T-ISSUE AND CELL, 1992 24 (3) 315-334 0 1902 Longman Group UK Ltd.

B. SATIAT-JEUNEMAITRE

SPAT IAL AND TEMPORAL REGULATIONS IN MATRICES: HELIC :OIDAL EXTRACELLULAR COMPARISON BETWEEN PLANT AND ANIMAL SYSTEMS Keywords:

Extracellular matrix, cell wall, hellcold self-assembly, directed assembly

morphogencsis.

cytoskelcton.

microlibrtl~.

ABSTRACT. This paper proposes an overview of the last few years‘ investigations regarding the helicoid formation in cxtraccllular matrices (ECMs). Despite the architectural polymorphism displayed among the layered ECM throughout the living kingdom, helicoidal structures arc often described in ECMs and appear as an optimal mechanical device. Helicoids correspond to complex two-phases composites. formation and regulation of which are still a source of debate. Taking the time-event into consideration, it is clear that helicoid in ECMs arc regulablc structures. On the other hand, analogies with helicoidal formations in cholcstcric liquid crystals strongly support the hypothesis of involvement of self-assembly processes. Therefore the balance between self-assemblies and cell regulation is questioned. By gathering animal and plant data on the topic and by analysing the characteristics of these helicoids m ECMs. it is clear that cells have the necessary machinery to interfere with the self-assembly proceaw in response to physiological or mechanical mechanisms. They arc able to modify the physicochemical conditions outside the plasma membrane, therefore acting on the pattern of selfassembly. Scvcral mechanisms are proposed to explain sudden variations occurrmg m the helicoidal formation with time.

I. The helicoidal arrangement

III. Events possibly involved in helicoid

1. Some fundamentals

formation and regu1ation 1. Self-organization processes 2. Intracellular structures involved in helicoid formation: are microtubules the chief actors? 3. Cellular controls of the self-assemblies

in ECMs in the definition of animals and plant ECMs 2. The polymorphism of the ECM and the helicoidal constancy 3. The molecular characteristics of helicoids 4. Biological advantages of helicoidal construction

Introduction II. Time regulation

The cell is characterised by a set of membraneous compartments. Outside the peripheral membrane, the plasma membrane, cellular activity is still expressed in the formation of the extracellular matrix (ECM). Most of the ECMs are complex systems composed of interwoven polymers synthesised independently that together perform a number of varied functions. These include physical support, protection, storage and release of many organic and inorganic substances.

in the helicoidal ECM

1. The biological clock, a fourth dimension in helicoid formation 2. Temporal organization of helicoidal ECM 3. Flexibility of the morphogenetic rhythm C.N.R.S.. Laboratoire des Biomembranes et Surfaces Cellulaires V&g&ales, Ecole Normale Sup&ieurc. 46 rue d’Ulm, Paris, France. Received

14 February

1992. 315

SATIAT-JEUNEMAITRE

316

The chemical and structural variability displayed by ECMs has intrigued researchers and stimulated investigation into the subcellular and molecular events underlying the ECM architecture, formation and function. The advancing fields of cellular and molecular biology and the sophistication of the techniques are rapidly improving the understanding of ECM metabolism. The identification of the great number and the variety of matrix components in both plant (Roberts, 1989) and animal cells (Bernfield, 1989) are proving constantly that extracellular matrices are more complex structures than previously thought. In support of such biochemical data, analyses of the ECM ultrastructure have been conducted, often revealing a wide range of ordered structures (see reviews in: Bouligand, 1972; Neville and Levy, 1984; Neville, 1988b; Roland et al., 1987). It is surprising that, despite the structural polymorphism displayed among the ECMs regarding the cell type or the state of development, observations have revealed the presence of a common highly organised pattern in most of the ECMs; some fibrillar components are often arranged in arced patterns, characteristic of three-dimensional helicoidal arrangements. Indeed, such helicoidal formation in ECMs appears to have a biological constancy throughout the entire living kingdom. It is thought that helicoids contribute to the establishment of the optimal framework essential for coping with spatial constraints generated in both cells and tissues. However, we do not have a clear idea of how these helicoids occur. Furthermore, it is clear that the helicoidal pattern itself varies from one ECM to another. The events involved in this process of helicoid variation are not fully understood. This paper provides a short overview within the framework of helicoidal arrangements in ECMs and questions the potentiality of a cellular regulation of helicoids. It takes into consideration both the animal and plant

Fig. 1. Comparison between ECM in animal and plant cells. In both cases, ECM results from cellular activities expressed outside the plasma membrane. However in animal cells the ECM often covers several cells. In contrast, in plant cells, ECMs (cell wall) are in fact a cellular compartment.

data on the spatial and temporal regulation of helicoids in ECM and thus emphasises the specificity and common characteristics of various helicoidal formations. In short, all the aspects of helicoid morphogenesis in ECM: the various expressions, the modulation and remodeling in time, the influence/involvement of extracellular and intracellular events on helicoid formation, will be discussed. I. The Helicoidal Arrangements

in ECMs

1.1. Some fundamentals in the definition of animal ECM and plant ECM

In order to understand the concept of ECM as described in this paper one must keep in mind certain fundamental concepts.

Fig. 2. Polylamellatcd aspect of fibrillar ECM. (A) Elytral outer exocuticle of scarabeid beetle (from Neville and Caveney, 1979). (B) Endocarp cell wall in hazelnut shell (courtesy of J. C. Roland); the zonation is due to an oblique section in a helicoidal arrangement. A x 12,900: B x11.000. Fig. 3. Distortion of the helicoidal texture. (A) Around transversely sectioned channels within the chorion of a moth (Smith ef al.. 1971). (B) Around pit fields in plum pit sclerocytes ECM (courtesy of J. C. Roland). A x40.000; B ~36,000.

..

SATIAT-JEUNEMAITRE

31s

As shown in Figure 1, in animal cells ECM is truly an extra-cellular product entirely outside the cell. This matrix is distributed over multiple cells and it can integrate cellular events into results at the tissue level. In contrast, in plants, the concept of ECM is slightly different: the ECM represents the cell wall and is in fact a special subcellular compartment outside the plasma membrane, limited in a ‘cellular’ space by a cuticle or a middle lamella. Hence it would be more appropriate to talk about an extracytoplasmic matrix rather than an extracellular one. In other words, in plants, ECM remains a cellular entity whereas in animals ECM becomes a tissue entity. Nevertheless one is justified in grouping such histologically different ECM since they are both ‘extramembrane compartments’ secreted by a cellular process, showing common physiological and architectural features. 1.2. The polymorphism

of the ECM and the

helicoidal constancy

In most cases, ECMs exhibit a layered aspect (Fig. 2). This zonation varies according to the organism, the species type, and the developmental stage of the observed tissue, revealing a wide polymorphism in ECM architecture. Although in some cases the observed layering is directly related to a chemical zonation (see for example Ryser et al., 1983), a closer look at the ultrastructure of many layered ECM shows that the layering is related to specific changes in the orientation of the fibrillar components. For example, the alternation of dark and light layers observed in the beetle cuticle (Fig. 2A) is a result of the differences in orientation of chitin polymers (for discussion see Neville and Caveney, 1969). Similarly, the layering observed in some plant ECMs (Fig. 2B) is not the cause of a chemical stratification, but the result of permanent change in the orientation of the cellulose polymers. These examples outline that the architectural polymorphism depicted throughout most of the ECMs often corresponds to distinct three

