SEARCH

SEARCH BY CITATION

Keywords:

  •  bryophytes;
  • cell walls;
  • immunocytochemistry;
  • land plant phylogeny;
  • ultrastructure;
  • vascular tissue

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • •  
    Although histologically much simpler than higher plants, bryophytes display a considerable degree of tissue differentiation, notably in those groups that possess an internal system of specialized water-conducting cells (WCCs). Here, using a battery of monoclonal antibodies, we examined the distribution of cell wall polysaccharide and glycoprotein carbohydrate epitopes in the gametophyte of four hepatics and eight mosses, with special reference to water-conducting cells.
  • •  
    CCRC-M7, an antibody against an arabinogalactan epitope, gave a highly consistent and generally specific labelling of WCCs; more variable results were obtained with other antibodies. The labelling patterns indicate that bryophytes exhibit cell and tissue complexity with respect to cell wall components on a par with higher plants.
  • •  
    A remarkable diversity in the immunocytochemical characteristics of WCCs was observed not only when comparing major bryophyte groups but also within the relatively small and well-circumscribed moss order Polytrichales, indicating that the cell wall biochemistry of WCCs may have been finely tuned in response to specific evolutionary pressures. The immunocytochemical data strengthen the notion that the WCCs in Takakia are not homologous with the hydroids of other mosses nor with the WCCs in Haplomitrium and metzgerialean liverworts.
  • •  
    The presence of several carbohydrate epitopes in hydroid walls runs strongly counter to the notion that their maturation involves hydrolysis of noncellulosic polysaccharides.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Recently, much progress has been made using polyclonal and monoclonal antibodies to plant cell wall components to determine their distribution in different tissues (Knox, 1997). Although major wall components – that is, cellulose, hemicellulose (often xyloglucans), pectins (homogalacturonan, rhamnogalacturonans and substituted galacturonans), glycoproteins and phenolics – are universally present in higher plants, immunocytochemical studies have revealed a very specific cell and tissue distribution of polysaccharide and glycoprotein carbohydrate epitopes. Surprisingly, to our knowledge none of the antibodies directed to cell wall epitopes in higher plants have been tested on bryophytes. Although these plants have a much simpler histological organization than higher plants, they may present a considerable degree of tissue differentiation, notably in those groups that possess an internal water-conducting system.

Specialized water-conducting cells (WCCs) occur in the majority of higher mosses (Polytrichales and Bryales) and, among the hepatics, in the Calobryales and few members of the Metgzeriales. Conducting tissues are unknown in the third group of plants traditionally placed in the bryophytes, that is, the hornworts or anthocerotes. As tracheids and vessel elements in tracheophytes, WCCs in bryophytes undergo programmed cytoplasmic lysis and lack a cytoplasmic content at maturity. A major difference, however, is that they do not deposit lignin in their walls; instead they contain polyphenolic compounds that may act as sealants (Scheirer, 1980). Contrasting with this major unifying character, WCCs in bryophytes exhibit remarkable morphological and developmental diversity (Ligrone et al., 2000).

WCCs in mosses, commonly referred to as hydroids, may occur in both the gametophyte and sporophyte generation. Hydroids are highly elongate cells with variously modified walls. In some members of the Polytrichales, including Polytrichum and Dendroligotrichum, the longitudinal hydroid walls are locally thickened before final cytoplasmic lysis and therefore may include several distinct wall layers. In other polytrichaceous mosses including Dawsonia and Atrichum, as well as in bryalean mosses, the hydroid walls are relatively thin throughout and consist of a homogeneous wall layer with a loose fibrillar or amorphous appearance and with low electron-opacity. On the basis of this highly distinctive appearance under the transmission electron microscope, it has been widely assumed that, allegedly as in tracheary elements, maturation of these walls involves the removal of noncellulosic polysaccharides (Hébant, 1977). Mature hydroids have no perforations in their walls as plasmodesmata are obliterated during late stages of differentiation (Ligrone et al., 2000). In a distinct position relative to bryoid mosses (viz Polytrichales and Bryales) stands Takakia: this enigmatic taxon, once considered a liverwort, is now generally recognised alongside the Sphagnidae and Andreaeidae as one of the basal moss lineages (Garbary et al., 1993; Renzaglia et al., 1997; Buck & Goffinet, 2000; Newton et al., 2000). As in bryoid mosses, a central strand of WCCs is present in both generations in Takakia whilst it is completely absent in Andreaea and Sphagnidae; however, in sharp contrast to hydroids, WCCs in Takakia show plasmodesma-derived perforations and are similar in shape to cortical parenchyma cells (Ligrone et al., 2000).

WCCs in hepatics are restricted to the gametophyte and possess numerous plasmodesma-derived pores. In the Calobryales they have thin walls and similar shape to ordinary parenchyma cells. By contrast, in the few metgzerialean genera possessing an internal water-conducting tissue, this consists of a single or multiple strands of highly elongated and thick-walled cells whose developmental design exhibits striking analogy with sieve cell development in tracheophytes (Ligrone & Duckett, 1996; Ligrone et al., 2000).

In this study we used a battery of monoclonal antibodies against polysaccharide and glycoprotein components of higher plants to map the distribution of cell wall components in a range of bryophytes including representatives of all groups known to possess an internal strand of water-conducting cells. Our principal aims were to discover whether or not tissue- and cell-specific wall epitopes occur in bryophytes and if so, whether differences in epitope distribution reflect the major structural differences between the WCCs of the various groups detailed above.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The species considered in the present study are reported in Table 1 along with their taxonomic position and sites of collection.

