pmc.ncbi.nlm.nih.gov

Cellular Differentiation in Moss Protonemata: A Morphological and Experimental Study

Abstract

Background and Aims

Previous studies of protonemal morphogenesis in mosses have focused on the cytoskeletal basis of tip growth and the production of asexual propagules. This study provides the first comprehensive description of the differentiation of caulonemata and rhizoids, which share the same cytology, and the roles of the cytoskeleton in organelle shaping and spatial arrangement.

Methods

Light and electron microscope observations were carried out on in vitro cultured and wild protonemata from over 200 moss species. Oryzalin and cytochalasin D were used to investigate the role of the cytoskeleton in the cytological organization of fully differentiated protonemal cells; time-lapse photography was employed to monitor organelle positions.

Key Results

The onset of differentiation in initially highly vacuolate subapical cells is marked by the appearance of tubular endoplasmic reticulum (ER) profiles with crystalline inclusions, closely followed by an increase in rough endoplasmic reticulum (RER). The tonoplast disintegrates and the original vacuole is replaced by a population of vesicles and small vacuoles originating de novo from RER. The cytoplasm then becomes distributed throughout the cell lumen, an event closely followed by the appearance of endoplasmic microtubules (MTs) in association with sheets of ER, stacks of vesicles that subsequently disperse, elongate mitochondria and chloroplasts and long tubular extensions at both poles of the nucleus. The production of large vesicles by previously inactive dictysomes coincides with the deposition of additional cell wall layers. At maturity, the numbers of endoplasmic microtubules decline, dictyosomes become inactive and the ER is predominantly smooth. Fully developed cells remain largely unaffected by cytochalasin; oryzalin elicits profound cytological changes. Both inhibitors elicit the formation of giant plastids. The plastids and other organelles in fully developed cells are largely stationary.

Conclusions

Differentiation of caulonemata and rhizoids involves a remarkable series of cytological changes, some of which closely recall major events in sieve element ontogeny in tracheophytes. The cytology of fully differentiated cells is remarkably similar to that of moss food-conducting cells and, in both, is dependent on an intact microtubule cytoskeleton. The disappearance of the major vacuolar apparatus is probably related to the function of caulonema and rhizoids in solute transport. Failure of fully differentiated caulonema and rhizoid cells to regenerate is attributed to a combination of endo-reduplication and irreversible tonoplast fragmentation. The formation of giant plastids, most likely by fusion, following both oryzalin and cytochalasin treatments, suggests key roles for both microtubules and microfilaments in the spatial arrangement and replication of plastids.

Key words: Bryophyta, caulonemata, cytoskeletal inhibitors, food-conducting cells, protonemal differentiation, rhizoids, sieve elements

INTRODUCTION

In the last three decades bryophytes (liverworts, mosses and hornworts), especially mosses, have increasingly been used as model systems for investigating the cellular and molecular basis of plant development and metabolism (see Cove et al., 1997, 2006, for reviews). The haploid gametophyte stage of their life cycle is particularly amenable to genetic manipulation and numerous mutants have been isolated, especially in Physcomitrella patens (Knight et al., 1988). Physcomitrella patens is proving an excellent system for dissecting the biological function of genes involved in stress responses (Reski, 1998; Schaefer, 2001; Frank et al., 2005), development and metabolism (see Cove et al., 2006, for a review) due, in no small part, to the development of homologous recombination techniques in this moss (Schaefer and Zrÿd, 1997; Reski, 1998, 1999; Hofmann et al., 1999) and the complete sequencing of its genome (Rensing et al., 2008). The majority of recent phylogenetic investigations unambiguously identify bryophytes as the present-living descendants of the earliest plant lineages that radiated successfully into terrestrial niches (Shaw and Renzaglia, 2004; Renzaglia et al., 2007). The study of bryophytes, therefore, may help us to unravel the evolutionary history of plant adaptation to the constraints of the terrestrial environment (Cuming et al., 2007).

In contrast to tip-growing filaments in all other land plants (including liverworts and hornworts), which are always unicellular, tip-growing filaments in mosses are multicellular and highly branched structures. These comprise three types of filaments: the chloronemata and caulonemata that form the juvenile (protonemal) phase of the gametophyte generation, and the rhizoids. Though often described as filaments produced only by the mature gametophores or leafy shoots of mosses (Crundwell, 1979), rhizoids intergrade in all respects with caulonemata (Goode et al., 1992; Duckett et al., 1998, 2004). Thus, both are treated together in the present generic account.

These three kinds of filament all elongate by tip growth (Menand et al., 2007a); however, the cytology, differentiation and functions of chloronemata are very different from those of caulonemata and rhizoids. In the former, all the cells, including the tip cells, exhibit essentially the same cytology, with organelles arranged peripherally around a large central vacuole. Their microtubule system comprises cortical helices similar to those in parenchyma cells. Nuclei in the chloronemata remain haploid (Kingham et al., 1995, 1997) and the individual cells are capable either of regenerating a new protonema or de-differentiating into spherical brood cells that in turn can produce a new protonema (Goode et al., 1993a). With a cytology closely resembling that of photosynthetic cells in the leafy shoots, the principal functions of chloronemata are assimilation (Tewinkel and Volkmann, 1987; Duckett et al., 1998) and asexual reproduction via brood cell or gemma formation (Sawidis et al., 1991; Goode et al., 1993b).

Caulonemata and rhizoids are very different to chloronemata (Schnepf, 1982; Jensen and Jensen, 1984; Doonan, 1991; Quader and Schnepf, 1989). Their apical cells are cytologically more similar to typical tip-growing cells such as pollen tubes (Steer and Steer, 1989), root hairs (Traas et al., 1985) and fungal hyphae (Tanaka, 1985). Endoplasmic and cortical microtubule and actin arrays extend from a centrally placed nucleus into the apical dome through a cytoplasmic cap containing abundant mitochondria, endoplasmic reticulum, dictyosomes and numerous gravity-sensing amyloplasts (Walker and Sack, 1990). A single large vacuole lies behind the nucleus and, following mitosis, the formation of an oblique cross-wall produces a subapical cell that is highly vacuolate. Significantly faster growth rates of rhizoids and caulonemata than chloronemata are most likely related to their more highly specialized tip-growing apparatus (Menand et al., 2007a).

All three kinds of filament produce side branches, usually immediately behind the cross-walls between two and five cells behind the apex (Schmiedel and Schnepf, 1979a, b, 1980; Doonan et al., 1986, 1988). This process involves microtubule-mediated migration of the nucleus from a central location in the peripheral cytoplasm to a bulge in the lateral wall that marks the position of the nascent bud. This is followed by an asymmetrical division that separates a branch initial with dense cytoplasm from the original filament cell. The latter remains highly vacuolate with widely scattered plastids, and its nucleus returns to a central position in the peripheral cytoplasm.

The roles of the cytoskeleton in growing protonemal cells and side-branch development have been elegantly demonstrated via the use of drugs in conjunction with immunocytochemistry (Doonan, 1991; Doonan and Duckett, 1988; Wacker et al., 1988), and most recently using the green-fluorescent protein talin (Finka et al., 2007). Caulonemal tip growth and side-branch formation have also been documented in considerable detail ultrastructurally (Jensen, 1981; Schnepf, 1982; Conrad et al., 1986). In contrast, apart from the studies by Kingham et al. (1995, 1997), which demonstrated genome endoreduplication up to 8C, little is known about cytological changes associated with maturation in caulonemal and rhizoidal cells. Even for fully differentiated cells, ultrastructural details are limited to only two species, Funaria and Bryum, and document a cytoplasmic organization similar to that described for food-conducting cells (FCCs) in the leafy shoot (Duckett et al., 1998). An account of the differentiation of protonemal cells from highly vacuolate to vacuole-less specialized cells with ‘food-conducting cytology’, together with experiments on the roles of the cytoskeleton is long overdue. Such information also becomes increasingly important for interpreting the molecular basis of moss development (Floyd and Bowman, 2007; Gremillon et al., 2007; Menand et al., 2007a).

Thus the aims of the present study were: (1) to provide the first comprehensive account, based on observations of both wild-collected and cultured specimens of over 200 taxa, of the ultrastructural changes that accompany the differentiation of moss caulonemal and rhizoid cells; (2) to compare critically the cytology of mature caulonemal and rhizoid cells with that of FCCs in moss gametophores and sporophytes and of sieve elements in vascular plants; and (3) to investigate the effects of disruption of the microtubule and microfilament cytoskeletons on mature caulonemal and rhizoid cell cytology.

MATERIALS AND METHODS

Plant material

The key features of protonemal differentiation described here are based on observations on over 200 moss taxa (Pressel, 2007), both in axenic cultures on Parker nutrient medium (Klekowski, 1969) solidified with 1 % on Phytogel (see Duckett and Ligrone, 1992 for full details) and in field-collected samples. Representative light-microscope images were selected from the following: Campylopus introflexus (Hedw.) Brid., Funaria hygrometrica Hedw., Physcomitrella patens (Hedw.) Bruch, Schimp. & Gumbel, Bryum tenuisetum Limpr., Rhizomnium punctatum (Hedw.) T. J. Kop., Bartramia pomiformis Hedw., Dicranum scoparium Hedw., D. flagellare Hedw., Tortula muralis Hedw., Eustichia longirostris Brid. and Encalypta streptocarpa Hedw. The first four taxa were selected for transmission electron microscopy; the inhibitor experiments were carried out, with identical results, on Funaria and Campylopus. The scanning electron micrographs of Bryum argenteum Hedw. and Plagiomnium cuspidatum (Hedw.) T. J. Kop. were selected as typical for bryoid mosses.

Light and electron microscopy

Living specimens were mounted in water and photographed with a digital camera under a Leica DM RXAZ microscope using differential interference contrast optics.