dimensional arrangements of the fibrillar components. However, in spite of such polymorphism, it is quite common, at the electron microscope level, to detect in each ECM, fibrillar components arranged in characteristic arced patterns, even if in a thin area of the ECM. (Fig. 4). The regularity of these arced patterns is more or less altered when in presence of defects in the texture (Fig. 3) (Mazur et al., 1982). When tilting the section, they can be inverted (Fig. 8A, B), revealing a specific three-dimensional arrangement in the ECM. The first three-dimensional interpretation of the arced patterns as helicoidal arrangements has been proposed by Bouligand in crab cuticle by analogy with similar patterns occurring in certain non-biological systems, such as in cholesteric liquid crystals (Bouligand, 1965). Since then, this analogy has been applied to other materials (Neville and Caveney, 1969; Neville and Luke, 1971; Bouligand, 1972, 1976; Livolant et al., 1978; Mazur et al., 1982; Reis, 1987). Thus the fundamentals of this analogy deserve to be described here. In the physics of liquid crystals, cholesteric assembly corresponds to a peculiar molecular arrangement in space (Fig. 5). Here, a helicoid is formed by successive molecular planes in which the subunits run in parallel to each other but with a rotation of the orientation from one plane to the next. It is an oblique section through this twisted pile of molecules which gives rise to the illusion of arced patterns in sections. In biological ECMs, fibrils instead of molecules exhibit the similar spatial pattern, giving an illusion of arced structures when observed in oblique section (compare Fig. 4 and Fig. 5). This feature of helicoidal arrangement can be seen directly by cryofixation and freeze fracture (Livolant and Bouligand, 1989; Leforestier and Livolant, 1991). Stereo-views of such helicoids in plant ECM after rapid freeze-deep etch techniques (Satiat-Jeunemaitre et al., 1992) also show clearly this regular and progressive change in orientation of fibrils from one sheet to the next (Fig. 6).

Fig. 4. Helicoids in ECMs. The arced patterns are similar in plant and animal ECMs. (A) Epithelial side of a moth chorion (Smith ef al.. 1971). (B) Plum pit sclerocyte (courtesy of I. C. Roland). A x48,ooO; B ~45,000.

320

SATIAT-JEUNEMAITRE

j[

Fig 5. Three-dimensional reconstitution of the helicoidal arrangement in liquid crystals (left side) and in ECMs (right side). In contrast to cholesteric liquid crystals, helicoid formation in ECM follows a temporal vector. The clocks on the right side of the drawing illustrate this rotative movement of the macromolecules with time.

Since their first interpretation, helicoids have been described as a common form of regular ECM architecture. They are expressed (i) continousiy; displaying either a monotonous helicoidal pattern, or a variability in the successive helicoidal pitch (SatiatJeunemaitre and Mosiniak, 1992), where they constitute the major part of the ECM; (ii) displaying either one monotonous heficoidal area inserted in the ECM (Filshie and Smith, 1980; Smith et al., 1971), or intermittent expression (Neville and Luke, 1969; Roland et al. , 1987)) where they form one or more small areas, or (iii) as a part of a complete rotation (Neville and Luke, 1969; Zelazny and Neville, 1972b; Roland and

Mosiniak, 1983). Examples are legion, and the following are cases of arced patterns in ECM which have been described in the literature duing the past few years, and which illustrate this helicoidal constancy independently of the tissue type or the living system. In animal cells: ecdysial membrane of moth antenna (Keil and Steiner, 1990), lobster innermost procuticle (Goudeau et al. , 1990), basement membrane in Peripatus integument (Wright and Luke, 1989), stroma of the rat cornea (Velasco and Hidalgo, 1988), Tenebrio molitor larval endocuticule (Quennedey and Quennedey, 1990) or adult cuticle (Lemoine et al., 1990), and see also Figure 10 in Binnington and Barrett (1988)

Fig. 6. Stereo-pair of replicas after freeze fracture/deep etch of plant epidermis showing the polylamellated organisation of the ECM with a shifting of fibrils from one layer to the next, a feature of helicoidal ECM (see also Satiat-Jeunemaitre er al., 1992). x82,ooO. Fig. 7. Immunostaining of microtubules in mung bean epidermis cells showing various aspect of helical pattern around the cell. (A) Oblique microtubules. (B) Flat helix gives rise to transverse microtubule arrays. (C) Stretched helix gives appearance of longitudinal microtubules. Such helical variation was thought to be related to the helicoidal ECM variations (see text and Satiat-Jeunemaitre, 1989, 1991, for further discussion). A x720; B x650; C x500.

322

where arced patterns in bowfly procuticle are outlined by a cytochemical reaction (for other references see Neville, 1967, 1975; Bouligand, 1978; Mazur et al., 1982). Inplant cells: growing epidermis (Roland et al., 1987), wood vessels (Satiat-Jeunemaitre, 1986), sclerocytes (Satiat-Jeunemaitre and Mosiniak, 1992), collenchyma cells (Roland, 1981), aerenchyma cells (Mosiniak and Roland, 1985), seed mucilage (Abeysekera and Willison, 1990), green alga spores (Mizuta and Wada, 1982), alga embryos (Peng and Jaffe, 1976), and fungal spores (Bonfante-Faso10 et al., 1984) (for other references see also Neville and Levy, 1984; Neville, 1988b, 1988~; Roland et al., 1987). One common feature of the helicoidal system both in animals and plants is that they show an extensive range of pitch which reflects a wide variation of angle between successive planes. In fact, modeling of the helicoidal pattern has shown that most of the fibrillar ECMs can be interpreted as variations around a basic helicoidal structural model, by simply modifying the rotation angle (Neville and Levy, 1984, 1985; Roland and Mosiniak, 1983; Roland et al., 1987; Levy, 1987). However the mechanisms leading to such changes in orientation are unfortunately still unknown. 1.3. The molecular characteristics of helicoids

Studies of the chemistry of the helicoids have contributed to the understanding of the structural constancy of helicoids in ECM and, further, have helped to partially solve the mechanisms of helicoid formation. The molecular characteristics of helicoids, described previously (Neville, 1985, 1986, 1988a; Roland et al., 1989) are summarised here. Helicoids are complex composite assemblies, usually comprised of linear polymers which are arranged in fibres and embedded in a different linear polymeric mixture. The fibrous polymers (for example, keratin-like proteins, chitin, cellulose or collagen) form the skeleton of the helicoids which constitutes the crystalline component of the extracellular matrix. The skeleton’s surrounding medium forms the amorphous phase of the system, and is made up of the linear polymers that are more flexible than the crystalline component, (i.e. proteo-

SATIAT-JEUNEMAITRE

glycans, xylans, xyloglucans). The association of the fibrous component and its embedding polymers covers various chemical combinations in the composite: helicoids are either composed of an homo eneous class of polymers (e.g. proteins Pproteins, or polysaccharides/polysaccharides), or an heterogeneous class of polymers (e.g. proteins/ polysaccharides/proteins or polysaccharides). Hence, it is clear that arrangements of helicoids are not dependent on one molecular type of ECMs. Analysis of all the common molecular features of these helicoids emphasise the similarities in the polymeric backbone shape. Some basic architectural principles common to all helicoids emerge which apply to ‘cholesteric molecules’ in ECM. These (i) should have a stiff ‘backbone’, (ii) should present a molecular assymetry and (iii) possess bulky side chains (Neville, 1986). In uivo, the exact molecular interactions in biological helicoids are not well known. However, it is feasible that possible variations in polymer quality and number can modify the molecular interactions in the composite structure and hence the supramolecular organisation of the ECM, which would then explain the variation of the helicoidal pitch. In addition, the main components of the ECM skeletons ( polysaccharides and proteoglycans) share the potential to be diverse by being able to change only a small radical within their chain. A wide structural diversity among proteoglycans and polysaccharides could also explain the polymorphism often observed in the ECMs. 1.4. Biological advantages of helicoidal construction

The presence of helicoidal structure in ECM as a general feature of organisms and cells appears to indicate some mechanical or physiological importance which may be related to specific functions. The more obvious functions of ECMs are indeed mechanical ones. In fact, ECMs have to assume the function of the hydrostatic skeleton of the tissue, i.e. it has a central role in the shape, extension and movement of the body/organ. It also controls and defines the organisation or deformation of the cell/ tissue. The ‘multiplywood construction’ found in