Table 1.  Liverwort and moss taxa investigated by immunocytochemistry
TaxaSite of collection
Hepatics
 Haplomitrium gibbsiae (Steph.) R. M. Schust. (Calobryales)New Zealand, Bridle Veil Path, Arthur's Pass
  Haplomitrium ovalifolium R. M. Schust. (Calobryales)New Zealand, Mount Arthur
  Symphyogyna undulata Colenso (Metzgeriales)New Zealand, Kelly Creek, Arthur's Pass
  Hymenophyton flabellatum (Labill.) Trevis (Metzgeriales)New Zealand, Kelly Creek, Arthur's Pass
Mosses
  Takakia lepidozioides Hatt. & H. InoueAleutian Islands (kindly provided by D.K. Smith and P. Davison, University of Tennesse)
  Takakia ceratophylla (Mitt.) GrolleBorneo, Mount Kinabalu, Laban Rata
  Dawsonia superba Grev. (Polytrichales)New Zealand, Kelly Creek, Arthur's Pass
  Dendroligotrichum dendroides (Hedw.) Broth. (Polytrichales)New Zealand, Kelly Creek, Arthur's Pass
  Polytrichum formosum Hedw.England
  Polytrichum strictum Brid. (Polytrichales)North Carolina
  Atrichum undulatum (Hedw.) P. Beauv. (Polytrichales)Tennessee
  Mnium hornum Hedw. (Bryales)Tennessee

Mature parts of the gametophyte were cut into 0.5 mm-long segments in a drop of 3% glutaraldehyde in 0.05 M piperazine-N,N′–bis(2-ethanesulfonic acid) (PIPES) buffer, pH 7.4 and transferred to a larger volume of the same fixative in scintillation vials for 2 h at room temperature. The samples were then rinsed in distilled water, dehydrated in a step gradient of ethanol with two exchanges of anhydrous ethanol at 4°C. LR White resin was added in 25% increments over 3 d, with two exchanges of 100% resin. The samples were then transferred to BEEM capsules with fresh resin and cured at 60°C for 24 h.

Light microscopy

A range of monoclonal antibodies from mouse or rat against plant cell wall components (Table 2) were probed on sections (0.5 µm) cut with a diamond histoknife and mounted on glass slides coated with chrome-alum to enhance section sticking. Rings were drawn around the sections with a wax pencil to maintain the fluid in subsequent steps of the protocol. The slides were then moved to an incubation chamber with high rh for the incubation steps. These were as follows: 1% (w/v) bovine serum albumin (BSA) in phosphate-buffered saline (PBS), 30 min; primary antibody diluted 1 : 8–1 : 80 depending upon the antibody, 4 h; three exchanges of PBS-BSA; 1 : 20 dilution of secondary antibody-gold (EY Laboratories, San Mateo, CA, USA) in PBS-BSA; three washes of PBS. The slides were then rapidly rinsed with double distilled water from a squirt bottle followed by six more repeats of double distilled water to remove residual chloride ions that would interfere with the silver intensification.

Table 2.  Monoclonal antibodies utilized for immunocytochemical characterization of bryophyte cell walls
AntibodyAntigen(s)/EpitopeReference/Source
AnticalloseCallose/(1–3)-β-linked penta- to hexa-glucanMeikle et al. (1991)/Biosupplies Ltd, Melbourne, Australia
AGPArabinogalactan-proteins/UnknownAnderson et al. (1984)/BioSupplies Ltd, Melbourne Australia
JIM8Arabinogalactan-proteins/UnknownPennell et al. (1991)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK
CCRC-M7Arabinogalactan-proteins, Rhamnogalacturonan-I/ Arabinosilated (1–6)-β-galactanPuhlman et al. (1994); Steffan et al. (1995)/M. Hahn, Complex Carbohydrate Research Center., University of Georgia, USA
LM5Galactan, Rhamnogalacturonan-I/ Tetra (1–4)-β-linked galactanJones et al. (1997)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK
LM6Arabinan, Rhamnogalacturonan-I/ Penta (1–5)-α-linked arabinanWillats et al. (1998)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK
JIM5Homogalacturonan/Low- or un-methyl-esterifiedWillats et al. (2000)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK
JIM7Homogalacturonan/Partially methyl-esterifiedWillats et al. (2000)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK
CCRC-M2Rhamnogalacturonan-I/UnknownPuhlman et al. (1994)/M. Hahn, Complex Carbohydrate Research Center., University of Georgia, USA
CCRC-M1Xyloglucan, Rhamnogalacturonan-I/ Fucosylated side groupPuhlman et al. (1994)/M. Hahn, Complex Carbohydrate Research Center, University of Georgia, USA
JIM12Extensin/UnknownSmallwood et al. (1994)/J.P. Knox, Centre for Plant Sciences, University of Leeds, UK

For silver intensification, the sections were incubated in freshly prepared solutions from the Amersham InstenSe (Amersham, Arlinngton, IL, USA) silver enhancement kit. An aliquot of 50–100 µl of the solution was pipetted on each slide and the slides returned to the humid chamber for 15–30 min. After the silver had developed, the sections were washed thoroughly with distilled water from a squirt bottle, dried with compressed air, and then mounted with Permount. Sections were photographed using a Zeiss Axioskop light microscope (Carlzeiss, Oberkochen, Germany). Serial sections to those used for immunocytochemistry were stained for 1 min with 1% toluidine blue in 1% sodium tetraborate for assessing the structural preservation and the general histological organization.

Immunogold-transmission electron microscopy

For electron microscopy only the antibodies which had been found to produce a distinct labeling of WCCs with the silver-intensification technique (Table 3) were probed. Sections from the same block faces used for light microscopy were cut with a diamond knife at 99 nm (pale gold reflectance) and mounted on 300 mesh uncoated gold grids. The grids were then processed through the localization protocol exactly as described for the slides in the immunogold silver protocols, except that the grids were floated on 4 µl drops of the solutions. After the water wash, the samples were dried and then stained for 2 min in 2%(w/v) uranyl acetate and 30 s in Reynold's lead citrate before observation with a Zeiss EM 10 CR (Zeiss) or a Jeol 100C electron microscope (Jeol).

Table 3.  Results of immunogold-silver labelling tests on transverse sections of the mature gametophyte in a range of bryophyte taxa
 CalloseAGPsJIM8CCRC-M7 LM5LM6JIM5JIM7CCRC-M2CCRC-M1JIM12
 WCCCPWCCCPWCCCPWCCCPWCCCPWCCCPWCCCPWCCCPWCCCPWCCCPWCCCP
  1. +   + + very strong labelling; ++ strong labelling; + weak labelling; ± very weak and scattered labelling; – no labelling; CP, cortical parenchyma; WCC, water conducting cells; (*) restricted to leptoids.

