane component associated with a cellulose microfibril synthesizing complex in higher plants. J. Cell Biol. 84, 315 326 Mueller, S.C., Brown, R.M., Scott, T.K. (1976) Cellulosic microfibrils : nascent stages of synthesis in a higher plant cell. Science 194, 949-951 Mtihlethaler, K. (1967) Ultrastructure and formation of plant cell walls. Annu. Rev. Plant Physiol. 18, 1-23 Neuhaus, G., Kiermayer, O. (1981) Formation and distribution of cell wall pores in desmids. In: Cytomorphogenesis in plants, pp. 215-228, Kiermayer, O., ed. Springer, Wien New York Noguchi, T., Tanaka, K., Ueda, K (1981) Membrane structure of dictyosomes, large vesicles and plasma membranes in a green alga, Micrasterias crux-melitensis. Cell Struct. Funct. 6, 217-229 Ohad, I., Danon, D. (1964) On the dimensions of cellulose microfibrils. J. Cell Biol. 22, 302-305 Oltmanns, F. (1922) Morphologie und Biologie der Algen. Gustav Fischer, Jena Peng, H.B., Jaffe, L.F. (1976) Cell wall formation in Pelvetia embryos. A freeze-fracture study. Planta 133, 57-71 Preston, R.D. (1974) The physical biology of plant cell walls. Chapman and Hall, London Robenek, H., Peveling, E. (1975) VerS.nderungen des Plasmalemmas w/ihrend der Zellwandregeneration an isolierten Protoplasten aus dem SproBkallus von Skimmia japonica Thunb. Planta 127, 281 284 Robenek, H., Peveling, E. (1977) Ultrastructure of the cell wall regeneration of isolated protoplasts of Skimmia japonica Thunb. Planta 136, 135 145 Schnepf, E., Herth, W. (1978) General and molecular cytology. Prog. Bot. 40, 1-11 Staehelin, L.A., Giddings, T.H. (1982) Membrane-mediated control of cell wall microfibrillar order. Developmental order: its origin and regulation, pp. 133-147. Alan R. Liss, Inc., New York
356 Van den Hoek, C. (1978) Algen. Einfiihrung in die Phykologie. Georg Thieme, Stuttgart Wada, M., Staehelin, L.A. (1981) Freeze-fracture observations on the plasma membrane, the cell wall and the cuticle of growing protonemata of Adiantum capillus-veneris L. Planta 151,462-468 Wilkinson, M.J., Northcote, D.H. (1980) Plasma membrane ultrastructure during plant protoplast plasmolysis, isolation and wall regeneration: a freeze-fracture study. J. Cell Sci. 42, 401-415
W. Herth: Microfibril formation of Spirogyra Willison, J.H.M., Cocking, E.C. (1975) Microfibril synthesis at the surface of tobacco mesophyll protoplasts, a freezeetch study. Protoplasma 84, 147-159 Willison, J.H.M., Klein, A.S. (1982) Cell wall regeneration by protoplasts isolated from higher plants. In: Cellulose and other natural polymer systems. Biogenesis, structure and degradation, pp. 61-86, Brown, R.M., ed. Plenum Press, New York London Received 3 May; accepted 13 June 1983
Note added in proof. Meanwhile a preliminary report describing similar rosette arrays for Spirogyra sp. Utex strain #918 is
in press (Hotchkiss AT, Roberts EM, Itoh T, Brown RM Jr (1983) Microfibril assembly among selected algae of the Zygnematales. J Cell Biol, ASCB abstract)