For transmission electron microscopy (TEM) protonemata and rhizoids were processed following the protocol of Ligrone and Duckett (1994); they were fixed in 3 % (v/v) glutaraldehyde, 1 % (v/v) formaldehyde (freshly prepared from paraformaldehyde) and 0·5 % (w/v) tannic acid in 0·05 m Na-phosphate buffer, pH 7·0 for 2 h at room temperature. After rinsing in 0·1 m buffer, the material was post-fixed with 1 % (w/v) osmium tetroxide in 0·1 m Na-phosphate buffer, pH 6·8, overnight at 4 °C, dehydrated through an ethanol series and embedded in Spurr's resin via propylene oxide. Thin sections were sequentially stained with 5 % (v/v) methanolic uranyl acetate for 15 min and lead citrate for 10 min and examined with a Jeol 1200 EX2 electron microscope. Critical-point drying for scanning electron microscopy (SEM) followed the protocol of Duckett and Ligrone (1995). Briefly, protonemata and rhizoids were dehydrated over 24 h in a graded ethanol : acetone series, critical-point dried, coated with a layer of gold about 20 nm thick, and observed with a Hitachi S570 scanning electron microscope operating at 20 kV.

Drug treatments

To investigate the roles of the cytoskeleton in determining protonemal cytology we used 10 µm oryzalin (a microtubule inhibitor) and 2 µm cytochalasin D (a microfilament inhibitor). The concentrations were based on previous studies (Schmiedel and Schnepf, 1979a, b, 1980; Wacker et al., 1988; Ligrone and Duckett, 1996) as well as on trials with different exposure times. Twelve hours was finally selected to ensure drug penetration in mature cells. The drugs were dissolved in dimethylsulphoxide (DMSO) to a concentration of 10 mm and, after dilution in distilled water to the desired final concentration, were applied as liquids. Controls were performed by incubating the samples in diluted DMSO only. This compound has no effect on protonemal morphogenesis up to a concentration of 1 % (Doonan et al., 1988).

RESULTS

Light microscopy

Figure 1 illustrates the typical appearance of chloronemal, caulonemal and rhizoid cells at different stages of development at the light-microscope level. Figure 2 highlights the degree of variation in mature caulonemal and rhizoid cells that one might expect to encounter in different taxa. Some of these data have been published before (e.g. Doonan, 1991; Jensen, 1981; Duckett et al., 1998) but are included in this account to serve as ‘controls’ and to aid interpretation of the ultrastructural changes that accompany differentiation in caulonemal and rhizoid cells, reported here for the first time.

Fig. 1.

Fig. 1.

Light micrographs of the key events in protonemal differentiation in mosses. All Funaria hygrometrica except (A, B) Tortula muralis, (D) Bartramia pomiformis, (F) Physcomitrella patens, (K) Encalypta streptocarpa, (L, M) Dicranum scoparium. (A) Chloronemal apical cell with peripheral chloroplasts and a large central vacuole. (B) Early onset of differentiation in cell 3 immediately following side-branch formation. (C) Caulonemal apical cell showing a basal vacuole, central nucleus and a plastid-free apical dome. (D) Phragmoplast in an apical cell. (E) Highly vacuolate subapical caulonemal cells before side-branch formation. (F) Cell 4 after side-branch formation remains highly vacuolate. (G) Onset of differentiation; increase in the cytoplasm around the nucleus and fine strands of cytoplasm extending through the vacuole. Note the plastid-free apical dome in the apical cell above. (H) Slighly later stage with more cytoplasmic strands traversing the vacuole. (I, J) Onset of longitudinal plastid alignment and loss of the vacuole. (K) Wild rhizoid showing cytoplasmic strands containing minute elongate plastids extending from the nucleus. (L) Wild rhizoid showing minute plastids, a spindle-shaped nucleus and a mucilage sheath outside the cell wall (arrowed). (M) Cultured caulonemal filament with cells packed with elongate chloroplasts. (N) Wild rhizoid showing an elongate nucleus and cytoplasmic strands containing minute plastids. The mucilage sheath is arrowed. (O) Fully differentiated caulonema. Note the pigmented walls and the polarized cytoplasm with the majority of the plastids at the distal end of the cell adjacent to the side branch. n, Nucleus. Scale bars: (B) = 100 µm; all others = 20 µm.

Fig. 2.

Fig. 2.

Light micrographs illustrating the range of cytologies in fully differentiated caulonemata and rhizoids. (A) Rhizomnium undulatum wild rhizoid with endoplasmic strands and scattered spindle-shaped plastids. (B, C) Eustichia longirostris, cultured rhizoids with swollen (B) and elongate (C) plastids packed with starch grains. (D) Funaria hygrometrica, wild rhizoid with highly elongate plastids. (E) Encalypta streptocarpa, wild caulonema with starch-filled plastids. (F) Dicranum flagellare, cultured rhizoid with minute plastids scattered along endoplasmic strands. (G, H) Dicranum scoparium, wild rhizoids with ovoid plastids scattered along cytoplasmic strands. (I) Encalypta streptocarpa, wild rhizoid containing over 100 cytoplasmic strands. (J) Funaria hygrometrica old caulonema packed with lipid droplets between elongate plastids. (K) Bryum tenuisetum, old rhizoid with lipid droplets aggregated at the apical end of the cell. l, Lipid droplets. Scale bars = 20 µm.

In chloronemata the apical cell and its derivatives are highly vacuolate with peripheral chloroplasts and a centrally located nucleus with a conspicuous nucleolus (Fig. 1A). Apical cells of caulonemata (Fig. 1B, C) have a plastid-free apical dome, a central spherical nucleus and a basal vacuole. The nucleus maintains a constant distance from the tip and the vacuole increases in size as the tip cell grows. Oblique divisions of the apical cell (Fig. 1D) produce highly vacuolate subapical cells (Fig. 1B, E) that contrast with the adjacent just-formed apical cells filled with cytoplasm. The subapical cells have small, usually rounded, plastids scattered in the peripheral cytoplasm and the nucleus is also confined to the periphery (Fig. 1E). The cells remain in this condition until after side-branch formation (Fig. 1B, F), which occurs as early as cells 2–3 in a few species (e.g. Tortula muralis) but more commonly in cell 4 or more from the apex.

The first sign of differentiation in caulonemal and rhizoid cells behind the site of side-branch formation is an increase in the relative volume of the peripheral cytoplasm and the formation of plastid-containing cytoplasmic strands extending through the central vacuole (Fig. 1G). These strands increase in frequency (Fig. 1H), thus dividing the central vacuole into several small vacuoles until suddenly (e.g. between cells 7 and 9 in Funaria) distinct boundaries between cytoplasm and vacuoles are lost (Fig 1I). Concomitantly the nucleus moves to a central location (Fig. 1I) and a longitudinal alignment of the plastids is increasingly apparent (Fig. 1J). Ultimately longitudinal strands of organelles become a ubiquitous feature of mature caulonemal and rhizoid cells (Fig. 1K–N; Fig. 2). The degree of brown pigmentation of cell walls varies considerably between taxa; within the same taxon the rhizoids tend to be more highly pigmented (Fig. 1K, N, O) than caulonemata, which instead usually contain more prominent plastids (Fig. 1M). A mucilage sheath outside the cell wall (Fig. 1L, N) is a feature found in all taxa.

In most cases an ovoid nucleus occupies a central location with endoplasmic strands radiating from it (Fig. 1K) but sometimes the nucleus becomes displaced to lie nearer the apical cross-wall (Fig. 1L, N). Other strands, completely separate from the nucleus, extend to the extremities of the cells. Some cells display distinct polarity with most of the plastids (Fig. 1O) and, in older filaments, also lipid droplets (Fig. 2J) lying towards the apical end of the cell.

The greatest variations between mature cells are in the number of endoplasmic longitudinal strands, which appear to depend on the diameter of the filaments in question. The main caulonemal and rhizoidal axes in Funaria, Physcomitrella, Bryum, Dicranum and Campylopus are examples of the most typical situation; these are approximately 20 µm in diameter and contain 20–30 strands per cell. The wider rhizoids of Rhizomnium (25 µm) contain between 40 and 60 strands (Fig. 2A, D–H) whilst the highest number we recorded was 80–100 in the 30-μm-wide rhizoids of Encalypta streptocarpa (Fig. 2I). At the other extreme are the caulonemata and rhizoids in Eustichia (Fig. 2B, C) and the Grimmiales (Pressel, 2007) that are about 15 µm in diameter and contain only 5–10 strands. In narrow rhizoidal side-branches in all taxa the number of endoplasmic strands is similarly reduced (not illustrated).

Filaments contain various plastid numbers and shapes. At one extreme are cells filled with elongate spindle-shaped plastids up to 20 µm in length, e.g. caulonemal cells of Bryum tenuisetum and Funaria (Figs 1M and 2J), at the other are widely scattered minute plastids less than 1 µm in diameter (Figs 1O and 2F, I). Other cells may have much larger, elongate plastids with (Fig. 2C) or without (Fig. 2A, D) starch grains. Alternatively in some taxa the plastids are ovoid or rounded, whether they be starch-filled (Fig. 2E) or starch-free (Fig. 2G, H). There is considerable variation in plastid shape, size and numbers even along the same filament and between filaments within a single protonemal colony or growing from the base of a single gametophore. Numerous spindle-shaped plastids tend to be most frequent in caulonemata growing on nutrient-rich media whereas minute plastids are more characteristic of caulonemata growing on nutrient-poor media and wild subterranean rhizoids.

Caulonemata and rhizoids have lipid droplets, often aligned along the endoplasmic strands (Fig. 2H), the accumulation of which is particularly prominent in ageing cultures (Fig. 2J, K).