REGULATIONS

IN HELICOIDAL

EXTRACELLULAR

helicoidal systems gives specific mechanical properties to the ECM. As the rigid skeleton elements are regularly orientated in all directions, the ECM displays an equal resistance wherever the direction of pressure may be. Hence, when forces are transmitted through the hydraulic skeleton (incompressible fluid space in the ECM) to a fibrillar architecture that is helicoidally arranged, the pressure will be distributed in all directions, minimising the deformation. Also if one considers not only the fibrillar component of the skeleton, but also the embedding polymers as well, then ultimately helicoids consist of a combination of a substance of high tensile strength with low elasticity (cellulose/chitin for example) and another of lower tensile strength and greater elasticity (hemicellulose, or proteoglycans). The importance of such a ‘two phases’ system is well known in mechanics, for it forms stronger materials than bulk or pure samples, and supports more effectively the cracks and the defects. In this way, animal cuticles and plant cell walls have both been compared to fibrous composites like fibre-glass (Neville, 1986; Roland et al., 1989). It is these precise properties which emphasise the helicoidal construction as the optimal mechanical device both when plastic areas are required, and when multidirectional forces are applied to the ECM. In addition, helicoids could act as articular surface (e.g. in transition zone between two areas varying by their texture) and could be useful in each ECM where structures should be flexible and highly resistant. Besides these mechanical interests, it is necessary to outline here the physiological importance of ordered constructions in ECMs. There has been a great deal of evidence to suggest that ECM and cell have physiological functions of reciprocity. In animal cells, such reciprocity effects are often mentioned (Bernfield, 1989) and in plant cells it is now known that some cell wall fragments may act as hormones or defence signals for the cell (Ryan and Farmer, 1991; Aldington et al., 1991). It is clear that the lack of membranous compartments in ECM which would be required to organise the signal recognition and transmission, has to be replaced by a specific molecular organisation. Thus, helicoids are probably an expression of such a complex molecular organisation,

MATRICES

33

for any small change in the molecular interconnections could be a message to instruct the cell or tissue of environmental variations. II. Time Regulation

in the Helicoidal ECM

II.1. The biological clock, a fourth dimension in helicoidal formation

As mentioned earlier. helicoids in ECM have often been compared to cholesteric assemblies in liquid crystals, because of the similar molecular pattern achieved. However, several essential differences between the two systems are apparent and must be taken into account when studying the mechanisms of biological helicoid formation. One of these differences deserves particular attention as it adds a fourth dimension to the helicoids: in ECMs, the helicoid formation follows a temporal vector (Fig. 5). In cholesteric liquid crystals, the molecules are simultaneously arranged in a helicoidal manner, as soon as the required physicochemical conditions are achieved. Contrarily, in biological helicoids, the molecular components are progressively apposed to the pre-existing structure, helicoids are growing with time. Therefore a helicoidal ECM can be described as the result of a continuous movement where periodic and regular shiftings of the fibrils allow a measurement of time. Such a system recalls the concept of a biological clock (Bunning, 1973; Millet and Manachere, 1983; Roland et al., 1983: Sweeney, 1987) suggesting that such an expressed periodicity in ECM formation could allow the cell to measure time, and to even organise itself with respect to external periodical variations. It also supposes the existence of a cellular mechanism which is able to regulate the formation of the ECM. Indeed, a relationship between a biological clock and the molecular orientation in ECMs has already been suggested (Neville, 1965). Since then, analyses of spontaneous or experimentally induced variations in helicoidal ECM morphogenesis have been performed in an attempt to characterise the temporal organisation of helicoidal ECM. 11.2. Temporal organisation of helicoidal ECM

In view of the distinctive temporal organisation, characteristic of the biological helicoids. helicoidal areas of ECM can be

SATIAT-JEUNEMAITRE

324

described as the expression of a rhythmic movement, i.e. a repetitive process allowing measurement of the motion of fibrous polymers in time. The periodicity of the rhythm corresponds to a complete rotation of microfibrils, that is to the pitch of the helicoid (Fig. 5). In other words, the period corresponds to the time taken to deposit a series of two arced patterns seen in oblique sections. This rhythm has been characterized as endogenous, since the successive fibrillar shifts are expressed in constant growing conditions: for example, the formation of a helicoidal area during the night in the insect cuticle (Neville and Luke, 1969)) or under continuous culture conditions for the plant epidermis (Roland et al., 1975; Satiat-Jeunemaitre, 1984a). In some cases, this rhythm needs to be triggered before it can run freely. Such triggering signals are different from one case to another: for instance one could be the passage from light to dark in the case of insect cuticles (based on Neville and Luke, 1969), or a light threshold or an ethylene signal in the case of parenchyma cells of maize (Satiat-Jeunemaitre, 1984a, 1984b). These helicoidal shifts can also be stopped, or strongly retarded, and here again the stop signal will vary with respect to the ECM considered: for example, passage from dark to light in insect cuticle (based on Neville and Luke, 1969), ethylene (Vian et al., 1982) or lanthanum in some epidermal cells (Satiat-Jeunemaitre, 1987b). Therefore we can assume that in helicoid formation instead of their being a direct relationship between a given external factor and the biological clock, a cascade of events occur in the signal transduction of the primary signals to the final response in helicoid formation (Satiat-Jeunemaitre, 1987b). To define more accurately the processes involved in ECM morphogenesis, the time component during helicoid formation has to be evaluated. A survey of the literature shows that the length of this period varies according to the ECM considered.

In cuticle of some insects, several helicoidal layers are deposited during the night, i.e. approximately in 12 hr (Neville and Luke, 1969; see Zelazny and Neville, 1972a, for discussion on the timing of endocuticle layer deposition). Consequently the period of the helicoid should be less than 12 hr, probably between 2 and 6 hr, according to the exact numbers of layers deposited. In plant cell walls, observations have shown that the periodicity of helicoid formation varies from 4-6 hr to 24 hr (Reis et al., 1985; Vian and Roland, 1987; Satiat-Jeunemaitre and Mosiniak, 1992). Such different values question the possible existence of various morphogenetic rhythms or perhaps an unique one which could be modifiable. Clearly at this stage insufficient data is available to draw firm conclusions and one can therefore only speculate. However, from plant cell wall observations, it is probable that the oscillatory processes observed are fundamentally ultradian (period of a few hours). An adjustment of this endogenous rhythm by external factors (having a circadian activity) could modulate the initial value of the period and masks the ultradian character of the morphogenetic rhythm (SatiatJeunemaitre and Mosiniak, 1992). Such properties will be of importance when looking for the cellular oscillators which may be involved in ECM morphogenesis. 11.3. Flexibility of the morphogenetic rhythm

The periodicity of helicoids is tissue specific. However, looking at the variability of the helicoidal expression both in animal and plant ECMs, several cases of rhythm modifications have been reported in the literature: for example, in the basement membrane of Peripatus integument, the helicoidal pattern is regularly interrupted and regional modifications in the helicoidal pitch occur (Wright and Luke, 1989); in plum pit sclerocytes, the helicoidal pitch also varies, according to the

Fig. 8. Spontaneous variations of helicoid pitch in ECMs: the case of plum pit sclerocytes. (A) The arced patterns in sclerocyte ECMs show various width suggesting some modulation of the helicoid morphogenesis. (B) Inversion of the arced patterns (*) when tilting the section demonstrates the helicoidal arrangement of the ECM components. (C) In another sclerocyte, the fibrillar shifts have been temporarily blocked (left side of the micrographs) before to process again in an more or less regular manner as in Figure 8A (see also Satiat-Jeunemaitre and Mosiniak, 1992). A x40.000: B x40.000; C x42.000.