Haplomitriumovalifolium ± ± + +   +   + +   + + ± +   + +   +   + + ± +   + + + + + + +
Haplomitrium gibbsiae
Symphyogynaundulata +   +   + + ± ± + +   +   + +   + ± +   + +   + + +   + +   +
Hymenophyton flabellatum +   +   + + +   + +   +   + + +   + ± +   + +   + + +   +   + +   +
Takakia lepidozioides, Takakia ceratophylla + + +   +   + + +   +   + + +   + + +   +   + +   +   + ± + +   +   + +   +   + + +
Dawsonia superba + + +   +   + +   +   + + +   +   + + +
Dendroligotrichum dendroides + +   +   + (*) +   + ± +   + +   + + +   + +   + + +
Mnium hornum + +   + + +   +   + + +   + + + + +
Atrichum undulatum + +   + +   + +   +   + + +
Polytrichum formosum Polytrichum strictum + +   + +   + +   + + +   +   + + +   + + + + + +   +   +

For both light and electron microscopy, controls were made by omitting incubation with the primary antibody.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The antibody against callose labelled WCCs only in Hymenophyton and Symphyogyna (Fig. 1a). The cortical parenchyma cells were weakly labelled for callose in all taxa except Dendroligotrichum, that showed no visible labelling. Immunocytochemical localization of callose at the ultrastructural level showed the labelling to be mainly associated with pores in WCCs of Symphyogyna and Hymenophyton (Fig. 1b).

image

Figure 1. Silver (a, c, e–i) and gold (b, d) immunolabelling of conducting cells (WCCs) in bryophytes. (a) The central strand of WCCs in Symphyogyna as labelled with the anticallose antibody. (b) Detail of WCC in Hymenophyton showing callose labelling (arrows) associated with the pores. (c, d) Localization of arabinogalactan-proteins in Dendroligotrichum; (c) the leptoids (arrows) surrounding the central strand of hydroids show a strong reaction whilst the hydroids are labelled weakly; (d) the anti-arabinogalactan proteins (AGP) labelling in leptoids is localized along the plasmalemma/wall interface (arrows). (e–i) CCRC-M7 labelling of WCCs in Symphyogyna(e), Haplomitrium gibbsiae(f), Takakia lepidozioides(g), Dawsonia(h) and Mnium(i); WCCs in Haplomitrium and Takakia are enclosed in a circle; the central strand in Dawsonia consists of heavily labelled hydroids (arrows) associated with thick-walled stereids that are labelled much less intensely. Numbers on bars are micrometers.

Download figure to PowerPoint

Antibodies against varied carbohydrate epitopes of arabinogalactan proteins gave different results. The ‘arabinogalactan proteins (AGP)’ antibody labelled the WCCs in all mosses, except Dawsonia, whilst no labelling was observed in hepatics (Table 3). Cortical parenchyma cells in mosses were also reactive to this antibody, particularly those cells located in proximity to the water-conducting central strand. A particularly strong reaction was observed in the leptoids ensheathing the central strand in Polytrichum and Dendroligotrichum (Fig. 1c). Electron microscopy showed that the AGP antibody produced a conspicuous labelling of the plasmalemma/wall interface in these cells (Fig. 1d).

The JIM8 antibody gave virtually no labelling in any species (Table 3), whilst CCRC-M7 produced a strong labelling of WCCs in both liverworts (Fig. 1e,f) and mosses (Fig. 1g–i). Usually cortical parenchyma cells reacted to CCRC-M7 much more weakly than WCCs, no labelling at all being observed in Dawsonia (Fig. 1h). The thick-walled cells (stereids) associated with hydroids in the central strand in Dawsonia gave a very weak reaction, in sharp contrast to the heavy labelling of the hydroids (Fig. 1h).

When probed at the ultrastructural level, CCRC-M7 also gave a strong labelling of WCCs in both hepatics and mosses. The labelling was mostly restricted to the middle lamella in both WCCs (Fig. 2a) and parenchyma cells in Haplomitrium. The walls of WCCs were labelled throughout in Symphyogyna and Hymenophyton (Fig. 2b,c), whilst in the cortical parenchyma labelling was restricted to the middle lamella area at the cell corners in Hymenophyton (Fig. 2d) and no labelling at all was observed in Symphyogyna (Fig. 2e). Labelling of WCCs in Takakia was essentially restricted to the wall matrix (the term ‘wall matrix’ is used in this report to indicate the whole cell wall to the exclusion of the middle lamella) whilst the middle lamella area at the cell corners was not labelled (Fig. 3a). CCRC-M7 produced an intense and highly specific labelling of the hydroids in Mnium (Fig. 3b,c), Atrichum (Fig. 3d) and Dawsonia (Fig. 3e) both at the level of the middle lamella and the wall matrix, including the thin longitudinal walls which lack a morphologically discernible middle lamella. In Polytrichum and Dendroligotrichum, two moss taxa in which the hydroid walls are locally thickened, the labelling was localized in the middle lamella and the adjoining wall layer on either side, whilst the innermost wall layer (or layers when more than one was visible) did not react (Fig. 3f). At the cell corners the labelling was mostly restricted to a thin layer relatively transparent to the electrons, located between the middle lamella and the innermost wall layer (Fig. 4a,b), whilst very little labelling was visible on the middle lamella. Unlike other mosses, the thin walls of hydroids in Polytrichum and Dendroligotrichum showed little or no labelling with CCRC-M7 (Fig. 4a). In all moss species the cortical cells were labelled only at the level of the wall/plasmalemma interface (not shown). In Dendroligotrichum, however, the CCRC-M7 antibody produced a strong labelling of the whole walls in the parenchyma cells in direct contact with hydroids (Fig. 4c).

image

Figure 2. Electron microscopy of CCRC-M7 labelling in liverworts. (a) Detail of WCCs in Haplomitrium gibbsiae; the labelling is mostly located on the middle lamella whilst it is almost absent from the wall matrix (arrows). (b, c) WCCs in Symphyogyna (b) and Hymenophyton (c). The walls of WCCs are labelled throughout (arrows); (d) cortical parenchyma cells in Symphyogyna do not react; (e) cortical parenchyma cells in Hymenophyton are labelled at the level of the expanded middle lamella at the cell corners. ML, middle lamella; PC, parenchyma cell. Numbers on bars are micrometers.