Scanning electron microscopy

A mucilage sheath is readily visible in Fig. 3A and its extremely delicate nature is revealed by images showing its more-or-less complete disruption (Fig. 3B, C), as often happens during preparation of specimens for TEM. Even more striking is the external wall layer exposed beneath the mucilage layer. Whereas this has little or no obvious substructure under TEM, under SEM it is highly ornamented in mature cells (Fig. 3C).

Fig. 3.

Fig. 3.

(A–C) Scanning electron micrographs of rhizoids showing the mucilage sheath and the highly ornamented outer wall layer beneath. (A) Bryum donianum, cultured rhizoid with the mucilage sheath intact apart from small splits. (B, C) Plagiomnium cuspidatum, wild rhizoids with the mucilage sheath extensively ruptured (B) and absent (C), exposing the highly ornamented outer wall layer. (D–F) Time-lapse images of a wild rhizoid of Encalypta streptocarpa. Stationary lipid droplets (arrowheads) act as reference points for the moving plastids (small arrows). After 15 min (E) plastid ‘a’ has moved towards plastid ‘b’. After 30 min (F) both ‘a’ and ‘b’ have moved in the opposite direction and the cluster of three plastids ‘c’ has separated. Scale bars: (A–C) = 10 µm; (D–F) = 20 µm.

Time-lapse photography

Results for mature caulonemal and rhizoid cells were very different from those obtained previously on tip cells (Walker and Sack, 1990, 1995) since movements in the former are clearly not associated with cell expansion or graviperception. Cytoplasmic streaming was never observed. A constant feature in all four species analysed was that the lipid bodies remained absolutely stationary thoughout the 6-h periods of observation (Fig. 3D–F). They thus formed useful reference points for the detection of plastid movement along the endoplasmic strands. No movements were detected for any of the plastids in cells packed with these (e.g. Figs 1M and 2E) and similarly the vast majority of more widely spaced elongate plastids remained stationary (e.g. Fig. 2A, D). The majority of the ovoid and spherical plastids in the rhizoids of Dicranum (Fig. 2G, H) and Encalypta (Fig. 3E, F) exhibited short saltatory movements of up to 2–3 µm along individual endoplasmic strands. Only in a few cases (Fig. 1L–N) were plastids seen to move uni- or bidirectionally for distances of up to 20 µm over a period of 1–6 h.

Transmission electron microscopy

Chloronema

In typical mature chloronemal cells the centrally placed nucleus is rounded, with a conspicuous fibrillar-granular nucleolus (Fig. 4A). Most of the cell lumen away from the nucleus is occupied by a large vacuole with scattered electron-opaque deposits. The cytoplasm around the nucleus and lining the cell walls contains swollen and ovoid (approx. 5 µm in length) chloroplasts with numerous starch grains and rounded-to-slightly-elongate mitochondria with saccate cristae and an electron-opaque matrix (not illustrated). The cell wall is unistratose, homogeneous and approx. 500 nm thick.

Fig. 4.

Fig. 4.

Transmission electron micrographs. All Funaria hygrometrica, except (D) Bryum tenuisetum and (G) Campylopus introflexus. (A) Chloronema cell with a large, central vacuole, swollen and ovoid, starch-containing plastids and a spherical, centrally located nucleus (cf. light micrograph Fig. 1A). (B, C) Undifferentiated caulonemal cells (cf. light micrograph Fig. 1 F). (D–G) The onset of differentiation. (D) Proliferation of sheets of cortical RER. (E–G) Membrane-bound crystalline inclusions (arrowed). d, Dictyosome; m, ovoid and spherical mitochondria; mi, microbodies; n, nucleus; p, plastids; v, vacuoles. Scale bars: (A–C) = 5 µm; (D, E) = 1 µm; (F, G) = 0·5 µm.

Caulonema and rhizoids

Subapical caulonemal and rhizoid cells, both before and immediately following side-branch formation, are highly vacuolate with all the organelles confined to a thin layer of peripheral cytoplasm along both the end (Fig. 4B) and side walls (Fig. 4C). Chloroplasts are discoidal (approx. 6 µm in length) and contain little or no starch. Mitochondria with saccate cristae and an electron-opaque matrix are rounded-to-slightly-elongate and the nuclei spherical-to-ovoid (Fig. 4C). The cytoplasm also contains scattered dictyosomes with small vesicles and short profiles of rough ER. The longitudinal cell wall, approx. 400 nm in thickness, is unistratose with a finely fibrillar texture (Fig. 4B, C).

Following side-branch initiation, an increase in the amount of cytoplasm, particularly at the apical ends of the cells, signals the onset of differentiation (Fig. 4D–G), although the cells remain highly vacuolate and the cytoplasm mostly peripheral. There are no changes in the plastids and mitochondria, the former containing little or no starch, and the dictyosomes remain inactive (Fig. 4F). Microbodies are often associated with both plastids and mitochondria (Fig. 4F). Sheets of rough ER are prominent along the walls (Fig. 4D) where occasional cortical microtubules are also visible, as noted previously by McCauley and Hepler (1992; not illustrated). A highly distinctive feature of these cells is the presence of crystalline inclusions in the lumina of tubular rough ER (Fig. 4F, G).

Slightly older cells are very different (Fig. 5). The plastids remain peripheral but the cytoplasm is much less electron-opaque and contains numerous cup-shaped single profiles of fenestrated smooth ER (Fig. 5A) associated with numerous vesicles 0·3–0·4 µm in diameter, often lying close together in pairs or triplets (Fig. 5B). The tonoplast then breaks down (Fig. 5C) and the bulk of the lumina of the cells becomes filled with a finely granular, electron-transparent material, mostly devoid of organelles, which instead remain located peripherally. Away from the cup-shaped ER profiles, the peripheral cytoplasm contains fenestrated sheets of ER, cortical microtubules and numerous dictyosomes with large peripheral vesicles, suggesting intense Golgi activity (Fig. 5D, E). In the cytoplasmic area immediately underneath are visible scattered endoplasmic microtubules with no preferred orientation (Fig. 5F). The cell wall retains a homogeneous structure (Fig. 5E) but has increased in thickness to 0·8–1 µm. In addition, at this time of intense dictyosome activity the nucleus becomes elongate, centrally positioned, and begins to develop tubular extensions from each end (Fig. 5G).

Fig. 5.

Fig. 5.

Transmission electron micrographs, early stages in differentiation. (A–C) Physcomitrella patens, (D–F) Funaria hygrometrica, (G) Campylopus introflexus, (H) Bryum tenuisetum. (A) Cup-shaped profiles of smooth ER adjacent to the nucleus (cf. light micrographs Fig. 1I, J). (B) Enlargement of (A) showing numerous groups of vesicles associated with the ER profiles. Note the intact tonoplast. (C) Slightly later stage with no signs of a tonoplast. (D) Dictyosomes with numerous large vesicles and short ER profiles. (E) Cortical microtubules (arrowed) and peripheral ER at the onset of wall thickening. (F) Randomly arranged microtubules (arrowed) associated with ER, Golgi bodies and a mitochondrion. (G) Tubular (arrowed) extension from a nucleus. (H) Microtubules associated with an elongate mitochondrion. d, Dictyosome; n, nucleus; m, mitochondria; p, plastids; s, cup-shaped profiles of smooth ER; t, tonoplast; v, vacuoles. Scale bars: (A, C) = 5 µm; (B) = 2 µm; (D–H) = 1 µm.

Further differentiation sees the establishment of longitudinal arrays of endoplasmic microtubules. Initially these are associated with elongate mitochondria (Fig. 5H) and sheets of rough ER (Fig. 6A), sometimes paralleling the membrane-bound fibrillar bundles (Fig. 6B) seen at the onset of differentiation (cf. Fig. 4E–G). Increasingly, however, they lie alongside fenestrated sheets of longitudinally orientated smooth ER (Fig. 6C) associated with a variety of vesicles, all with electron-transparent contents (Fig. 6C–G). Prominent initially but rarely present in fully differentiated cells are spherical stacks, approx. 0·4–0·5 µm in diameter, of up to 20 tightly packed, flattened vesicles (Fig. 5E). The numerous vesicles observed in close proximity to these stacks are characteristically lunate and 0·6–0·8 µm in diameter (Fig. 6D), while the remainder, mostly aligned along microtubules, are generally spherical and over 1 µm in diameter (Fig. 6F, G).

Fig. 6.

Fig. 6.

Transmission electron micrographs, caulonemal differentiation continued. All Funaria hygrometrica except (H) Bryum tenuisetum. (A, B) Longitudinally aligned microtubules associated with fenestrated sheets of rough ER. In (B) note the crystalline inclusion (arrowed). (C, D) Fenestrated sheets of smooth ER associated with microtubules. In their vicinity are numerous vesicles, some with lunate profiles (arrowed in D) and stacks of vesicles. (E) Detail of a stack of vesicles and smooth ER. (F) Row of vesicles aligned along endoplasmic microtubules (arrowed). (G) Transverse section showing vesicles and their associated microtubules (arrowed). (H, I) Longitudinal profiles through end walls. No polarity is evident in (H) (cf. light micrograph Fig. 1 N) but in (I) the cytoplasm is much denser at the apical end of the cell (cf. light micrograph Fig. 1O). Note the abundant plasmodesmata in the cross-wall in (I). (J) Elongate nucleus, plastids with numerous starch grains and vesicles scattered through the cytoplasm. Scale bars: (C, H–J) = 5 µm; (A, B, D, F) = 1 µm; (E, G) = 0·5 µm.