SATIAT-JEUNEMAITRE

state of cell differentiation (Figs SA, C) and, as for other plant ECMs, the rhythm appears inconstant (Harche, 1986, 1990; Satiat-Jeunemaitre and Mosiniak, 1992). Such spontaneous variations which presumably occur to accommodate certain constraints, suggest that the cell is able to modify helicoidal shifts and regulate helicoidal ECM formation. This hypothesis is reinforced when taking into account the experimental data both in animal and plant cells on helicoidal ECM. In plants, the morphogenetic rhythm involved in helicoid formation can be accelerated, deaccelerated, blocked, or halted temporarily (Roland et al., 1987 and Satiat-Jeunemaitre, 1987b for reviews). In animals, the chitin orientation changes can be induced as well during the deposition of insect cuticle (Neville, 1967): If locusts are grown in constant conditions, the endocuticle no longer shows helicoidal layering. Interestingly, however, Neville reports that, in some regions of the skeleton, helicoids still occur, showing that the helicoid metabolism may vary according to the cell condition. In short it appears that, contrary to the situation in cholesteric liquid crystals, helicoids in ECM are probably a result of an unitary morphogenetic process that is regulated by the cell. Furthermore if one is to elucidate cellular mechanisms possibly involved in helicoid formation or regulation, we must search for those which may reflect the rapid change of microfibril rotation, the flexibility and the sensitivity of the system.

III. Events Possibly Involved in Helicoid Formation and Regulation

In such helicoidally arranged composites the components must be assembled in an ordered fashion. The vexing question is to understand how this assembly is dictated and eventually regulated. It is well known that some steps of ECM formation are not controlled by the living protoplasm but result from selfassembly mechanisms (for instance, formation of microfibrils, control of microfibril diameter in the crystallisation process). Are helicoids just another example of a selfassembly process resulting from intrinsic molecular properties of their components, or is there a cellular system able to transform the apparent continuous efflux of cellular

products into versatile ture?

and periodic

struc-

III. 1. Self-organization processes In non-biological systems as in cholesteric liquid crystals, helicoid formation is the result of a self-assembly process due to the intrinsic properties of the molecules (Gray, 1962; Bouligand, 1972; Mazur et al., 1982). In appropriate conditions, the molecules are able to self-assemble in a twisted pile, i.e. able to organise themselves without any energy requirements or enzyme actions. The twisting is provoked by a combination of molecule shapes, the temperature, the pressure and the forces developed at the surface of the system. Because some animal ECMs present the same twisted ply structure, and sometimes some distance away from the cell surface, it has been suggested that helicoids in ECMs could arise by extracellular selfassembly of molecular components (Neville and Caveney, 1969; Smith et al., 1971; Neville and Luke, 1971). This hypothesis has since been overtaken for plant ECMs as well (Bouligand, 1972; Neville and Levy, 1985; Roland et al., 1987; Abeysekera and Willison, 1990). Application to helicoids in ECMs.

Recent findings indicate that some of the physical requirements established for helicoid formation in liquid crystals are in some ways achieved in biological ECMs. Firstly, a prime reason for self-organization is due to ECM components being confined to a narrow gap between plasma membrane and an external rigid layer (epicuticle, middle lamella, another membrane). Such trapping between more or less parallel and rigid barriers gives specific dynamic behaviour, fibrils in viscous medium will tend to organize in strata parallel to the constraining layers. In a similar way, in vitro reconstitution of helicoidal arrangements has been successively performed only when the medium was trapped between two glass sheets (Reis, personal communication). In vivo, the necessity of such constraining layers to achieve a regular helicoid has recently been emphasized (Neville, 1988~). Secondly, helicoids similar to those described in cholesteric liquid crystals have been successfully achieved in vitro with biological polymers. For example, DNAs,

REGULATIONS

IN HELICOIDAL

EXTRACELLULAR

RNAs, polysaccharides such as xanthan or hydroxypropylcellulose, and polypeptides such as poly-y-ethyl-glutamate have shown their ability to self-assemble in helicoidal, cholesteric assembly (Maret et al., 1981; Werbowyj and Gray, 1984; Livolant and Bouligand, 1989 and references therein; Livolant, 1991). Thirdly, the fluid state of skeleton. polymers (cellulose) displaying helicoidal pattern has now been described in vivo (Abeysekera and Willison, 1990; Roland et al., 1989), supporting the analogy with the liquid state of cholesteric mesophases. The working hypothesis is that most of ECMs go through a liquid phase, even if a brief one (Bouligand, fY72: Favard and Bouligand, 1980). Finally, analyses of some ECM embedding polymers such as hemicelluloses show clearly that they have the cholesteric requirements: their molecular linear backbone has assymetric side chains (Neville, 198.5; Roland et al., 1989). Some in vitro assays with purified fractions of hemicelluloses show a clear tendency to self-organize (Reis, 1978, 1987). Furthermore, in situ localizations of ECM components often show an accumulation of hemicelluloses where helicoidal shifts of the skeleton polymers are observed (Vian et al., 1986). However, the fibrous components of ECMs as cellulose or chitin do not have the required characteristics to be cholesteric. Although the suggestion made more than 20 years ago from animal ECM that the embedding polymers act as orientating agent on the skeleton (Neville. 1967), it is actually in plant cells that recent experimental data on helicoidal ECM have supported such an hypothesis. They are tightly associated with skeleton fibers and. because of their cholesteric properties, hemicelluloses are believed to be responsible for manipulating cellulose microfibrils into helicoidal arrays (Vian etal.. 1986; Neville, 1985, 1988a). Finally, at this stage of investigation, it could be assumed that helicoids in ECM would result essentially from planar geometrical confinement of microfibrillar elements embedding in cholesteric molecules. Thespecificity ofbiological ECMs. However, although the physical analogy is convincing enough to admit the occurrence of selfassembly processes in the formation of helicoidal ECM, distinct or additional formation

MATRICES

377

mechanisms are suggested when regarding the biological characteristics of ECMs: a) Helicoid growth is time dependent. and displays a wide variation of its pitch. suggesting an adaptative mechanism, as seen in section II. b) Helicoids result from a heteromolecular complex of biological polymers instead of the homogeneous one described in the in vitro experiments or in liquid crystals, suggesting more complex molecular interactions. Moreover. in ECM helicoid components co-exist with other non-structural components (enzymes, ions, secondary products, other polymers such as pectins etc.), and the numerous covalent linkages and hydrogen bonds occurring between the helicoid and the other ECM components have to be considered in the mechanisms of the threedimensional pattern construction as they constitute a target for cellular regulation. c) The helicoids are irreversible structures: they are either consolidated (often by the impregnation of secondary products) or temporarily seen and finally destroyed (as in the instance of the primary wall in plants): in all cases they never act as reversible structures as in liquid crystals where the cholesteric assembly is not a stable construction but a reversible equilibrium state. d) In ECM the molecular pattern is fixed between two biological surfaces (plasma membrane and cuticle for example. or plasma membrane and middle lamella.); therefore they could be under the influence of signals emerging from either or both surfaces. e) ECM helicoids are made by long macromolecules, the lengths of which are actually quasi-infinite (Roland etal., 1989). extending over the whole plant cell/animal tissue. Whatever the architectural patterns described, they are often expressed with a constant aspect around the cell or over the body, suggesting a cellular/tissue synchronisation. For example the pattern observed in Figure 8 is the same all around the cell, and is described in neighboring cells as well. suggesting a controlled variation in the periodicity of the helicoids. Such synchronisation is also suggested when, in plant tissues, a quasi-symmetric pattern of irregular helicoids is seen each side of the middle lamellae (Mosiniak and Roland. lY85:

SATIAT-JEUNEMAITRE

328

Roland et al., 1987; see also discussion Leitch, 1989).

in

The above points demonstrate that the self-assembly processes possibly occurring in ECMs must in some way be controlled by the cell. 111.2. Intracellular structures involved in helicoid formation: are microtubules the chief actors?