Download figure to PowerPoint

image

Figure 3. Electron microscopy of CCRC-M7 labelling in mosses. (a) Takakia, the reaction is localized to the wall matrix (arrows); (b) and (c) Mnium, the hydroids are labelled at the level of both the wall matrix (arrows) and the middle lamella, whilst no reaction is visible in adjacent parenchyma cells; (d) Atrichum, detail of a thin wall between adjacent hydroids; (e) Dawsonia, detail of a hydroid, a neighbouring parenchyma cell and a stereid (S); the hydroid walls are labelled throughout whilst no reaction is visible in the parenchyma cell nor in the stereid. (f) Detail of a thickened hydroid wall in Polytrichum formosum, showing labelling of the middle lamella and two adjacent wall layers on either side (arrows). H, hydroid; ML, middle lamella; PC, parenchyma cell; S, stereid. Numbers on bars are micrometers.

Download figure to PowerPoint

image

Figure 4. CCRC-M7 labelling in Polytrichum formosum (a) and Dendroligotrichum (b) and (c). (a) The labelled wall layer visible in thickened walls (arrows) is interrupted at the level of the thin walls; note the absence of labelling on the expanded middle lamella area at the cell corner. (b) As in Polytrichum, labelling of hydroids in Dendroligotrichum is localized to the middle lamella of thickened walls and a thin wall layer (arrows) on either side; (c) the walls of parenchyma cells abutting the hydroids in Dendroligotrichum are heavily labelled (arrows). H, hydroid; ML, middle lamella; PC, parenchyma cell; TW, thin hydroid wall. Numbers on bars are micrometers.

Download figure to PowerPoint

Three further antibodies against epitopes occurring in the side chains of rhamnogalacturonan-I (RG-I), LM5, LM6 and CCRC-M2, also produced sharply different labelling patterns.

LM5 gave a strong and highly specific labelling of WCCs in Dawsonia (Fig. 5a) and Polytrichum (Fig. 5b) and a weaker and nonspecific labelling in Takakia (Fig. 5c) and Mnium (Fig. 5f). Weak or no labelling was observed in WCCs in Dendroligotrichum (Fig. 5e) and Atrichum as well as in the hepatics Haplomitrium and Symphyogyna. LM6 produced a strong and specific labelling of WCCs in Dawsonia (Fig. 7a), whilst only a very weak labelling was observed in the other polytrichaceous mosses and in Mnium. A strong but nonspecific labelling was observed with this antibody in Takakia (Fig. 7d) and a very weak or no labelling was recorded in Haplomitrium and Symphyogyna. CCRC-M2 produced weak labelling of both WCCs and parenchyma cells in Haplomitrium and only of WCCs in Polytrichum, whilst no reaction was observed in any other taxon with this antibody.

image

Figure 5. LM5 silver (a–c, e, f) and gold (d, g) labelling of water-conducting cells (WCCs) in mosses. (a) Dawsonia, the hydroids are heavily labelled both in the central strand and in the leaf traces, while the stereids intermingled with hydroids in the central strand show no labelling; (b) Polytrichum formosum, the thickened walls of hydroids are heavily labelled, whilst thin wall areas show no reaction (arrows); (c, d) Takakia ceratophylla; (c) the WCCs (encircled) are more intensely labelled than adjoining cortical cells; (d) the labelling in WCCs is localized to the wall matrix (arrows); (e) Dendroligotrichum, the hydroids (encircled) show very little labelling whilst the cortical cells are heavily labelled; (f) and (g) Mnium; (f) the hydroids (encircled) are weakly labelled, mostly at cell corners; (g) electron microscopy shows that the labelling is restricted to a thin layer between the middle lamella and the wall matrix at the cell corners (arrows). L, leaf trace; S, stereids. Numbers on bars are micrometers.

Download figure to PowerPoint

image

Figure 7. LM6 labelling of water-conducting cells (WCCs) in mosses. (a–c) Highly specific labelling of hydroids in Dawsonia shown by light (a) and electron microscopy (b, c); hydroids in leaf traces (arrows) also are labelled, whilst the stereids scattered among hydroids in the central trand show no labelling. (d) and (e) Takakia; (d) the antibody reacts with both WCCs (encircled) and the living cells of the cortex including the peripheral stereids; (e) electron microscopy showing the labelling localized to the middle lamella of WCCs (arrows); (f) Polytrichum formosum(g) Dendroligotrichum; in both species the labelling of hydroids by LM6 is restricted to the middle lamella and two wall layers (arrows) on either side in thickened walls. L, leaf trace; ML, middle lamella; S, stereids. Numbers on bars are micrometers.

Download figure to PowerPoint

At the ultrastructural level, LM5 strongly reacted with the wall matrix in Dawsonia hydroids whilst no labelling was found on the expanded middle lamella areas at the corners between hydroids (Fig. 6a) and only weak labelling on parenchyma cell walls (Fig. 6b). A similar pattern was observed in Takakia (Fig. 5d), whilst in Mnium LM5 labelled the border of the middle lamella areas at the corners between hydroids but not the wall matrix (Fig. 5g). A completely different pattern was found in Polytrichum; here LM5 strongly labelled a wall layer interposed between the middle lamella and a more internal layer abutting the hydroid lumen (Fig. 6c). The labelled wall layer varies in thickness and may in places be absent (Fig. 6c,d). It is always missing in the thin wall areas. Although possessing a distinct layered structure, the hydroid walls in Dendroligotrichum did not react with LM5 (Fig. 6e). In Polytrichum LM5 gave weak labelling of cortical cells, including leptoids and peripheral stereids, and this was limited to the plasmalemma/wall interface (not shown). The immunocytochemical pattern of LM6 in Dawsonia (Fig. 7b,c) was much the same as observed with LM5, whilst in Takakia LM6 mainly labelled the middle lamella areas, notably at the cell corners, both in WCCs (Fig. 7e) and in parenchyma cells. In Polytrichum (Fig. 7f) and Dendroligotrichum (Fig. 7g) LM6 gave weak labelling of the middle lamella and an adjacent wall layer.

image

Figure 6. LM5 labelling of hydroids in mosses. (a, b) Dawsonia; (a) the hydroid walls are labelled throughout (arrows) except at the cell corners; (b) the strong reaction of hydroid walls (arrows) contrasts with the very weak labelling of adjacent parenchyma cells. (c, d) Polytrichum formosum; only the thick walls are labelled, with the labelling being restricted to a layer of varying thickness on one or both sides (arrows) of the middle lamella; (e) hydroids in Dendroligotrichum show no reaction to LM5. ML, middle lamella; PC, parenchyma cell. Numbers on bars are micrometers.