Mature cells have longitudinally aligned elongate plastids, up to 20 µm in length and starch-free in Bryum tenuisetum (Fig. 6H), both ovoid and elongate and packed with starch grains in Funaria (Fig. 6J), and abundant mitochondria up to 3–5 µm in length (Fig. 6H). Cytoplasmic polarity, when present, is now clearly evident with denser cytoplasm towards the apical ends of the cells (Fig. 6I). The nuclei are highly elongate and contain a fragmented nucleolus (Fig. 6J), as described by Kingham et al. (1995). Numerous microtubules line the nuclear envelope and extend along its polar tubular extensions, the latter now sometimes over 20 µm in length and reaching the ends of the cells (Fig. 7A–C). Numerous vesicles are scattered throughout the electron-transparent cytoplasm and, as in younger cells, are frequently aligned along endoplasmic microtubules (not shown). Longitudinally aligned microtubules and lipid droplets are also frequent in the cortical cytoplasm (Fig. 7D). In the internal cytoplasm microtubules are associated with single fenestrated profiles of mainly smooth ER (Fig. 6E). The internal cytoplasm also contains scattered polysomes (Fig. 7E) but most of the ribosomes are aggregated around the plastids (Fig. 6G). Microbodies (Fig. 7G) remain the same as in undifferentiated cells (cf. Fig. 4E) as also do the crystalline aggregations within tubular rough ER (Fig. 7H). Dictyosomes with small vesicles (Fig. 7F), often associated with partially coated reticulum, are a further constant feature scattered throughout the cytoplasm of mature cells. Although longitudinally aligned microtubules can still be seen adjacent to the plastids and mitochondria (Fig. 8A), their numbers are reduced to usually only one per mitochondrion compared with five or more earlier (Fig. 5H), and several scattered microtubules per plastid (Fig. 8A) compared with several groups of six or more earlier (not illustrated).

Fig. 7.

Fig. 7.

Transmission electron micrographs, fully differentiated caulonemal cells. All Funaria hygrometrica. (A) Longitudial profile of a tubular extension from the nuclear envelope (arrowed). (B) Transverse section of a vesicle and the tubular extension of the nuclear envelope (arrowed) with associated microtubules. (C) Tubular extension from the nuclear envelope (arrowed) near a cross-wall and over 30 µm from a centrally located nucleus. (D) Grazing section showing cortical microtubules and lipid droplets. (E) Endoplasmic microtubules associated with smooth ER in electron transparent cytoplasm containing scattered polysomes. (F) Inactive dictyosomes and partially coated reticulum. (G) A plastid, an elongate mitochondrion and microbodies surrounded by free ribosomes. Note the irregular, electron-lucent invagination of the cell wall (arrowed). (H) Rough ER with crystalline inclusions. mi, microbodies; p, plastid. Scale bars: (A, C, D, G) = 2 µm; (E, F) = 1 µm; (B, H) = 0·5 µm.

Fig. 8.

Fig. 8.

Fully differentiated caulonemal cells. (A–D) Funaria hygrometrica. (A) Transverse section showing microtubules (arrowed) associated with the envelopes of a plastid and mitochondrion. (B) Highly convoluted plasmalemma–wall interface associated with ER and cortical microtubules. (C) Section showing distict layering of the cell wall and a highly irrregular innermost stratum (arrowed). (D) Longitudial section showing highly elongate starch-filled plastids in a rhizoid (cf. light micrograph Fig. 2D). (E–G) Variations in plastid ultrastructure in mature cells in Bryum tenuisetum. (E) Caulonemal cell packed with elongate chloroplasts containing little or no starch. (F) Chloroplast with giant grana. Note the saccate cristae filling the adjacent elongate mitochondrion and the inactive dictyosomes. (G) Highly attenuated rhizoid plastid. Local swellings contain grana and lipid droplets. d, Dictyosome. Scale bars: (C–E) = 5 µm; (A, B, F, G) = 2 µm.

In mature cells the walls, usually highly pigmented under the light microscope (e.g. Fig. 1L, N, O), are 1·5–2·5 µm in thickness and comprise three more-or-less distinct layers (Fig. 8A–C). The original outer wall has a fibrillar granular appearance and is highly electron-opaque, while the middle layer deposited during cellular maturation is more uniformly fibrillar, less opaque and sometimes granular in the area closer to the plasmalemma (Fig. 8A). Plasmodesmata are very numerous in the end walls (Fig. 8B), where the wall–plasmalemmal interface may be highly irregular (Fig. 8B). Irregular invaginations of electron-transparent wall material often extend into the lumen at the extremities of the cells (Fig. 8C).

Selected examples of the range of plastid morphologies found in fully mature cells are illustrated in Fig. 8D–G. Typical of rhizoids, both wild and in culture, are higly elongate plastids generally with numerous starch grains (Fig. 8D) in comparison with more often starch-free plastids in caulonemata (Fig. 8E). The plastids tend to contain grana with generally less than ten thyakoids, although some may have much larger stacks extending across the full width of the organelles (Fig. 8F). Sometimes the plastids present highly irregular profiles with swollen areas, some packed with thylakoids, others containing plastoglobuli (Fig. 8G).

Effects of oryzalin and cytochalasin D on mature caulonema and chloronema cells (Table 1)

Table 1.

Effects of oryzalin and cytochalasin D on moss caulonemata and rhizoids compared with those on leptoids sensu stricto

Control
 10 µm oryzalin
 2 µm cytochalasin
Cellular features Caulonemata/rhizoids FCCs Caulonemata/rhizoids FCCs Caulonemata/rhizoids FCCs
Polarity Sometimes pronounced Pronounced Absent Pronounced Sometimes pronounced Pronounced
Microtubules
Endoplasmic MTs Long, numerous Long, numerous Short, scattered Short, associated with nuclear envelope only Long, numerous Long, numerous
Cortical MTs Frequent Sparse Absent Absent Frequent Sparse
Nucleus
Shape Spindle-shaped, with tubular extensions Spindle-shaped Irregular with tubular extensions Ovoid Spindle-shaped with tubular extensions Spindle-shaped
Position Usually central Distal Variable Distal Usually central Distal
Organelle alignment Pronounced Pronounced Lost completely Lost completely Reduced Pronounced
Endomembrane domains
ER Scattered single profiles of SER Abundant longitudinal sheets of RER Scattered single profiles of SER Abundant convoluted sheets of RER Scattered profiles of SER Abundant longitudinal sheets of RER
Crystalline inclusions Yes Yes Yes Yes Yes Yes
Golgi bodies Scattered with small vesicles Scattered with small vesicles Scattered with small vesicles Scattered with small vesicles Scattered with large vesicles Scattered with large vesicles
Trans-Golgi cisternae Sparse Abundant Sparse Sparse Sparse Abundant
Membrane-bounded vesicles Abundant, spheroidal 0·5–1·0 µm Abundant, spheroidal 0·5–1·0 µm Abundant, small >0·5 µm Abundant, large (3 µm), highly irregular Abundant, spheroidal 0·5–1·0 µm Abundant, spheroidal 0·5–1·0 µm
Plastids Elongate, with and without starch Elongate, starch-free Pleomorphic, starch-filled, often extremely large Pleomorphic, starch-free Elongate, with and without starch, often extremely large Elongate, starch-free
Mitochondria Usually elongate Elongate, spheroidal to ellipsoidal Spheroidal to ovoidal Pleomorphic Usually elongate Spheroidal to ovoidal

The profound effects of the treatment with 10 µm oryzalin on the cytology of mature cells are clearly visible under the light microscope (Fig. 9A, B). The longitudinal alignment of the plastids is lost completely in both rhizoids and caulonemata (Fig. 9A) and many cells, particularly in chloronema filaments, contain giant pleomorphic plastids (Fig. 9B).

Fig. 9.

Fig. 9.

Mature protonemata of Funaria hygrometrica treated with drugs. (A–E) 10 µm oryzalin. (A, B) Light micrographs. (A) Caulonema showing clumped plastids with no longitudinal alignment. (B) Caulonema with giant pleomorphic plastids. (C–E) Electron micrographs. (C) Highly pleomorphic nucleus. Note that the tubular extensions of the nuclear envelope remain (arrowed). (D) Caulonema with giant starch-filled plastid and no longitudianl arrangement of the ER. (E) Giant chloroplast with abundant starch grains in a chloronemal cell. (F–K) 2 µm cytochalasin D. (F, G) Light micrographs. (F) Longitudinal alignment of the organelles and polarity are retained in the caulonema (F) but some cells contain giant plastids (G). (H–L) Electron micrographs. (H) Giant starch-filled caulonemal plastid. (I) Numerous longitudinally aligned microtubules (arrowed) associated with the nuclear envelope. (J) Longitudinally aligned microtubules (arrowed) associated with vesicles and a dictyosome with large vesicles. (K) Detail of dictyosome from (J). (L) Inactive dictyosome from an untreated cell. d, Dictyosome. Scale bars: (A, B, F, G) = 20 µm; (C-E, H) = 5 µm; (I, J) = 2 µm; (K, L) = 0·5 µm.

At the ultrastructural level endoplasmic microtubules vanish in all but a few cells, where short lengths remain scattered in the cytoplasm with no preferred orientation (not illustrated). The cytoplasmic organization is greatly perturbed, with a loss of the longitudinal organelle alignment. The shape of the nuclei ranges from ovoid to highly irregular (Fig. 9C) but the long tubular extensions of the nuclear envelope found in untreated cells remain unchanged, although often tracing a more undulating course (Fig. 9C). The plastids are scattered in the cytoplasm with no preferred orientation and are often massive, highly irregular in outline and packed with large starch grains (Fig. 9D, E). These giant plastids may be up to 40 µm in length and more than 10 µm in width. The ER system is almost completely disrupted, leaving only few scattered and short profiles with no preferred orientation (Fig. 9C, D), whilst the Golgi system shows no apparent alteration relative to untreated specimens (Fig. 9L). Cytoplasmic vesicles are scattered randomly through the cytoplasm.