Since organized ECMs have been described, the question of the cellular control has been asked, and various hypothesis have arisen for animal ECMs (see Neville, 1967, and WeisFogh, 1970, for reviews) as in plant ECMs (Roland et al., 1975; Wardrop et al., 1979; Preston, 1982; Staehelin and Giddings, 1982). A survey of the recent literature on this topic demonstrates that progress has been slow, over the last few years. However it is in plant cells that the topic has been most extensively investigated, probably stimulated by the observations of two subcellular structures which seem to be related to the ECM formation. Firstly, the microtubules, because they have often been seen underlying the cellulose microfibrils on the other side of the plasma membrane, are thought to be involved in the control of microfibril orientation. A second exciting element for plant biologists, which does not appear in animal cells, has been the observation of terminal rosette-shaped complexes associated to the membrane (Herth, 1985; Quader, 1986; Giddings and Staehelin, 1988): these have been identified as the synthetases responsible for glucan chain synthesis, able to move between the plasma membrane sheets therefore acting on the cellulose orientation (Delmer, 1987; Seitz and Emmerling, 1989). Although these complexes have not yet been isolated, they may help to explain a direct functional relationship between microtubules in cytoplasm, and microfibrils in extracellular compartments. The legacy of debates regarding this hypothesis of a direct control of microtubules on microfibril orientation via cellulose symthetase complexes will not be reported here (see Robinson and Quader, 1982; Gunning and Hardham, 1982; Heath and Seagull, 1982; Emons, 1982; Lloyd and Wells, 1985; Traas et al., 1985; Herth, 1985; Lloyd, 1987; Preston, 1988; Satiat-Jeunemaitre, 1989, 1991; Derksen et al., 1990).

However, as microtubule control has been proposed as well to explain helicoid formation (Roberts et al., 1985), it seems appropriate to further evaluate this hypothesis, taking into account the helicoid characteristic described previously in this paper. In plant cells displaying helicoidal ECM, microtubules are organised in helices around the cells (Fig. 7), sometimes corresponding to inner wall fibrils (Lloyd and Wells, 1985). The helix pitches vary, spontaneously or under experimental conditions, and the changes occurring in fibril orientation have been related to the dynamics of these microtubular helices (Lloyd and Seagull, 1985; Roberts et al., 1985; Satiat-Jeunemaitre, 1991). However, if the microtubules are directly responsible for helicoid construction they must comply with the previously established facts; i.e. they should mirror the constant change direction, and achieve rotation of 360” in a few hours. Therefore, microtubules have to go through many depolymerisation/repolymerisation cycles, or, if microtubule changes are due to stretching/flattening of the microtubular helices, microtubular helices have to inverse their polarity to follow a 360” rotation achieved by cellulose fibrils (see Satiat-Jeunemaitre 1989, 1991 for further discussions). These two assessments seem unlikely. Other arguments questioning the validity of the hypothesis in the case of helicoidal ECM can be made. Firstly, a careful three-dimensional analysis of the ECM ultrastructure shows that the helicoidal pattern is not cancelled after inhibiting microtubule polymerisation, but merely slowed down (Vian et al., 1982; Satiat-Jeunemaitre, 1984b). Secondly, a descriptive study of microtubule/microfibril shifts each side of the plasma membrane in cells presenting helicoidal ECM shows that the variation of microtubular helices appears to be related to variation in cell polarity and do not mirror the shifts observed in helicoidal ECM (Satiat-Jeunemaitre, 1991). Thirdly, helicoid formation in plant ECM has been reported to occur without microtubules (Mizuta and Wada, 1982), or with no correlation between microtubules and microfibrils (Mizuta et al., 1989). In animal cells, large numbers of microtubules may run perpendicularly to the surface plane and therefore the helicoidal

REGULATIONS

IN HELICOIDAL

EXTRACELLULAR

surface, countering the hypothesis of a structural link between microtubules, plasma membrane and ECM fibrils. Their specific orientation is related to the strengthening of the muscle insertion on the tegument, and they are probably involved in the transmission of tension from contractile cell to exoskeleton (Smith et al., 1969; Neville, 1975). Microtubules are not thought to be directly involved in the organisation of ECMs, and no relation with chitin synthetase have been shown, even indirectly. Finally, from most animal data and in some instances in plants, it is clear that microtubules are not necessarily involved in the helicoid formation. The exact nature of the relationship between microtubules and microfibrils in plant cells displaying helicoidal ECMs is still an open question. i11.3. Cellular controls on self-assembly Broadly speaking, the cell could act on ECM ultrastructure either by (i) changing the physice-chemical characteristics of the extracellular compartment and where they modify the conditions of self-assembly processes, or (ii) by altering the synthesis of the ECM components. and where they modify the mutual interactions between ECM components, or (iii) by acting on the cross-linkages between the helicoid components, where they modify the ECM stability and properties. Modifications of the physicochemical environment in the extracellular compartment. As mentioned before, self-organ-

isation of molecules depends on the characteristics of the surrounding medium. Temperature, pH. ion concentration, pressure and various forces are some of the parameters involved in the construction of the organised phases of liquid crystals, and any variation of them could affect the final molecular construction (Gray, 1962; Mazur et al.. 1982). Consequently, by changing such parameters just outside the plasma membrane, where the self-assembly processes are supposed to occur, cells could change the conditions of cholesteric assembly and then act on the orientation of the skeleton microfibrils. On the plasma membrane for instance, the differential activity of proton pumps, ionic pumps, or ionic channels could play the role of regulating agents,

MATRICES

329

which will modulate the pattern of helicoidal construction over time by adjusting the pH or the ionic concentration of the extracellular medium close to the plasma membrane. Considering pressure variations, it is well known that ECMs have to face a wide range of pressure over time: for instance between two ecdyses in insect growth, fibrillar arrangements in the forming cuticle are under an increasing pressure, and go through a quite sudden release of tension when the old integument is shed. It would not be surprising if such changes in pressure could play an essential role in the behaviour of fibrillar components outside the plasma membrane. Similarly, in plants, changes in fibril orientation could be due to variation in turgor pressure inside the cells. Actually, a decrease of osmotic pressure has been shown to alter helicoidal ECM formation (Reis etal., 1985). Finally, it is clear that the cell is able to adjust the self-assembly processes, and an abrupt change in the helicoidal pattern (or the induction/arrest of the helicoidal shifting) could mirror some drastic variation in the physico-chemical conditions of the extracellular medium. Modijications of the synthesis of the helicoidal complex. Skeleton fibres are polymerized

and crystallised (fibrillogenesis) outside the plasma membrane both in plant (Reis, 1987: Bolwell, 1988) and animal systems (Neville, 1967). Conversely, embedding polymers are synthesised intracellularly and have to be transported from the Golgi apparatus to the cell surface in Golgi vesicles eventually fusing with the plasma membrane. Obviously, the formation of the helicoidal composite outside the plasma membrane will depend on the conjunction between these two synthetic pathways, and any regulating process acting on one of these pathways will also act on helicoid formation. Considering first skeleton synthesis; experimental data demonstrates that the formation of the helicoidal composite is altered when skeleton synthesis is inhibited (Satiat-Jeunemaitre, 1987a). Therefore, a metabolic switch of skeleton synthetases on the plasma membrane could be an important step in ECM regulation. Actually, in animal cells. it has been suggested that a temporary depletion in chitin synthesis could be a necessary signal to promote a change in chitin micro-