Download figure to PowerPoint

JIM5 and JIM7, two antibodies that bind to two different partially methyl-esterified epitopes of homogalacturonan, gave sharply distinct results. The former gave a strong labelling of WCCs in the moss Dendroligotrichum and the hepatic Haplomitrium whilst the latter gave little or no labelling of WCCs but produced a stronger reaction with cortical cells in all the hepatics. The ultrastructural localization of the JIM5 epitope in Dendroligotrichum and Haplomitrium was quite different, the epitope being localized in both the middle lamella and the innermost wall layer (but not the intermediate layer, that reacts with LM6) in Dendroligotrichum (Fig. 8a) and throughout the wall except the middle lamella in Haplomitrium (Fig. 8b). The thin wall areas of hydroids in Dendroligotrichum showed no labelling.

image

Figure 8. (a) Dendroligotrichum, thickened hydroid walls as labelled with JIM5; the antibody reacts with the middle lamella and a thick layer abutting the cell lumen whilst a thinner layer in between (arrows) is not labelled; (b) Haplomitrium ovalifolium, JIM5 labelling of water-conducting cells (WCCs); the wall matrix (arrows) is heavily labelled whilst the middle lamella shows no reaction; (c–e) CCRC-M1 labelling in Takakia ceratophylla; (a) the central strand of WCCs (encircled) is strongly labelled; (d) and (e) details of WCCs showing the labelling localized to the middle lamella (arrows) including enlarged areas at cell corners; very little labelling is visible on the wall matrix. Numbers on bars are micrometers.

Download figure to PowerPoint

CCRC-M1, an antibody against fucosylated side groups of xyloglucan and RG-I, reacted strongly with WCCs and less so with parenchyma cells in Takakia (Fig. 8c). No reaction to this antibody was observed in any other moss besides Takakia. The WCCs in the hepatics Symphyogyna and Hymenophyton were also negative to CCRC-M1 whilst only a weak reaction was observed in Haplomitrium (Table 2). By contrast, parenchyma cells showed a strong reaction in both Symphyogyna and Hymenophyton. The CCRC-M1 epitope in WCCs in Takakia was localized to the middle lamella, notably at the cell corners, whilst the wall matrix showed little or no labelling (Fig. 8d,e). The same distribution was observed in parenchyma cells.

The antibody against extensin, JIM12, reacted weakly with WCCs in Haplomitrium. No reaction was observed with mature WCCs in any other taxon. Parenchyma cells showed a strong reaction with JIM12 in Polytrichum (not shown) and much a weaker reaction in Takakia and Dendroligotrichum.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Patterns of cell wall labelling in bryophytes

As far as we are aware, this is the first study in which antibodies raised against cell wall components of higher plants have been used for immunolabelling of cell walls in bryophytes. The fact that a greater or lesser labelling of certain cell types was observed with all the antibodies clearly indicates that molecules containing those epitopes are included in the biochemical arsenal used in cell wall construction in bryophytes and, as in higher plants, they are deployed in different ways in different species. Particularly noteworthy are the discoveries that WCCs in bryophytes have highly distinctive immunocytochemical properties relative to adjacent cortical cells, and WCCs in different taxa show different labelling patterns.

With some exceptions, notably the highly specific CCRC-M7 reaction with WCCs in Symphyogyna, Dawsonia and Atrichum, differences in the immunogold-silver labelling between WCCs and cortical cells appeared to be essentially quantitative with most antibodies. Electron microscopy, however, showed that in most cases the two cell types are clearly distinct in the location of the antibodies within the walls, implying a tissue-specific spatial regulation of the epitope-containing molecules.

Distribution of arabinogalactan epitopes

Arabinogalactans in higher plant cell walls occur mainly as side chains in arabinogalactan proteins (AGPs) but can also be associated with the pectic polysaccharide RG-I (Gaspar et al., 2001; Willats et al., 2001). The AGPs are ubiquitous proteoglycans associated with plant cell walls and especially the plasmalemma/wall interface (Gaspar et al., 2001). Their many proposed functions include protoplast/wall adhesion, cell differentiation, targeting sites for new wall synthesis, stabilization of pectic polysaccharides in secretory vesicles and the signalling for programmed-cell-death. In many cases strong labelling of developing vascular tissue has been observed in higher plants with antibodies to AGPs, leading to the suggestion that these molecules are involved in cell-specific changes associated with differentiation (Knox et al., 1991; Schindler et al., 1995). Many monoclonal antibodies have been generated to this group of proteins, although for none of them is epitope specificity known in detail (Knox, 1997). Most of the AGP monoclonal antibodies recognize all types of AGP but JIM8 appears to be specific for a subset that are more lipophilic (Gaspar et al., 2001). In this study, the three AGP antibodies also gave strikingly different results.