After treatment with 2 µm cytochalasin D the cells retain both the longitudinal alignment of the organelles and polarity, as in the controls (Fig. 9F). The microtubule cytoskeleton appears unaffected by the treatment, with numerous longitudinally aligned microtubules clearly visible along the nuclear envelope (Fig. 9I) and in association with vesicles in the cytoplasm (Fig. 9J). The nuclei, mitochondria and the ER network are generally indistinguishable from the controls and the tubular extensions of the nuclear envelope remain (not illustrated). The treatment does, however, lead to the appearance of giant plastids in a few cells of both caulonemata and chloronemata (Fig. 9G). These plastids are up to 35 µm in length, and their envelopes are greatly distended because of the accumulation of particularly large starch deposits (Fig. 9H). Golgi bodies are now associated with large vesicles (cf. Fig. 9K and L). The crystalline inclusions in the ER are unaffected by either treatments (not illustrated).

DISCUSSION

This study shows that cellular differentiation in caulonemata and rhizoids, but not in chloronemata, follows much the same pattern and involves remarkable cytological changes including several novel events, some of which show parallels with processes occuring during sieve element development in tracheophytes (Table 2). The cytological reorganization from highly vacuolate cells to vacuole-less cells with strands of endoplasmic microtubules and longitudinally aligned organelles involves profound changes in virtually every component except the microbodies. Whereas the cytological events in apical and subapical cells are principally associated with nuclear and cell division, producing new apical cells and side-branch initials respectively, the cytological changes occurring during maturation in caulonemal and rhizoid cells probably reflect specialization in solute transport. The present study, therefore, provides further support for the notion that caulonemata and rhizoids are morphologically and functionally equivalent structures, the only major difference being their association respectively with the juvenile and mature phase of moss gametophytes (Goode et al., 1992; Duckett et al., 1998).

Table 2.

Comparison of the principle features of mature moss rhizoids and caulonemata, food-conducting cells in moss gametophores and sporophytes and sieve elements in lower tracheophytes

Cellular feature Caulonemata and rhizoids Food-conducting cells1 Sieve elements2
Polarity Sometimes pronounced Pronounced Absent
Microtubules
Endoplasmic Long, numerous Long, numerous Absent
Cortical Frequent Sparse Absent
Nucleus
Endoreduplication Yes Most likely Yes, before degeneration
Degeneration No Sometimes Yes
Shape Spindle-shaped, with tubular extensions Spindle-shaped N/A
Position Usually central, occasionally distal Distal N/A
Longitudinal organelle alignment Pronounced Pronounced Absent
Cell wall
Multilayered Yes Yes Yes
Inner layer irregular Yes Yes Yes
Specialized plasmodesmata No Yes Yes
Desmotubules Present Present Absent
Tonoplast breakdown Yes Probable Yes
p-protein No No Yes
Ribosomes Free, aggregated around plastids and mitochondria Mainly free, aggregated around plastids and mitochondria Absent
Endomembrane domains
ER Scattered, single profiles of SER Abundant, longitudinal sheets of rough/smooth ER Abundant, stacks of SER in some taxa
Crystalline inclusions* Yes, scattered Yes, prominent* Yes, prominent*
Golgi bodies Scattered with small vesicles Scattered with small vesicles Absent
Trans-Golgi network Sparse Abundant Sparse or absent
Vesicles Abundant, spheroidal 0·5–1·0 µm Abundant, spheroidal 0·5–1·0 µm Absent
Membrane-bounded tubules Scarce Abundant Absent
Regeneration No No No
Plastids Most commonly elongate, with or without starch Elongate, starch-free; starch-filled in conducting parenchyma Ovoid, starch-free
Mitochondria Usually elongate (associated with 1 MT) Elongate, spheroidal to ellipsoidal (associated occasionally with several MTs) Spheroidal to ellipsoidal (no association with MTs)
Desiccation Major cytological reorganization Major cytological reorganization Accumulation of non-reducing sugars – trehalose?

Just as the proper characterization of tip-cell organization was an absolute prerequisite for the recent demonstration of a common role of helix-loop-transcription factors in controlling the development of root hairs in Arabidopsis and rhizoids in Physcomitrella (Menand et al., 2007b), the cytological data in the present study open the way for specifically targeted molecular studies on, for example, auxin transporter genes (Floyd and Bowman, 2007) and plastid replication (Gremillon et al., 2007).

Developmental considerations

The starting point for caulonemal and rhizoid differentiation is a highly vacuolate cell similar to those forming chloronemata (Fig. 1). Most of the cytoplasm and organelles either remain in the apical derivative, following apical cell division, or are subsequently sequestered into the nascent side-branch initials.

Endoplasmic reticulum

The earliest cytological change associated with the onset of differentiation is the appearance of tubular ER profiles containing crystalline inclusions; these persist in the fully differentiated cells. McCauley and Hepler (1992) reported crystalline inclusions in the lumen of cortical ER of freeze-substituted caulonema cells of Funaria hygrometrica but not in branch initial cells or buds. The present study confirms that these structures are unique to caulonemal cells within the protonemal system.

Similar electron-opaque granular inclusions have been reported in the vacuoles and membranous tubules of moss food-conducting cells (Hébant 1970; 1972; Ligrone and Duckett, 1994). Hébant (1970, 1972) interpreted these structures as being ‘globules denses’ (refractive globules or spherules), commonly found in the sieve elements of lower vascular plants (Kruatrachue and Evert, 1974; Perry and Evert, 1975). In the leaf of Isoetes, accumulation of crystalline and fibrillar material in RER cisternae is unique to developing sieve elements and distinguishes the latter from parenchyma cells (Kruatrachue and Evert, 1974).

Whilst during differentiation the ER network increases markedly and consists mainly of RER, at maturity it decreases and becomes predominantly smooth, with the majority of ribosomes forming conspicuous aggregates around various organelles, and in particular the plastids. This shift from a predominantly RER to a mainly SER system suggests that mature caulonemal and rhizoid cells are highly stable and that the intense metabolic activity, which characterizes their development, slows down considerably once differentiation is complete. A similar shift from a predominantly RER to a smooth-surfaced ER is also seen in developing sieve elements in lycopods (Burr and Evert, 1973; Kruatrachue and Evert, 1974).

Vacuole and vesicles

Following the appearance of ER inclusions, the next major event in differentiation is the fragmentation and disappearance of the tonoplast, leading to mixing of the vacuolar contents with the cytoplasm as seen in sieve element differentiation (Kruatrachue and Evert, 1974).

Concomitant with tonoplast disintegration is the appearance of long, straight or cup-shaped RER profiles associated with numerous vesicles and stacks of vesicles. The close association and apparent continuity between the ER profiles and the vesicles and vacuoles (and the absence of dictyosomes from their immediate vicinity) suggests that these originate from the ER (Fig. 6C–E). We interpret these micrographs as the generation of stacks of flattened vesicles from the ER that then disperse and expand. The lunate vesicular profiles in close proximity to the stacks are most likely an intermediate stage between the highly flattened and compressed state of the vesicles in the stacks and their full expansion leading to a typical rounded shape. Thus, the various vacuoles found in mature caulonemal and rhizoid cells are most likely derived de novo from the ER and are not the result of fragmentation of the large single vacuole seen at the onset of differentiation.

‘Stacks’ of vesicles, virtually indistinguishable from those described here, are also present in close association with the ER in FCCs of mosses (see Fig. 2B in Ligrone and Duckett, 1994). Jensen and Jensen (1984) reported dilations of the ER in protonema tip cells comparable with the RER enlargements implicated in vacuole formation in other tip-growing systems, e.g. root tip cells, pollen tubes and laticifer cells (Jensen and Jensen, 1984 and citations therein) and argued that in protonemal tip cells the ER is also involved in vacuole formation.

Microtubule cytoskeleton

In differentiating caulonemal and rhizoid cells the change in organelle distribution, from apparently randomly arranged to longitudinally aligned, coincides with the genesis of endoplasmic microtubules. It may therefore be envisaged that, while the position of the organelles throughout the cell lumina is a direct consequence of the disappearance of the original large vacuole, their characteristic longitudinal alignment, as noted previously by Schnepf and Quader (1987) and Duckett et al. (1998), correlates directly with their association with the axial arrays of endoplasmic microtubules. These associations are remarkably similar to those found in FCCs of bryoid mosses (Tables 1 and 2; Ligrone and Duckett, 1994; 1996) and Sphagnum (Ligrone and Duckett, 1998). The only differences are the greater abundance of cortical microtubules and just a single microtubule associated with each mitochondrion in caulonemal and rhizoid cells.

Extensive light microscopy observations of living cells confirm that cytoplasmic streaming does not occur in moss filament cells; time-lapse photography reveals the overall static position of the plastids in mature cells. This finding is not altogether surprising since cytoplasmic strands associated with streaming in plant cells almost invariably contain actin ‘cables’ (Kachar and Reese, 1988). Thus the close association between MTs and plastids, mitochondria and other organelles in mature caulonemal and rhizoid cells determines their position and alignment but is not involved in movement.

The nucleus

During differentiation of these cells, the changes in nuclear shape from spherical and ovoid to highly elongate and spindle-shaped closely mirror those in food-conducting cells (R. Ligrone and J. G. Duckett, pers obs.). Further studies on the latter are now needed to establish whether these morphological changes reflect endoreduplication, as previously demonstrated for caulonemata (Kingham et al., 1995).

One of the most striking features of caulonemal and rhizoid cell differentiation is that as the nucleus becomes elongate and associates with endoplasmic MTs, the nuclear envelope (NE) at both poles forms long, linear tubules that extend towards the ends of the cells. Similar extensions of the nuclear envelope, although never as conspicuous as in protonemal cells, are also present in FCCs in the leafy shoot of mosses (S. Pressel et al., pers obs.).