330

fibril orientation (Neville, 1967). Such a metabolic switch may occur in plant cells, when changes in ECM structure are related to a change of glucan synthase activity on the plasma membrane. It is thought that, in peculiar conditions, cellulose synthetase is able to switch from cellulose production to callose production (Delmer, 1987; Seitz and Emmerling 1990). Such metabolic switches could explain some sudden ruptures/ alterations of helicoidal pattern, as a temporary skeleton fiber depletion will change the physical characteristics of the selfassembly medium (molecular interactions, viscosity of the medium, pressures), and then create a new starting point for another set of self-assembly processes. Considering now the embedding polymers, it has been mentioned previously that a slight change in the molecule conformation would give rise to different cholesteric properties and therefore would change the final helicoidal arrangement. Then, by altering the sorting and the processing of these structural polymers from Golgi apparatus to plasma membrane, the cell can control the final helicoidal pattern in ECM. Therefore the ECM regulation would result from a differential molecular packaging from the Golgi to the plasma membrane. Another way to regulate the polymer synthesis is to control the routing and the direction of the vesicles transporting skeleton synthetases and embedding polymers from the Golgi to the plasma membrane, and also to control the turnover of these synthetases at the cell surface. For example, increasing release of cytoplasmic vesicles into plant ECM induced experimentally has been reported to be concomitant with an increase of embedding polymer deposition (Okuda and Mizuta, 1987). Attempts to measure vesicular traffic and cell surface turn-over have already been performed (Steer, 1988 and references therein), and no doubt that regulation at this level will interfere with the final ECM composition, and consequently the final helicoidal arrangement. Cross-linkings. The presence of various cross-linkages in ECM is well documented especially in plants (Fry, 1986; MacCann et al., 1990; Satiat-Jeunemaitre et al., 1992). To study the possible cell control on helicoid formation by tne establishment of cross-links

SATIAT-JEUNEMAITRE

between ECM components, the first question is to ask if they are formed at the same time as the helicoid, or if they fixed instantaneously the unstable helicoidal arrangements made by the self-assembly processes. The strong analogy with liquid crystals supports the second hypothesis. Actually, such a phenomenon was suggested in animal cells: the intermutual orientation effect between chitin and protein molecules would take place before the cross-linking of the protein to the chitin and perhaps even that the two polymers have to be properly oriented by self-assembly to allow the cross linkages to be forged (Neville, 1967). In that case, cross links would be essentially the stabilisers of the helicoidal structure. Therefore, at first sight, their eventual role in the orientation of the skeleton fibers would be relevant from post-morphogenetic events, and should not be considered in a paper concerning the mechanisms involved in helicoid formation. However, one must keep in mind that the helicoid formation is conditioned by the preexisting ECM structures. Therefore, the degree of ‘fixation’ or the denaturation of the formed helicoids will affect the next assembly processes. For example, absence or breakages of cross linkings will probably increase the fluidity of these external ‘constraining layers’, meanwhile high levels of cross-linkage will increase the hardness of the structure. Consequently one could say that all the metabolic events which could act on such cross-linkings are a way to modulate the forming ECM. Conclusion Although similar helicoids occur in plants and animal ECMs, two distinct formation mechanisms have been generally proposed: (i) a microtubule-directed assembly for plant cells, (ii) self assembly for animal cells. The conjunction of the data in the two living kingdoms gives a better view of the whole phenomena, and allow mapping of the mechanisms possibly involved. Analogy with liquid crystals and in vitro reconstitution experiments clearly indicate that, in helicoidal ECM the question of macromolecular orientation may be solved at the physicochemical levels. However, because of the time component involved in helicoid formation, the variation occurring in helicoidal for-

REGULATIONS

IN HELICOIDAL

EXTRACELLULAR

mation and other characteristics of helicoids in ECM, a cellular control of the helicoid formation and stabilisation must be inferred. It appears clear that the cells have the necessary metabolic machinery to manipulate the self-assemblies of the biological polymers, and therefore act on the ECM morphogenesis. On the other hand, by producing hydrolytic enzymes or changing the physico-chemical environment of the extracellular compartment, cells are also able to change and even degrade the ECM. Such induced changes in the ECM can interact in return with the function of the cell. This dynamic reciprocity is an open field of investigation in which the plasma membrane is obviously a key structure in the mediation of the signals from one compartment to the other. However, this cell surface is continuously remodelled by the vesicular traffic from the Golgi apparatus to the cell surface

MATRICES

331

(exocytosis), and from the cell surface to the inside of the cell (endocytosis). In addition, the facts discussed in this paper show that ECM architecture actually begins its story in the secretory vesicles. Therefore, a better understanding of the ECM morphogenesis will arise from a better knowledge of the vesicular traffic inside the cell.

Acknowledgements I am grateful to Dr Neville, Prof. Smith and Prof. Roland for providing documents and for valuable discussion. I would like to thank the department of Biological and Molecular Sciences, Oxford Polytechnic, for giving me access to the facilities for the production of this manuscript. This work was partially supported by a travel grant ‘ALLIANCE’ from CNRS/British Council.

References Aheysekcra R. M. and Willison, J. H. M. 1990. Architecture of the fluid cellulosic arrays in the epidermis 01 the quince seed. Biol. Cell, 68, 251-257. Aldington, S.. MacDougall. G. J. and Fry. S. C. 1991. Structure-activity relationships of biologically active oligosaccharides. Plant. Cell and Environment, 14, 625-636. Bernfield. M. 1989. Extracellular matrix. Cur. Opin. in Cell Biol., 1, 953-955. Binnington. K. C. and Barrett, F. M. 1988. Ultrastructural localization of phenoloxidases in cuticle and haemopoiettc tissue of the bowfly Lucilia cuprina. Tissue Cell, 20, 405-419. Bolwell G. P. 1988. Synthesis of cell wall components: aspects of control. Phytochemistry, 27, 1235-1253. Bonfante-Fasolo, P.. Vian. B. and Testa. B. 1984. Wall texture in the spore of a vesicular-arbuscular mycorrhizal fungus. Protoplasma, 120, 51-60. Bouligand, Y 1965. Sur une architecture torsadee repandue dans de nombreuses cuticulcs d’arthropodes. C.K. Hrbd Seanc. Acad. Sci., Paris, 261, 3665-3668. Bouligand. Y. 1972. Twisted fibrous arrangements in biological materials and cholesteric mesophases. Trssue (LII. 4, 189-217. Bouligand. Y. 1976. Les analogues biologiques des cristaux liquides. La Recherche, 67, 474-476. Bouligand. Y. 1978. Cholesteric order in biopolymers. In Mesomorphic order in polymers and polymerization m lrqu~d crystalline media (ed. A. Blumstein), American Chemical Society, Washington. Bunning. C. E. 1973. The physiological clock, circadian rhythms and biological chronometry. The English Unnersny Press, Ltd. London. Delmer. D. P. 1987. Cellulose biosynthesis. Ann. Rev. Plant Physiol., 38, 259-290. Derksen. J.. Wilms, H. A. and Pierson, E. S. 1990. The plant cytoskeleton: its significance in plant development. ~crtr Bot. Neerl., 39, l-18. Emons. A. M. 1982. Microtubules do not control microfibril orientation m a hehcoidal cell wall. Protoplasmu. 113, 85-93, Favard, P. and Bouligand. Y. 1980. La phCnomenologie des autoassemblages biologiqucs. In La morphogemw. De la biologic aux mathLmatiques (ed. Y. Bouligand). pp 101-113. Maloine. Paris. Filshie. B. K. and Smith D. S. 1980. A proposed solution to a fine structural puzzle: the organization of gill cuticle in a crayfish (Panulirus). Tissue Cell, 12, 209-226. Fry. S. C. 1986. Cross-linking of matrix polymers in the growing cell walls of angiosperms. Ann. Reo. Plant phnwi., 37, 165-186. Giddings, T. H. and Staehelin, L. A. 1988. Spatial relationship between microtubules and plasma membrane rosettes during the deposition of primary wall microfibrils in Closterium sp. Plonru. 173, 22-30.