The most consistent and specific labelling of WCCs in both liverworts and mosses was obtained with CCRC-M7 antibody. This antibody therefore appears to be suitable as a marker of WCC differentiation in bryophytes. CCRC-M7 recognizes an epitope containing at least three (1–6)-β-linked galactosyl residues and possibly also including arabinosyl residues (Steffan et al., 1995). Although CCRC-M7 marked the WCCs in most taxa with remarkable selectivity, the subcellular localization of the labelling was highly variable. Thus CCRC-M7 labelling was found throughout the wall of WCCs in the metzgerialean hepatics Symphyogyna and Hymenophyton as well as in the mosses Mnium and Dawsonia, whilst it was restricted to the inner hydroid walls in Takakia, the middle lamella and an adjacent wall layer in Polytrichum and Dendroligotrichum, and only the middle lamella in Haplomitrium. Most likely the pattern of WCC labelling by CCRC-M7 mainly reflects the distribution of the epitope associated with RG-I (this also takes account of the fact that WCCs have no plasmalemma at maturity), whilst the labelling of the plasmalemma/wall interface in cortical cells probably reflects the presence of AGPs. CCRC-M7 labelling of the wall in cortical cells was only observed in Hymenophyton (but not in the allied species Symphyogyna), with a strong labelling of the middle lamella. In Dendroligotrichum, similar labelling was limited to the parenchyma cells in contact with hydroids. A dual localization of CCRC-M7 has been found in Arabidopsis root parenchyma cells (Freshour et al., 1996), although the labelling pattern of living cells in other species indicates a specificity for plasmalemma-associated AGPs (Vaughn, 2002a,b).

The AGP mouse monoclonal antibody was elicited against stylar extracts of tobacco. Unlike the CCRC-M7 monoclonal, in higher plants the AGP antibody appears to recognize only the plasmalemma/wall interface and thus probably reacts only with AGPs and not with arabinogalactans in the middle lamella. Bryophytes present a similar picture (Fig. 1d). Remarkably, in all the mosses except Dawsonia, but including Takakia, the stronger labelling with this antibody was found in parenchyma cells closely associated with WCCs, which, on the basis of their leptoid-like cytology, are considered to be specialized for the transport of assimilates (Ligrone et al., 2000).

The JIM8 monoclonal antibody recognizes a small subset of AGPs and in general gave a very low level of labelling in bryophytes (Table 3). In contrast this antibody often strongly recognizes vascular cells in higher plants (K.C., Vaughn, unpublished).

Distribution of pectic polysaccharide epitopes

Several different antipectin monoclonal antibodies were used. The monoclonal antibodies JIM5 and JIM7 are specific to homogalacturonan and both bind preferentially to partially methyl-esterified epitopes although these are not yet defined (Willats et al., 2001). JIM7 mainly labelled the wall matrix whereas JIM 5 bound predominantly to the middle lamella. In addition JIM5 and JIM7 often displayed two mutually exclusive labelling patterns, that is, if a given cell type was labelled with one antibody, then it was not labelled with the other. For example, the WCC in Takakia were unlabelled with JIM 5 but strongly labelled with JIM7.

The CCRC-M2 monoclonal antibody labels RG-I but the epitope recognized is not known. Only Haplomitrium reacted with CCRC-M2, suggesting that the corresponding epitope is rare in bryophyte cell walls.

The LM5 and LM6 monoclonal antibodies recognize galactan (LM5) and arabinan (LM6) side chains of RG-I pectins with high specificity. In higher plants these epitopes are highly developmentally and spatially regulated (Willats et al., 2001). The LM5 epitope generally appears during cell differentiation whilst the LM6 epitope is often abundant in the walls of meristematic cells (Willats et al., 1999; Bush et al., 2001; Serpe et al., 2001; Vaughn, 2002b). In bryophytes both WCCs and cortical cells may react to both antibodies, thus indicating that galactan and arabinan RG-I side chains may coexist in fully differentiated cells (Table 3). The WCCs in liverworts reacted only to LM6 whilst in mosses the results were more variable; some species reacted only to LM6 (i.e. Dendroligotrichum and Atrichum) others to both antibodies. In the latter group, however, the distribution of the two epitopes was highly variable. The LM5 epitope was localized to the wall matrix in Takakia, the whole wall except the middle lamella at cell corners in Dawsonia, the border of the expanded middle lamella at cell corners in Mnium and only a specific layer in the thickened hydroid walls in Polytrichum. The fact that the LM5-labelled layer in Polytrichum hydroids is absent in the thin wall areas suggests that this layer is not homologous with the wall matrix in Dawsonia.

In the two species where LM6 gave the stronger labelling, that is, Takakia and Dawsonia, the distribution of this epitope was sharply different, being restricted to the middle lamella and the wall matrix, respectively, that is, the opposite of the patterns found in these genera with LM5. The distribution of LM6 in Polytrichum (the middle lamella and two adjacent wall layers), coinciding with that observed for CCRC-M7, indicates that in this species at least arabinan and arabinogalactan epitopes are colocalized to the same wall domains, probably as side chains of RG-I pectins.

In line with the results obtained with other antibodies, the strong reactivity of WCCs in Takakia to CCRC-M1 clearly sets these cells apart from the hydroids in other mosses, all of which were completely negative with this antibody. CCRC-M1 recognizes an epitope containing a terminal fucosyl residue α-1-2-linked to a galactosyl residue. This epitope is most prevalent in xyloglucans from many (but not all) plants, but is also found to a low level in RG-I (Puhlman et al., 1994). The fact that the reaction in Takakia was mainly localized to the middle lamella at the cell corners strongly suggests that it reflects the distribution of a fucosylated RG-I rather than xyloglucan, although a nonfucosylated xyloglucan may still be present. Interestingly, the results obtained indicate that the fucosylated epitope of CCRC-M1, lacking in all mosses except Takakia, also occurs in the four liverworts tested although here it is mainly associated with parenchyma cell walls.

Callose

The callose associated with pits in WCCs in Symphyogyna and Hymenophyton is probably a remnant of the callose plugs which are deposited around the plasmodesmata and subsequently removed during pit development (Ligrone & Duckett, 1996).