The cell wall

Although the possibility that dictyosomes contribute to the vesicle and vacuole system in differentiating caulonemal and rhizoid cells cannot be excluded, their principal role is most likely the production of wall materials. Dictyosomes remain quiescent throughout vesicle generation, with large vesicles appearing only during the subsequent period of wall thickening. When the additional wall layers are being deposited the dictysomes are concentrated in close proximity to the cell wall, whereas the vesicle/vacuole apparatus is located centrally. Tubules and fenestrated lamellar elements of the cortical ER exhibit an intimate association with the plasma membrane, as previously described by McCauley and Hepler (1992) in caulonema cells of Funaria hygrometrica. McCauley and Hepler (1992) suggested that such an intimate association might indicate that the ER is anchored in some way to the plasma membrane. Another likely explanation is that the cortical ER is also involved, together with the dictyosomes, in the deposition of the additional cell wall layers. The extent of the cortical ER progressively decreases following the thickening of the cell wall and is inconspicuous in fully differentiated caulonema cells. Similarly, the population of dictyosomes and dictyosome vesicles declines at maturity.

The multilaminate walls of caulonemata, rhizoids and food-conducting cells are remarkably similar, even extending to the irregular ingrowths (Fig. 8C). A unique feature of the caulonemata and rhizoids, however, is the mucilage sheath. This hitherto unsuspected component of caulonemal and rhizoid cell walls is largely transparent under the light microscope and therefore is distinct from rhizoid papillosity. This latter feature, extensively used by taxonomists (Crundwell, 1979), develops early in differentiation and consists of the same material forming the original unthickened wall. The most obvious function of the mucilage sheath is protection against desiccation but involvement in other functions, including interaction with soil micro-organisms, is also possible.

Functional considerations

Inhibitor studies (Table 1)

The present results indicate that, as in FCCs of mosses (Ligrone and Duckett, 1996), the highly characteristic cytological organization of fully developed caulonemal and rhizoid cells is intrinsically dependent on their cytoskeleton, with distinctly different roles for the microtubules and microfilaments. Removal of the microtubules by oryzalin treatment results in a total loss of the organelle longitudinal alignment, with the plastids tending to aggregate in clumps, a phenomenon previously described by Wacker et al. (1988). Thus, as for moss FCCs, it may be concluded that organelle alignment is dependent on microtubules.

Oryzalin also induces changes in organelle shape, with plastids, mitochondria and nuclei all being affected. In particular the nucleus, following removal of the microtubule cytoskeleton, loses its spindle shape and develops highly irregular contours. Noting that the microtubules radiating from the nuclear poles not only extend into the cytoplasm but also interact with non-polar regions of the nuclear envelope, Ligrone and Duckett (1994) suggested that, in moss FCCs, microtubules might be responsible for nuclear shaping. The present data suggest a similar role for the microtubules in mature caulonemal and rhizoid cells.

In untreated materials the tubular extensions of the nuclear envelope are closely associated with endoplasmic microtubules (Fig. 7A, B) and thus they would appear to be dependent on these for their formation. However, their persistence following both oryzalin and cytochalasin treatments (Fig. 9D) indicates that, once formed, their structural integrity is not dependent on microtubules nor microfilaments. One obvious possibility for future investigation is that the nuclear extensions are maintained by intermediate filaments/nuclear lamins (Aebi et al., 1986; Fisher et al., 1986; Goldman et al., 2002), implying that these extensions have the exclusively structural role of anchoring the nucleus in a more-or-less central location.

Changes in the endomembrane system after oryzalin treatment, in particular a reduction in the ER, also resemble closely the results reported by Ligrone and Duckett (1996) in moss FCCs. The highly irregular appearance of the vacuoles and the proliferation of small spherical vesicles, especially in proximity to the ER, are clearly indicative of a close relationship between the ER, vesicles and vacuoles and the microtubule cytoskeleton.

In contrast to the major changes following oryzalin treatment, those with cytochalasin (Table 1) are more-or-less limited to the accumulation of large vesicles adjacent to the dictyosomes. Thus, as in moss FCCs (Ligrone and Duckett, 1996), actin arrays do not appear to determine organelle alignment but, as in caulonemal tip cells (Finka et al., 2007), are probably involved in the transport of secretory vesicles from dictyosomes to the cell periphery. It should be noted, however, that visualization of actin with GFP-talin (Finka et al., 2007; Fig. 5I) reveals colocalization with the chloroplasts in differentiated caulonemal cells.

Giant plastids

By far the most striking and, at first sight, unexpected result from the inhibitor treatments was the formation of giant plastids, particularly in chloronemal cells. Previous studies have already implicated both microtubules and microfilaments in plastid replication in protonemata. Abel et al. (1989) reported the presence of one single, large, plate-like chloroplast per protonema cell in the PC22 mutant of Physcomitrella patens. This mutant was found to be deficient in chloroplast division and protonemal differentiation. Since the actin, but not the microtubular cytoskeleton, appeared to be more irregular than in the wild type, the defective chloroplast division was attributed to deficiencies in the actin cytoskeleton. Numerous minute chloroplasts are produced in protonemal cells of Sphagnum exposed to colchicine or oryzalin (Goode et al., 1993b). Because plastid numbers were unaffected by either CIPC (isopropyl N-3-chlorophenyl carbamate) or cytochalasins, it was concluded that oryzalin and colchicine affect plastids via microtubule-organizing centres located on the plastid envelope and that the microtubules are essential for both replication and positioning of the plastids.

Several isoforms of the protein FtsZ, regarded as the ancestor of tubulin and essential for cell division in prokaryotes, have recently been identified in land plants (Kiessling et al., 2004 and citations therein) and are involved in chloroplast division as well as maintenance of chloroplast shape and integrity in Physcomitrella patens (Suppanz et al., 2007 and citations therein). It may well be, therefore, that the presence of giant plastids in oryzalin-treated protonemata is due to inhibition of FtsZ proteins, and hence plastid division, by oryzalin. However, this cannot explain the formation of giant plastids in cytochalasin-treated filaments; an alternative scenario for the origin of giant plastids following our inhibitor studies is that these may arise not by misdivision but by fusion (as occurs during spermatogenesis and sporogenesis in mosses and in a range of tissues in ferns; Duckett and Ligrone, 1993). We envisage that, in untreated specimens, spatial separation between the plastids is maintained by microtubules and particularly microfilaments. Their absence would allow the plastids to come into close proximity, thus predisposing them to fusion.

When Ligrone and Duckett (1994) first decribed the cytology now well established as characteristic of FCCs in bryophytes (Ligrone et al., 2000) they suggested that long-distance solute transport might be via vesicle trafficking along the axial arrays of microtubules. Subsequent investigations (Schmid, 1998) revealed that not only was solute transport much faster than vesicle trafficking but that it was also completely unaffected by oryzalin, which disrupted the microtubule system in FCCs (Ligrone and Duckett, 1996) in an almost identical manner as reported here for caulonemata and rhizoids. Thus it was concluded that solute transport in bryophyte food-conducting cells was by mass flow as in phloem, but no alternative explanation was offered for the unique cytology of these cells. More recent experiments on moss gametophores have showed that the microtubule/vesicle apparatus was intimately associated with the ability of FCCs to withstand desiccation (Pressel et al., 2006). Similar de- and rehydration experiments are now indicating that the same is also true for rhizoids and caulonemata, as expected from their closely similar cytology (Pressel, 2007).

In addition to the suite of structural commonalities detailed above (and summarized in Table 1), further features indicative of such a role in desiccation include the highly irregular plasmalemma–wall interface, which might protect the cells from plasmolysis, and the abundance of microbodies, possibly associated with the removal of superoxide radicals produced during desiccation stress (Smirnoff, 1993; Minibayeva and Beckett, 2001; Mayaba et al., 2002). The presence of abundant and metabolically active mitochondria, as evidenced by their electron-opaque matrix and saccate cristae, throughout cellular differentiation reflects a major requirement for ATP (presumably for both solute transport and recovery from desiccation) even at maturity and throughout the length of the filaments. Retention of more organelles in mature FCCs and caulonemal cells than in sieve elements probably reflects the dual and conflicting roles of solute transport and survival through desiccation.

A final feature common to caulonemal cells and FCCs is that both are incapable of regeneration. Although excised mature rhizoid and caulonemal filaments can produce new filaments these are always from dormant side-branch initals and never from mature cells. Similarly regeneration from excised regions of moss stems is always from cortical parenchyma cells (Clymo and Duckett, 1986) and never from more internal cells including FCCs. Kingham et al. (1995) attributed the loss of totipotency along protonemata to endoreduplication. This is probably not the only cause; disintegration of the tonoplast is most likely the point at which differentiation becomes irreversible.

A question of homology

Setting aside the endoplasmic microtubule/vesicle apparatus in moss FCCs, caulonemata and rhizoids as features most likely associated with poikilohydry (Pressel et al., 2006), the remarkable list of common features between these and sieve elements not set out previously (Table 2) raises the important issue of possible homology. Do similarities imply homology or simply reflect an independent evolution of mass-flow transport systems, as is certainly the case in brown algal ‘sieve tubes’ (Raven, 2003) and fungal rhizomorphs (Lew, 2005)? Just as the loss of cell contents reduces the resistance to water flow in water-conducting cells by several orders of magnitude (Raven, 1993), it is highly likely that the mixing of vacuole contents and cytoplasm following tonoplast rupture reduces the viscosity of the cytosol in both moss food-conducting cells and sieve elements. The general electron transparency of the cytoplasm of both these elements supports this notion. Although an unusual event in plant differentiation, on its own tonoplast disintegration cannot be considered evidence for homology. The presence of p-protein bodies in the cytoplasm of young sieve elements and maturational nuclear degeneration versus the absence of these features in mosses argue against homology (van Bel, 2003). In addition, from a functional standpoint, the presence of desmotubules and constricted ends in the plasmodesmata in moss food-conducting cells might be a major impediment to mass flow (Raven, 2003). Thus, from our current state of knowledge the similarities between moss food-conducting cells and sieve elements are best considered an example of homeoplasy. Searches in the Physcomitrella genome for sequences diagnostic of vascular plant phloem would now seem highly appropriate.