SATIAT-JEUNEMAITRE

332

Goudeau, M., Lachaise, F., Carpentier, G. and Goxe, B. 1990. High titers of ecdysteroids are associated with the secretory process of embryonic envelopes in the European lobster. Tissue Cell, 22, 269-281. Gray, G. W. 1962. Molecular structure and the properties of liquid crystals. Academic Press, London. Gunning, B. E. S. and Hardham, A. R. 1982. Microtubules. Ann. Reu. of Pht Physiol., 33, 651-698. Harche, M. 1986. Un type original d’architecture parietale: l’epiderme foliaire de I’Alfa (Stipa tenacissima). Cr. Acad. Sci., 303, 131-134. Harche, M., Chadli, R. and Catesson, A. M. 1990. Diversity of cellulose microfibtil arrangement in the cell walls of Lygeum spartum leaves. Annals of Botany, 65, 19-86. Heath, I. B. and Seagul, R. W. 1982. Oriented cellulose fibrils and the cytoskeleton: a critical comparison of models. In The cytoskeleton in plant growth and development (ed. C. C. Lloyd), pp 163-173. London, Academic Press. Herth, W. 1985. Plant cell wall formation. In Botanical microscopy (ed. A. W. Robards). Oxford University Press. Keil, T. A. and Steiner, C. 1990. Morphogenesis of the antenna of the male silkmoth, Antheraea polyphemus. I. The leaf-shaped antenna of the pupa from diapause to apolysis. Tissue Cell, 22, 319-336. Leforestier, A. and Livolant, F. 1991. Cholesteric liquid crystalline DNA - a comparative analysis of cryotixation methods. Biol. Cell, 71, 115-122 Leitch, A. R. 1989. Formation and ultrastructure of a complex multilayered wall around the oospore of chara and lamprothamnium (characeae). Br. Phycol. .I., 24, 229-236. Lemoine, A., Millot, C., Curie, G. and Delachambre, J. 1990. Spatial and temporal variations in cuticle proteins as revealed by monoclonal antibodies. Immunoblotting analysis and ultrastructural immunolocalization in a bettle, Tenebrio molitor. Tissue Cell, 22, 177-189.

Levy, S. 1987. A 3-D computer representation of helicoidal superstructures in biological materials, exemplified by the Nitella cell wall. Eur. J. Cell Biol., 44, 27-33. Livolant. F., Giraud, M. M. and Bouligand, Y. 1978. A goniometric effect observed in sections of twisted fibrous materials. Biol. Cell, 31, 159-168. LivoIant, F. 1991. Ordered phases of DNA in vivo and in vitro. Physica A., 170, 176, 117-137 Livolant, F. and Bouligand, Y. 1989. Freeze-fractures in cholesteric mesophases of polymers. Mol. Cryst. Liq. Cryst., 166,91-loo. Lloyd, C. W. 1987. The plant cytoskeleton: the impact of fluorescence microscopy. Ann. Reu. Plant Physiol., 38, 119 221. Lloyd, C. W. and Seagull, R. W. 1985. A new spring for plant cell biology microtubules as dynamic helices. TIBS, 10, 476-478. Lloyd, C. W. and Wells, B. 1985. Microtubules are at the tips of root hairs and form helical patterns corresponding to inner wall fibrils. J. CeN Sci., 75, 225-238. MacCann, M. C., Wells, B. and Roberts, K. 1990. Direct visualization of cross-links in the primary plant cell wall. J. Cell Sci., 96, 323-334. Maret, G., Milas, M. and Rinaudo, M. 1981. Cholesteric order in aqueous solutions of the polysaccharide xanthan. Polym. Bull., 4, 291-297.

Mazur, G. D., Regier, .I. C. and Kafatos, F. C. 1982. Order and defects in the silkmoth chorion, a biological analogue of a cholesteric liquid crystal. In Insect ultrastructure (eds. R. C. King and H. Akai), Vol. 1. Plenum. Millet, B. and Manachere. G. 1983. introduction b l’Ptude des rhytmes biologiques. Themes Vuibert, Paris. Mizuta, S. and Wada, S. 1982. Effects of light and inhibitors on polylamellation and shift of microtibril orientation in Boergesenia cell wall. PI. Cell Physiol., 23, 257-264. Mizuta, S., Kurogi. U., Okuda, K. and Brown, R. M. 1989. Microfibrillar structure, cortical microtubule arrangement and the effect of amiprophosmethyl on the microfibril orientation in the thallus cells of the filamentous green alga, Chamaedoris orientalis. Annals of Botany, 64, 383-394.

Mosiniak, M. and Roland, J. C. 1985. Variations spontanees du rhythme d’assemblage des parois cellulaires a texture helicoidale: l’exemple du papyrus. Ann. Sci. Nat. Bot., 7, 175-212.r Neville, A. C. 1965. Circadian organization of chitin in some insect skeletons. Quart. J. Misc. Sci., 106, 315-325. Neville, A. C. 1967. Chitin orientation in cuticle and its control. Ado. § Physiol., 4, 213-286. Neville, A. C. 1975. Biology of the arthropod cuticle. Springer-Verlag.,New York. Neville, A. C. 1985. Molecular and mechanical aspects of helicoid development in plant cell walls. Bioessays, 3, 4-9. Neville, A. C. 1986. The physics of helicoids. Phys. Bull., 37, 74-76. Neville, A. C. 1988a. A pipe-cleaner molecular model for morphogenesis of helicoidal plant cell walls based on hemicellulosic complexity. J. theor. Biol., 131, 243-254. Neville. A. C. 1988b. The helicoidal arrangement of microfibrils in some algal cell walls. In Progress in phycological research (eds. F. Round and Y. Chapman), 6, pp 1-21. Biopress Ltd. Neville, A. C. 1988~. The need for a constraining layer in the formation of monodomain helicoids in a wide range of biological structures. Tissue Cell, 20, 133-143. Neville, A. C. and Caveney, S. 1969. Scarabaeid beetle exocuticle as an optical analogue of cholesteric liquid crystals, Biological reuiews, 44, 531-562.

REGULATIONS

IN HELICOIDAL

EXTRACELLULAR

MATRICES

333

Neville. A. C. and Levy, S. 1984. Helicoidal orientation of cellulose microfibrils in Nitella opaca internode cells: ultrastructure and computed theoretical effects of strain reorientation during wall growth. Planta,162, 37&384. Neville, A. C. and Levy, S. 1985. The helicoidal concept in plant cell wall ultrastructure. In Biochemistry ofplnnr cell walls (eds. C. T. Brett and J. R. Hillman), pp. 99-124. Cambridge University Press. Neville, A. C. and Luke, B. M. 1969. A two system model for chitin-protein complexes in insect cuticle\. Tissue Cell. 1, 689-607. Neville. A. C. and Luke, B. M. 1971. A biological system producing a self-assembling cholesteric protein liquid crystal. J. Cell Sci., 8, 9>1G9. Okuda, K. and Mizuta, S. 1987. Accumulation of cytoplasmic vesicles in cell walls during cell modification in Chaetomorpha moniligera. Jap. J. Phycol., 35, 113-120. Pcng, H. B. and Jaffe, L. F. 1976. Cell wall formation in Pelvetia embryos. A freeze fracture study. Planta.133, 107116. Preston, R. D. 1988. Cellulose-microfibril-orienting mechanisms in plant cell walls. Planta, 174, 67-74. Preston, R. D. 1982. The case for multinet growth in growing walls of plant cells. Plantu. 159, 356363. Quader, H. 1986. Cellulose microfibril orientation in Oocystis solitaria: proof that microtubules control the alignment of the terminal complexes. J. Cell Sci., 83, 223-224. Quennedey. A. and Quennedey. B. 1990. Morphogenesis of the wing anlagen in the mealworm beetle Tenehrm molitor during the last larval instar. Tissue CeN, 22, 721-740. Rcls. D. 1978. Precisions cytochimiques sur I’assemblage in vitro des hemicelluloses de I’hypocotyle de soja (Phuwolus aureus Roxb.). Ann. Sci. Nat.. 19, 163-193. Reis. D. 1987. Cholesteric like pattern in plant cell walls: different expressions. MO/. Cryst. Liq. Cryst.. 153, 3.x53. Reis, D., Roland. J. C. and Vian, B. 1985. Morphogenesis of twisted cell wall: chronology following an osmotic shock. Protoplasma. 126, 3&46.