Concluding remarks

The remarkable diversity observed in the labelling pattern of WCCs in different major groups is wholly consistent with the hypothesis of independent evolution of these cells in the different bryophyte lineages (Ligrone et al., 2000). Notably, the data obtained in this work reinforce the notion that WCCs in Takakia are not homologous with the hydroids in other mosses nor with the WCCs in Haplomitrium though with the latter they share a similar morphology and plasmodesma-derived perforations. However, it is important to note that the same data also point to the existence of a pronounced immunocytochemical diversity within the polytrichaceous mosses, a relatively small and well-circumscribed group that has recently been confirmed as a monophyletic entity (Hyvonen et al., 1998). At least in part this may be related to the fact that in certain polytrichaceous genera, including Polytrichum and Dendroligotrichum, the hydroid walls are locally thickened by deposition of additional layer(s) presumably providing additional strengthening for their relatively large shoots. By contrast in other genera, for example, Dawsonia, the hydroids have thin walls throughout and adequate mechanical strength is presumably ensured by the presence of thick-walled living cells (stereids) associated with hydroids in the shoot central strand (Hébant, 1977). It is interesting to note that the thickened hydroid walls in Dendroligotrichum differ from those in Polytrichum in their apparent lack of galactan-substituted RG-I. This finding suggests that hydroids with thickened walls may have evolved independently in the two genera, starting from a thin-walled ancestral condition. This appears to be in line with the dendrogram recently proposed by Hyvonen et al. (1998) in which Dawsonia is resolved in a basal position and Dendroligotrichum and Polytrichum are widely separated. WCCs in the two allied liverwort genera Symphyogyna and Hymenophyton exhibit close immunological congruence, although, coherently with current classification of the two genera in separate families (Crandall-Stotler & Stotler, 2000), the different labelling pattern of cortical cells with CCRC-M7 is probably an indication of evolutionary divergence. By contrast the three pairs of species within the same genus (Polytrichium, Takakia, Haplomitrium) gave exactly the same labelling patterns.

The present results invite comparative immunocytochemical investigations on other mosses. These should include a range of advanced bryalean taxa where the water-conducting strands are much reduced or completely absent (e.g. Orthotrichales, Hookeriales). Could it be that the latter groups retain an immunological signature of the former presence of hydroids? On the other hand, absence of similar markers in Sphagnum and Andreaea would reinforce the idea that the absence of WCCs in both these groups is the primitive condition in mosses.

The wide range of polysaccharide and glycoprotein epitopes detected in the present study clearly indicates that hydroid walls have a highly complex biochemical organization that belies their apparently simple structure. The presence of pectin-associated epitopes in particular is probably the strongest argument to date against the hypothesis, long held as perhaps the most compelling feature linking tracheary elements and hydroids (Hébant, 1977), that the removal of noncellulosic polysaccharides accompanies hydroid maturation.