ACKNOWLEDGEMENTS

Silvia Pressel thanks NERC for financial support from a CASE studentship with the Royal Botanic Gardens, Kew. The authors thank K. Pell (QMUL) for technical assistance and Karen Renzaglia for the use of electron microscope facilities at the University of Southern Illinois.

LITERATURE CITED

  1. Abel WO, Knebel W, Koop H-U, Marienfeld JR, Quader H, Reski R, Schnepf E, Spörlein B. A cytokinin-sensitive mutant of the moss, Physcomitrella patens, defective in chloroplast division. Protoplasma. 1989;152:1–13. [Google Scholar]
  2. Aebi U, Cohn J, Buhle L, Gerace L. The nuclear lamina is a meshwork of intermediate-type filaments. Nature. 1986;323:560–564. doi: 10.1038/323560a0. [DOI] [PubMed] [Google Scholar]
  3. van Bel AJE. The phloem, a miracle of ingenuity. Plant, Cell and Environment. 2003;26:125–149. [Google Scholar]
  4. Burr FA, Evert RF. Some aspects of sieve-element structure and development in Selaginella kraussiana. Protoplasma. 1973;78:81–97. [Google Scholar]
  5. Clymo RS, Duckett JG. Regeneration of Sphagnum. New Phytologist. 1986;102:589–614. [Google Scholar]
  6. Conrad PA, Steucek GL, Hepler PK. Bud formation in Funaria: organelle redistribution following cytokinin treatment. Protoplasma. 1986;131:211–223. [Google Scholar]
  7. Cove DJ, Knight CD, Lamparter T. Mosses as model systems. Trends in Plant Science. 1997;2:99–105. [Google Scholar]
  8. Cove DJ, Bezanilla M, Harries P, Quatrano R. Mosses as model systems for the study of metabolism and development. Annual Review of Plant Biology. 2006;57:497–520. doi: 10.1146/annurev.arplant.57.032905.105338. [DOI] [PubMed] [Google Scholar]
  9. Crundwell AC. Rhizoids and moss taxonomy. In: Clarke GCS, Duckett JG., editors. Bryophyte systematics. London: Academic Press; 1979. pp. 347–363. [Google Scholar]
  10. Cuming AC, Cho SH, Kamisugi Y, Graham H, Quatrano RS. Microarray analysis of transcriptional responses to abscisic acid and osmotic, salt and drought stress in the moss. Physcomitrella patens. New Phytologist. 2007;176:275–287. doi: 10.1111/j.1469-8137.2007.02187.x. [DOI] [PubMed] [Google Scholar]
  11. Doonan JG. The cytoskeleton in moss morphogenesis. In: Lloyd CE., editor. The cytoskeletal basis of plant growth and form. London: Academic Press; 1991. pp. 289–301. [Google Scholar]
  12. Doonan JH, Duckett JG. The bryophyte cytoskeleton: experimental and immunofluorescence studies of morphogenesis. Advances in Bryology. 1988;3:1–31. [Google Scholar]
  13. Doonan JH, Jenkins GI, Cove DJ, Lloyd CW. Microtubules connect the migrating nucleus to the prospective division site during side branch formation in the moss, Physcomitrella patens. European Journal of Cell Biology. 1986;41:157–164. [Google Scholar]
  14. Doonan JH, Cove DJ, Lloyd CW. Microtubules and microfilaments in tip growth: evidence that microtubules impose polarity on protonemal growth in Physcomitrella patens. Journal of Cell Science. 1988;89:533–540. [Google Scholar]
  15. Evert RF. Seedless vascular plants. In: Behnke H-D, Sjolund RD., editors. Sieve elements. Comparative structure, induction and development. Berlin: Springer; 1989. pp. 33–62. [Google Scholar]
  16. Duckett JG, Ligrone R. A survey of diaspore liberation mechanisms and germination patterns in mosses. Journal of Bryology. 1992;17:335–354. [Google Scholar]
  17. Duckett JG, Ligrone R. Plastid dividing rings in ferns. Annals of Botany. 1993;72:619–627. [Google Scholar]
  18. Duckett JG, Ligrone R. The formation of catenate foliar gemmae and the origin of oil bodies in the liverwort Odontoschisma denudatum (Mart). Dum. Jungermmaniales: a light and electron microscope study. Annals of Botany. 1995;76:405–419. [Google Scholar]
  19. Duckett JG, Ligrone R. A comparative cytological analysis of fungal endophytes in the sporophyte rhizomes and vascularized gametophytes of Tmesipteris and Psilotum. Canadian Journal of Botany. 2005;83:1443–1456. [Google Scholar]
  20. Duckett JG, Schmid AM, Ligrone R. Protonemal morphogenesis. In: Bates JW, Ashton NW, Duckett JG., editors. Bryology for the Twenty-first Century. Leeds, UK: Maney and the British Bryological Society; 1998. pp. 223–246. [Google Scholar]
  21. Duckett JG, Fletcher P, Francis R, Matcham HW, Read JT, Russell AJ, Pressel S. In vitro cultivation of bryophytes; practicalities, progress, problems and promise. Journal of Bryology. 2004;26:3–20. [Google Scholar]
  22. Finka A, Schaefer DG, Saidi Y, Goloubinoff P, Zrÿd J-P. In vivo visualization of F-actin structures during development of the moss Physcomitrella patens. New Phytologist. 2007;174:63–76. doi: 10.1111/j.1469-8137.2007.01989.x. [DOI] [PubMed] [Google Scholar]
  23. Fisher D, Chaudhary N, Blobel G. cDNA sequencing of nuclear lamins A C reveals primary secondary structural homology to intermediate filament proteins. Proceedings of the National Academy of Science of the USA. 1986;83:6450–6454. doi: 10.1073/pnas.83.17.6450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Floyd SK, Bowman JL. The ancestral developmental tool kit of land plants. International Journal of Plant Science. 2007;168:1–35. [Google Scholar]
  25. Frank W, Ratnadewi D, Reski R. Physcomitrella patens is highly tolerant against drought, salt and osmotic stress. Planta. 2005;220:384–394. doi: 10.1007/s00425-004-1351-1. [DOI] [PubMed] [Google Scholar]
  26. Goldman RD, Gruenbau Y, Moir RD, Shumaker DK, Spann TP. Nuclear lamins: building blocks of nuclear architecture. Genes and Development. 2002;16:533–547. doi: 10.1101/gad.960502. [DOI] [PubMed] [Google Scholar]
  27. Goode JA, Duckett JG, Stead AD. Towards an understanding of developmental relationships between chloronema, caulonema, protonemal plates and rhizoids in mosses; a comparative study. Cryptogamic Botany. 1992;3:50–59. [Google Scholar]
  28. Goode JA, Alfano F, Stead AD, Duckett JG. The formation of aplastidic abscission (Tmema) cells and protonemal disruption in the moss Bryum tenuisetum Limpr. is associated with transverse arrays of microtubules and microfilaments. Protoplasma. 1993;174:158–172. [Google Scholar]
  29. Goode JA, Duckett JG, Stead AD. Protonemal morphogenesis in Sphagnum: the cytoskeleton and effects of growth regulators. Advances in Bryology. 1993;5:129–151. [Google Scholar]
  30. Gremillon L, Kiessling J, Hause B, Decker E, Reski R, Sarnighausen E. Filamentous temperature sensitive Z (FtsZ) isoforms specifically interact in the chloroplasts and in the cytosol of Physcomitrella patens. New Phytologist. 2007;176:299–310. doi: 10.1111/j.1469-8137.2007.02169.x. [DOI] [PubMed] [Google Scholar]
  31. Hébant C. Aspects infrastructuraux observés au cours de la différenciation du phloèm (leptome) dans la tige feuillée de quelques Mousses Polytrichales. Comptes rendu hebdomadaire des séances de l' Académie des Sciences. 1970;271:1361–1363. [Google Scholar]
  32. Hébant C. Les tissues conducteurs des Bryophytes. L'Information Scientifique. 1972;5:207–219. [Google Scholar]
  33. Hofmann AH, Codon AC, Ivascu C, Russo VEA, Knight C, Cove D, et al. A specific member of the Cab multigene family can be efficiently targeted and disrupted in the moss Physcomitrella patens. Molecular and General Genetics. 1999;261:92–99. doi: 10.1007/s004380050945. [DOI] [PubMed] [Google Scholar]
  34. Jensen LCW. Division, growth and branch formation in protonema of the moss Physcomitrium turbinatum: studies of sequential cytological changes in living cells. Protoplasma. 1981;107:301–307. [Google Scholar]
  35. Jensen LCW, Jensen CG. Fine structure of protonemal apical cells of the moss Physcomitrium turbinatum. Protoplasma. 1984;122:1–10. [Google Scholar]
  36. Kachar B, Reese TS. The mechanism of cytoplasmic streaming in characean algal cells: sliding of endoplasmic reticulum along actin filaments. The Journal of Cell Biology. 1988;106:1545–1552. doi: 10.1083/jcb.106.5.1545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Kiessling J, Martin A, Gremillon L, Rensing SA, Nick P, Sarnighausen E, Decker E, Reski R. Dual targeting of plastid division protein FtsZ to chloroplasts and the cytoplasm. EMBO Reports. 2004;5:889–894. doi: 10.1038/sj.embor.7400238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kingham KI, Duckett JG, Glyn MC, Leitch AR. Nuclear differentiation in the filamentous caulonema of the moss Funaria hygrometrica. New Phytologist. 1995;131:543–556. doi: 10.1111/j.1469-8137.1995.tb03090.x. [DOI] [PubMed] [Google Scholar]