Roberts. I. N., Lloyd, C. W. and Roberts, K. 1985. Ethylene-induced microtuhule reorientations: mediation by hellcal arrays. Planta, 164, 439-447. Roberts, K. 1989. The plant excracellular matrix. Cur. Opin. in Ceil Biol., 1, 102@1027. Robinson. D. G. and Quader, H. 1982. The microtubule-microfibril syndrome. In The cytoskeleton in plant growth and deoelopment (ed. C. W. Lloyd). Academic Press. London, New York. Roland, J. C. 1981. Comparison of arced patterns in growing and non growing polylamellate cell walls of higher plants. In Cell walls’ 111, Proceedings of the second cell wall meeting held in Gottingen (eds. D. G. Robinson and H Quader). pp. 162-170. Roland. J. C. and Mosiniak. M. 1983. On the twisting pattern, texture and layering of the secondary cell walls of lime wood. Proposal of an unifying model. fAWA Bull.. 4, 15-26. Roland, J. C., Reis. D., Vian, B. and Roy, S. 1989. The helicoidal plant cell walls as a performing-based compo$itc. Rio/. Cell. 67, 209.220. Roland, J. C.. Reis, D., Vian, B. and Satiat-Jeunemaitre, B. 1983. Traduction du temps en espace dans les parols de cellules vegCtales. Ann. Sci. Nat., 5, 173-192. Roland, J. C., Reis, D., Vian, B., Satiat-Jeunemaitre, B. and Mosiniak. M. 1987. Morphogenesis of plant ccl1 walls at the supramolecular level: internal geometry and versatility of helicoidal expression. Protoplasma, 140, 75-91. Roland. J. C.. Vian, B. and Reis, D. 1975. Observations with cytochemistry and ultracryotomy on the fine structure of the expanding walls in actively elongating plant cells. J. Cell Ser.. 19, 239-259. Ryan. C. A. and Farmer. E. E. 1991. Oligosaccharide signals in plants: a current assessment. Ann Rev. Plum f’hwo/. Mol. Biol.. 42, 651-674. Ryscr. U., Meier. H. and Holloway, P. J. 1983. Identification and localization of suberin in the cell walls of green cotton Iibres (Gossypium hirsutum L var. Green Lint). Proroplasmu. 117, 196-205. Satiat-Jeunemaitre. B. 1984a. Croissance et texture pariCtale du colCoptile de mah g I’obscurite et en lumierc contmue. Physiol. Veg., 22, 745-753.

Satiat-Jeunemaitre, B. 1984b. Experimental modifications of the twisting and rhythmic pattern in the cell wall\ of maize coleoptile. Biol. Cell, 51, 373-380. Satiat-Jeunemaitre, B. 1986. Cell wall morphogenesis and structure in troplcal tension wood. IAWA Bull.. 7, l55S 164.

Sanat-Jeunemaitre, B. 1987a. Inhibition of the helicoidal assembly of the cellulose-hemicellulose complex by 2.6. Dichlorobenzonitrile (DCB). Biol Cell, 59, 89-96. Satiat-Jeunemaitre, B. 1987b. Spatio-temporal organization of the apoplast as a signal effector in cell growth polarity: experimental evidence. In The cell surface in signal transduction (eds. E. Wagner. H. Greppin and B. Millet) NATO asi series, Vol. H12, pp 129-138. Springer Verlag. Berlin, Heidelberg. Satlat-Jeunemaitre, B. 1989. Microtubules, microfibrilles pariCtales phogen&se veg&ale: cas des cellulcs en extension Bullerin de la SociM Botanique de France, 136, ActualitCs Botaniques, 2. 87-98. Satrat-Jeunemaitre. B. 1991. Microtubule and microfibril arrangements during cell wall expansion along the growth gradient of the mung bean hypocotyl. Ann. Sci. Nat.. (in press).

SATIAT-JEUNEMAITRE

334

Satiat-Jeunemaitre, B. and Mosiniak, M. 1992. Variations in morphogenetic rhythm of secondary cell wall deposition in prunus (sp) sclerocytes. Submitted for publication. Satiat-Jeunemaitre, B., Martin, B. and Hawes, C. R. 1992. Helicoidal plant cell walls revealed by rapid-freezing and deep-etching. Pro~oplasma (in press). Seitz, H. U. and Emmerling, M. 1990. The cellulose synthase problem. Bat. Acta, 103, 7-8. Smith, D. S., Telfer, W. H. and Neville, A. C. 1971. Fine structure of the chorion of a moth, Hyalophora cecropia. Tissue Cell, 3, 411-498.

Smith, D. S., Jarlfors, U. and Russell, F. E. 1969. The fine structure of muscle attachments in a spider (Lafrodecrus mactans, Fabr.). Tissue Cell, 1, 67%687. Staehelin, L. A. and Giddings, T. H. 1982. Membrane mediated control of cell wall microfibrillar order. In Deuelopmental order: ifs origin and regulation. (ed. Alan R. Liss), pp. 133-147. Inc. New York. Steer M. 1988. Plasma membrane turnover in plant cells. J. Exp. Rot., 39, 987-996. Sweeney, B. M. 1987. Rhythmic phenomena in plan& (eds. Harcourt Brace and Jovanovitch). Acad. Press, Inc. Traas, J. A., Braat, P., Emons, A. M., Meekes, H. and Derksen, J. W. 1985. Microtubules in root hairs. J. Cell Sci., 76, 303-320. Velasco, A. and Hidalgo, J. 1988. Ultrastructural demonstration of proteoglycans in adult rat cornea. Tissue Cell, 20, 561-575. Vian, B. and Roland, J. C. 1987. The helicoidal cell wall as a time register. New Phyfol., 105, 345-357. Vian, B., Mosiniak, M., Reis, D. and Roland, J. C. 1982. Dissipative process and experimental retardation of the twisting in the growing plant cell walls. Effect of ethylene-generating agent and colchicine: a morphogenetic revaluation. Biol. Ceil, 46, 301-310. Vian, B., Reis, D., Mosiniak, M. and Roland, J. C. 1986. The glucuronoxylans and the helicoidal shift in cellulose microfibrils in linden wood: cytochemistry in muro and on isolated molecules. Protoplasma, 131, 185-199. Wardrop, A. B., Welter-arts, M. and Sassen, M. M. A. 1979. Changes in microtibril orientation in the walls of elongating plant cells. Acta Bat. Neerl., 28, 31%333. Weis-Fogh, T. 1970. Structure and formation of insect cuticle. In Insect ultrastructure (ed. A. C. Neville). Roy. Ent. Sot. Sympos., 5, 165-184. Werbowyj, R. and Gray, D. Cl. 1984. Optical properties of hydroxypropylcellulose liquid crystals. Cholesteric pitch and polymer concentration. Macromolecules, 17, 1512-1520. Wright, J. C. and Luke, B. M. 1989. Ultrastructural and histochemical investigations of peripatus integument. T&we Cell, 21, 605-625. Zelazny, B. and Neville, A. C. 1972a. Endocuticle layer formation controlled by non-circadian clocks in beetles. J. Insect Physiol., 18, 1967-1979. Zelazny, B. and Neville, A. C. 1972b. Quantitative studies on libril orientation in beetle endocuticle. J. Insect Physiol., 18,2095-2121.