Information on immunocytochemical properties of tracheids/vessel elements in tracheophytes is currently restricted to very few antibodies (Schindler et al., 1995; McCann, 1997; Awano et al., 1998; Maeda et al., 2000), none of which could be obtained for use in the present study, apart from JIM8 that gave virtually no reaction in bryophytes. A greatly extended comparative immunocytochemical study of WCCs and tracheids in different bryophyte and tracheophyte groups now becomes particularly desirable not only for assessment of homology/homoplasy in the context of embryophyte interrelationships and phylogeny (Kenrick & Crane, 1997; Mishler & Churchill, 1984, 1985; Mishler et al., 1994; Ligrone et al., 2000; Renzaglia et al., 2000), but also for a better understanding of functional relationships between cell wall molecular organization and specialization in long-distance transport of water.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Dr M. Hahn (Complex Carbohydrate Research Center, University of Georgia, USA) for generously supplying some of the antibodies used for this study. Thanks are extended to Lynn Libous-Bailey (Southern Weed Science Research Unit, Agricultural Research Service, U.S. Department of Agriculture, Stoneville, MS) for technical assistance and the Department of Conservation in New Zealand for providing collecting permits in National Parks for the New Zealand taxa. This work was supported by a grant to R. Ligrone by CNR (Italy) for a short-term stay in the USA (2000). The observations were in part performed at the eISME (University “Federico II”, Napoli, Italy), whose technical staff is gratefully acknowledged. Mention of a trademark, proprietary property or vendor does not constitute an endorsement by the USDA.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Anderson MA, Sandrin MS, Clarke AE. 1984. A high proportion of hybridomas raised to a plant extract secrete antibody to arabinose or galactose. Plant Physiology 75: 10131016.
  • Awano T, Takabe K, Fujita M. 1998. Localization of glucuronoxylans in Japanese beech visualized by immunogold labelling. Protoplasma 202: 213222.
  • Buck WR, Goffinet B. 2000. Morphology and classification of mosses. In: ShawAJ, GoffinetB, eds. Bryophyte biology. Cambridge, UK: Cambridge University Press, 71123.
  • Bush MS, Marry M, Huxham IM, Jarvis MC, McCann MC. 2001. Developmental regulation of pectic epitopes during potato tuberisation. Planta 213: 869880.
  • Crandall-Stotler B, Stotler RE. 2000. Morphology and classification of the Marchantiophyta. In: ShawAJ, GoffinetB, eds. Bryophyte biology. Cambridge, UK: Cambridge University Press, 2170.
  • Freshour G, Clay RP, Fuller MS, Albersheim P, Darvill AG, Hahn MG. 1996. Developmental and tissue-specific structural alterations of the cell-wall polysaccharides of Arabidopsis thaliana roots. Plant Physiology 110: 14131429.
  • Garbary DJ, Renzaglia KS, Duckett JG. 1993. The phylogeny of land plants: a cladistic analysis based on male gametogenesis. Plant Systematics and Evolution 188: 237269.
  • Gaspar Y, Johnson KL, McKenna JA, Bacic A, Schultz CJ. 2001. The complex structure of arabinogalactan-proteins and the journey towards understanding their function. Plant Molecular Biology 47: 161176.
  • Hébant C. 1977. The conducting tissues of bryophytes. Bryophylum Bibliotheca, Vol. 10. Vaduz, Lichtenstein: J. Cramer.
  • Hyvonen J, Hedderson TA, Smith Merrill GL, Gibbings JG, Koskinen S. 1998. On phylogeny of the Polytrichales. Bryologist. 101: 489504.
  • Jones L, Seymour GB, Knox JP. 1997. Localization of pectic galactan in tomato cell walls using a monoclonal antibody specific to (1(r) 4)-β-D-Galactan. Plant Physiology 113: 14051412.
  • Kenrick P, Crane PR. 1997. The origin and early diversification of land plants: a cladistic study. Smithsonian Series in Comparative Evolutionary Biology. Washington DC, USA: Smithsonian Institution Press.
  • Knox JP. 1997. The use of antibodies to study the architecture and developmental regulation of plant cell walls. International Review of Cytology 171: 79110.
  • Knox JP, Linstead PJ, Peart J, Cooper C, Roberts K. 1991. Developmentally regulated epitopes of cell surface arabinogalactan proteins and their relation to root tissue pattern formation. Plant Journal 1: 317326.
  • Ligrone R, Duckett JG. 1996. Development of water-conducting cells in the antipodal liverwort Symphyogyna brasiliensis (Metzgeriales). New Phytologist 132: 603615.
  • Ligrone R, Duckett JG, Renzaglia KS. 2000. Conducting tissues and phyletic relationships of bryophytes. Philosophical Transactions of the Royal Society London B 355: 795813.
  • Maeda Y, Awano T, Takabe K, Fujita M. 2000. Immunolocalization of glucomannans in the cell wall of differentiating tracheids in Chamaecyparis obtusa. Protoplasma 213: 148156.
  • McCann MC. 1997. Tracheary element formation: building up to a dead end. Trends in Plant Science 2: 333338.
  • Meikle PJ, Bonig I, Hoogenraad NJ, Clarke AE, Stone BA. 1991. The location of (1[RIGHTWARDS ARROW]3)-β-glucans in the walls of pollen tubes of Nicotiana alata using a (1[RIGHTWARDS ARROW]3)-β-glucan-specific monoclonal antibody. Planta 185: 18.
  • Mishler BD, Lewis LA, Buchheim MA, Renzaglia KS, Garbary DJ, Delwich CF, Zechman FW, Kantz TS, Chapman RL. 1994. Phylogenetic relationships of the ‘green algae’ and ‘bryophytes’. Annals of the Missouri Botanical Gardens 8: 451483.
  • Mishler BD, Churchill SP. 1985. Transition to a land flora: phylogenetic relationships of the green algae and bryophytes. Cladistics 1: 305328.
  • Mishler BD, Churchill SP. 1984. A cladistic approach to the phylogeny of the ‘bryophytes’. Brittonia 36: 406424.
  • Newton AE, Cox CJ, Duckett JG, Wheeler JA, Goffinet B, Hedderson TAJ, Mishler BD. 2000. Evolution of major moss lineages: phylogenetic analysis based on multiple gene sequences and morphology. Bryologist 103: 187211.
  • Pennell RI, Janniche L, Kjellbom P, Scofield GN, Peart JM, Roberts K. 1991. Developmental regulation of a plasma membrane arabinogalactan-protein in oilseed rape flowers. Plant Cell 3: 13171326.
  • Puhlmann J, Bucheli E, Swain MJ, Dunning N, Albersheim P, Darvill AG, Hahn MG. 1994. Generation of monoclonal antibodies against plant cell wall polysaccharides. I. Characterization of a monoclonal antibody to a terminal α-(1[RIGHTWARDS ARROW]2)-linked fucosyl-containing epitope. Plant Physiology 104: 699710.
  • Renzaglia KS, Duff RJ, Nickrent DL, Garbary DJ. 2000. Vegetative and reproductive innovations of early land plants: implications for a unified phylogeny. Philosophical Transactions of the Royal Society London B 355: 769793.
  • Renzaglia KS, McFarland KD, Smith DK. 1997. Anatomy and ultrastructure of the sporophyte of Takakia ceratophylla (Bryophyta). American Journal of Botany 84: 13371350.
  • Scheirer DC. 1980. Differentiation of bryophyte conducting tissues: structure and histochemistry. Bulletin of the Torrey Botanical Club 107: 298307.
  • Schindler T, Bergfeld R, Schopfer P. 1995. Arabinogalactan-proteins in maize coleoptiles: developmental relationship to cell death during xylem differentiation but not to extension growth. Plant Journal 7: 2536.
  • Serpe MD, Muir AJ, Keidel AM. 2001. Localization of cell wall polysaccharides in non-articulated laticifers of Asclepias speciosa Torr. Protoplasma 216: 215226.
  • Smallwood M, Beven A, Donovan N, Neill SJ, Peart J, Roberts K, Knox JP. 1994. Localization of cell wall proteins in relation to the developmental anatomy of the carrot root apex. Plant Journal 5: 237246.
  • Steffan W, Kovàc P, Albersheim P, Darvill AG, Hahn MG. 1995. Characterization of a monoclonal antibody that recognizes an arabinosylated (1[RIGHTWARDS ARROW]6)-b-D-galactan epitope in plant complex carbohydrates. Carbohydrate Research 275: 295307.
  • Vaughn KC. 2002a. Attachment of the parasitic weed dodder to the host. Protoplasma 219: 227237.
  • Vaughn KC. 2002b. Dodder hyphae invade the host: a structural and immunocytochemical characterization. Protoplasma (In press.)
  • Willats WGT, Limberg G, Buchholt HC, Van Alebeek G, Benen J, Christensen TMIE, Visser J, Voragen A, Mikkelsen JD, Knox JP. 2000. Analysis of pectic epitopes recognised by hybridoma and phage display monoclonal antibodies using defined oligosaccharides, polysaccharides, and enzymatic degradation. Carbohydrate Research 327: 309320.
  • Willats WGT, Marcus S, Knox JP. 1998. Generation of a monoclonal antibody specific to (1[RIGHTWARDS ARROW]5)-α-L-arabinan. Carbohydrate Research 308: 149152.
  • Willats WGT, McCartney L, Mackie W, Knox JP. 2001. Pectin: cell biology and prospects for functional analysis. Plant Molecular Biology 47: 927.
  • Willats WGT, Steele-King CG, Marcus SE, Knox JP. 1999. Side chains of pectic polysaccharides are regulated in relation to cell proliferation and cell differentiation. Plant Journal 20: 617628.