  39. Kingham KI, Duckett JG, Gasdova B, Kovarik A, Bezdek M, Leitch AR. The role of DNA methylation in nuclear and cell differentiation in the filamentous caulonema of the moss Funaria hygrometrica. New Phytologist. 1997;138:567–577. [Google Scholar]
  40. Klekowski EJ., Jr. Reproductive biology of the Pteridophyta. II. A study of the Blechnaceae. Journal of the Linnean Society, Botany. 1969;62:361–377. [Google Scholar]
  41. Knight C, Cove DJ, Boyd PJ, Ashton NW. The isolation of biochemical and developmental mutants in Physcomitrella patens. In: Glime JM., editor. Methods in bryology. Nichinan, Japan: Hattori Botanical Laboratory; 1988. pp. 47–58. [Google Scholar]
  42. Kruatrachue M, Evert RF. Structure and development of sieve elements in the leaf of Isoetes muricata. American Journal of Botany. 1974;61:253–266. [Google Scholar]
  43. Lew RR. Mass flow and pressure-driven hyphal extension in Neurospora crassa. Microbiology. 2005;151:2685–2692. doi: 10.1099/mic.0.27947-0. [DOI] [PubMed] [Google Scholar]
  44. Ligrone R, Duckett JG. Cytoplasmic polarity and endoplasmic microtubules associated with the nucleus and organelles are ubiquitous features of food-conducting cells in bryoid mosses (Bryophyta) New Phytologist. 1994;127:601–614. [Google Scholar]
  45. Ligrone R, Duckett JG. Polarity and endoplasmic microtubules in food-conducting cells of mosses: an experimental study. New Phytologist. 1996;134:503–516. [Google Scholar]
  46. Ligrone R, Duckett JG. The leafy stems of Sphagnum (Bryophyta) contain highly differentiated polarized cells with axial arrays of endoplasmic microtubules. New Phytologist. 1998;140:567–579. doi: 10.1111/j.1469-8137.1998.00298.x. [DOI] [PubMed] [Google Scholar]
  47. Ligrone R, Duckett JG, Renzaglia KS. Conducting tissues and phyletic relationships of bryophytes. Philosophical Transactions of the Royal Society of London, Series B. 2000;355:795–813. doi: 10.1098/rstb.2000.0616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Mayaba N, Minibayeva F, Beckett RP. An oxidative burst of hydrogen peroxide during rehydration following desiccation in the moss Atrichum androgynum. New Phytologist. 2002;155:275–283. [Google Scholar]
  49. McCauley MM, Hepler PK. Cortical ultrastructure of freeze-substituted protonemata of the moss Funaria hygrometrica. Protoplasma. 1992;169:168–178. [Google Scholar]
  50. Menand B, Calder G, Dolan L. Chloronemal and caulonemal cells both expand by tip growth in the moss Physcomitrella patens. Journal of Experimental Botany. 2007;58:1843–1849. doi: 10.1093/jxb/erm047. [DOI] [PubMed] [Google Scholar]
  51. Menand B, Keke Y, Jouannic S, Hoffmann L, Ryan E, Linstead P, Schaefer DG, Dolan L. An ancient mechanism controls the development of cells with a rooting function in land plants. Science. 2007;316:1477–1480. doi: 10.1126/science.1142618. [DOI] [PubMed] [Google Scholar]
  52. Minibayeva F, Beckett RP. High rates of extracellular superoxide production in bryophytes and lichens, and an oxidative burst in response to rehydration following desiccation. New Phytologist. 2001;152:333–341. [Google Scholar]
  53. Perry JW, Evert RF. Structure and development of the sieve elements in Psilotum nudum. American Journal of Botany. 1975;62:1038–1052. [Google Scholar]
  54. Pressel S. Experimental studies of bryophyte cell biology, conservation and physiology. University of London; 2007. Ph.D. Dissertation. [Google Scholar]
  55. Pressel S, Ligrone L, Duckett JG. Effects of de- and rehydration on food-conducting cells in the moss Polytrichum formosum Hedw.: a cytological study. Annals of Botany. 2006;98:67–76. doi: 10.1093/aob/mcl092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Quader H, Schnepf E. Actin filament array during side brand initiation in protonema cells of the moss Funaria hygrometrica: an actin organizing center at the plasma membrane. Protoplasma. 1989;151:167–170. [Google Scholar]
  57. Raven JA. The evolution of vascular plants in relation to quantitative functioning of dead water-conducting cells and stomata. Biological Reviews. 1993;68:337–363. [Google Scholar]
  58. Raven JA. Long-distance transport in non-vascular plants. Plant, Cell and Environment. 2003;26:73–85. [Google Scholar]
  59. Rensing SA, Lang D, Zimmer AD, Terry A, Salamov A, Shapiro H, et al. The Physcomitrella genome reveals evolutionary insights into the conquest of land by plants. Science. 2008;319:64–69. doi: 10.1126/science.1150646. [DOI] [PubMed] [Google Scholar]
  60. Renzaglia KS, Duff RJ, Ligrone R, Shaw J, Mishler BD, Duckett JG. Bryophyte phylogeny: advancing the molecular and morphological frontiers. Bryologist. 2007;110:179–213. [Google Scholar]
  61. Reski R. Physcomitrella and Arabidopsis: the David and Goliath of reverse genetics. Trends in Plant Science. 1998;3:209–210. [Google Scholar]
  62. Reski R. Molecular genetics of Physcomitrella. Planta. 1999;208:301–309. [Google Scholar]
  63. Sawidis T, Quader H, Bopp M, Schnepf E. Presence and absence of the preprophase band of microtubules in moss protonemata: a clue to understanding its function? Protoplasma. 1991;163:156–161. [Google Scholar]
  64. Schaefer DG. Gene targeting in Physcomitrella patens. Current Opinion in Plant Biology. 2001;4:143–150. doi: 10.1016/s1369-5266(00)00150-3. [DOI] [PubMed] [Google Scholar]
  65. Schaefer DG, Zrÿd J-P. Efficient gene targeting in the moss Physcomitrella patens. The Plant Journal. 1997;11:1195–1206. doi: 10.1046/j.1365-313x.1997.11061195.x. [DOI] [PubMed] [Google Scholar]
  66. Schmid A. A new photoassimilate translocation mechanism in the Polytrichales? University of London; 1998. Ph.D. Dissertation. [Google Scholar]
  67. Schmiedel G, Schnepf E. Side branch formation and orientation in the caulonema of the moss Funaria hygrometrica: normal development and fine structure. Protoplasma. 1979;100:367–383. [Google Scholar]
  68. Schmiedel G, Schnepf E. Side branch formation and orientation in the caulonema of the moss Funaria hygrometrica: experiments with inhibitors and with centrifugation. Protoplasma. 1979;101:47–59. [Google Scholar]
  69. Schmiedel G, Schnepf E. Polarity and growth of caulonema tip cells of the moss Funaria hygrometrica. Planta. 1980;147:405–413. doi: 10.1007/BF00380180. [DOI] [PubMed] [Google Scholar]
  70. Schnepf E. Morphogenesis in moss protonemata. In: Lloyd CW., editor. The cytoskeleton in plant growth and development. New York: Academic Press; 1982. pp. 321–344. [Google Scholar]
  71. Schnepf E, Quader H. Functions of microtubules in plant cells. In: Wohlfarth-Bottermann KE., editor. Nature and function of cytoskeletal proteins in motility and transport. Vol. 34. New York: Gustav Fisher Verlag; 1987. pp. 115–124. Fortschritte der Zoologie, Vol Stuttgart. [Google Scholar]
  72. Shaw AJ, Renzaglia KS. Phylogeny and diversification of bryophytes. American Journal of Botany. 2004;10:1557–1581. doi: 10.3732/ajb.91.10.1557. [DOI] [PubMed] [Google Scholar]
  73. Smirnoff N. The role of active oxygen in the response of plants to water deficit and desiccation. New Phytologist. 1993;125:27–58. doi: 10.1111/j.1469-8137.1993.tb03863.x. [DOI] [PubMed] [Google Scholar]
  74. Steer MW, Steer JM. Pollen tube tip growth. New Phytologist. 1989;111:323–358. doi: 10.1111/j.1469-8137.1989.tb00697.x. [DOI] [PubMed] [Google Scholar]
  75. Suppanz I, Sarnighausen E, Reski R. An integrated physiological and genetic approach to the dynamics of FtsZ targeting and organisation in a moss, Physcomitrella patens. Protoplasma. 2007;232:1–9. doi: 10.1007/s00709-007-0284-5. [DOI] [PubMed] [Google Scholar]
  76. Tanaka I. Cytoskeleton in fungal cells. Transaction of the Mycological Society of Japan. 1985;26:89–103. [Google Scholar]
  77. Traas JA, Braat P, Emons AMC, Meekes H, Derksen J. Microtubules in root hairs. Journal of Cell Science. 1985;76:303–320. doi: 10.1242/jcs.76.1.303. [DOI] [PubMed] [Google Scholar]
  78. Tewinkel M, Volkmann D. Observations on dividing plastids in the protonema of the moss Funaria hygrometrica Sibth. Planta. 1987;172:309–320. doi: 10.1007/BF00398659. [DOI] [PubMed] [Google Scholar]
  79. Wacker I, Quader H, Schnepf E. Influence of the herbicide oryzalin on cytoskeleton and growth of Funaria hygrometrica protonemata. Protoplasma. 1988;142:55–67. [Google Scholar]
  80. Walker LM, Sack FD. Amyloplasts as possible statoliths in gravitropric protonemata of the moss Ceratodon purpureus. Planta. 1990;181:71–77. [PubMed] [Google Scholar]
  81. Walker LM, Sack FD. Microfilament distribution in protonemata of the moss Ceratodon. Protoplasma. 1995;189:229–237. doi: 10.1007/BF01280177. [DOI] [PubMed] [Google Scholar]