| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Anatomy and Morphology |
Department of Plant Biology, Southern Illinois University, Carbondale, Illinois 62901-6509 USA
Received for publication September 19, 2005. Accepted for publication February 20, 2006.
ABSTRACT
The presence of Nostoc in longitudinally oriented schizogenous canals is a feature that separates Leiosporoceros from all other hornworts and represents a novel symbiotic arrangement in land plants. In surface view, Nostoc canals are visible as elongated, dichotomously branched blue-green strands. All other hornworts develop numerous discrete globose colonies through continuous production of mucilage clefts as avenues for multiple invasions within a single thallus. To elucidate the anatomy and development of the unusual Nostoc strands in Leiosporoceros, we examined sporeling development in culture and the structure of strands in field-collected plants using light and electron microscopy. Rosette-like sporelings have mucilage clefts scattered along swollen apices. All field specimens were strap-shaped, contained Nostoc, and lacked mucilage clefts. Nostoc strands are located in the center of the thallus and develop behind the apical cell by separation of the middle lamella between apical derivatives. Strands elongate and branch in synchrony with apical growth, and thus only a single invasion is required for strand production. Two distinct ultrastructural morphotypes in the collections suggest nonspecificity of Nostoc. We speculate that Nostoc enters the thallus in the sporeling stage through mucilage clefts, and once colonies are established, cleft production ceases.
Key Words: Anthocerotophyta • cyanobacteria • hornworts • Leiosporoceros • neotropics • nitrogen fixation • Nostoc strands • symbiosis
Cyanobacteria played a crucial role in the oxygenation of Earth through a novel photosynthetic mechanism that has been releasing oxygen into the atmosphere since Precambrian times (Adams and Duggan, 1999
; Schopf, 2000
; Raven, 2002a
). The production of atmospheric oxygen made possible the development of aerobic metabolism and the evolution of a multitude of aquatic and terrestrial organisms, including all eukaryotes (Raven et al., 1999
). Diverse life strategies and physiological adaptations allowed cyanobacteria to colonize literally every habitat on Earth (Whitton and Potts, 2000
). In addition, the ability of some filamentous cyanobacteria to fix atmospheric nitrogen has enabled them to enter symbioses with diverse marine and terrestrial organisms (Adams, 2000
). Among the filamentous cyanobacteria, the heterocyst-forming genera Nostoc Vauch. and Anabaena Bory ex Bornet et Flahault are the most commonly reported symbionts (Naidu, 1977
; Potts, 2000
; Lechno-Yossef and Nierzwicki-Bauer, 2002
).
In land-inhabiting green plants, symbiotic associations with cyanobacteria are thought to have evolved ca. 500 million years ago (mya) (Raven, 2002b
), and today they are found sporadically in all major plant lineages. Due to the ecological importance of the plant–cyanobacteria relationship, the mode of establishment of the symbiosis and the physiological and cytological mechanisms involved have been intensively studied (Meeks, 1998
; Rai et al., 2002
).
Among tracheophytes, the eudicot Gunnera L. is the only known angiosperm that harbors a cyanobacterial symbiont (Bergman et al., 1992
; Bergman, 2002
). In gymnosperms, cyanobacterial symbionts are limited to but widespread in cycads, where they occur in modified "coralloid" roots (Lindblad et al., 1985
). Nostoc colonies form a dark band between the inner and outer cortex of the root, with elongated cortical cells presumably related to metabolite transport interconnecting the two partners (Lindblad et al., 1985
; Adams, 2000
). The aquatic heterosporous fern Azolla Lam. is the only pteridophyte with a cyanobacterial association (Duckett et al., 1975
; Adams, 2000
). The cyanobiont in this plant is located in an extracellular cavity that forms by the infolding of a dorsal, chlorophyllous lobe above the water (Duckett et al., 1975
; Adams, 2000
).
Among bryophytes, Blasia L. and Cavicularia Steph. (Blasiales) are the only liverworts that have permanent symbiotic associations with cyanobacteria (Renzaglia, 1982
; Dalton and Chatfield, 1985
; Renzaglia et al., 2000
). The Nostoc endophyte is restricted to auricles that develop from ventral, three-celled slime hairs on the liverwort. Mucilage produced by internal and external slime papillae attracts the Nostoc, which becomes "trapped" in the auricle as the auricle grows and closes off to form a globose colony (Renzaglia, 1982
; Renzaglia et al., 2000
). Liverwort cells associated with the cyanobacteria have wall ingrowths that presumably increase the area of absorption (Duckett et al., 1977
).
In mosses, cyanobacterial filaments are usually epiphytic on moss gametophytes (Solheim and Zielke, 2002
). Endophytic cyanobionts in hyaline cells of the peat moss, Sphagnum L., have also been reported and probably contribute to its nitrogen balance and growth (Granhall and Hofsten, 1976
; Solheim and Zielke, 2002
).
Discrete globose cyanobacterial colonies, usually on the ventral side of the gametophyte, are a universal feature of hornworts (Leitgeb, 1878
; Campbell, 1895
; Renzaglia, 1978
). The typical pattern in hornworts is for the cyanobacteria to invade the thallus through ventral mucilage clefts that form continuously near the apex throughout the lifespan of the hornwort. Because each cleft is a potential site for Nostoc invasion, multiple discrete colonies typically occur on one thallus (Renzaglia, 1978
). When the cyanobacteria become established, the middle lamella between internal cells separates to form a schizogenous space, and a globose colony is formed (Duckett et al., 1977
; Rodgers and Stewart, 1977
; Adams, 2002
).
The only known exception to the previously mentioned organization and development of Nostoc colonies in hornworts is found in the monotypic genus Leiosporoceros Hässel (Duff et al., 2004
; Villarreal et al., 2005
). In recent morphological and phylogenetic studies (Duff et al., 2004
; Villarreal et al., 2005
), this taxon was identified as the most genetically and morphologically divergent hornwort. Among the several morphological innovations that distinguish this plant are the absence of ventral clefts in mature plants and the presence of cyanobacteria in strands that parallel the main axis of the thallus (Villarreal et al., 2005
). This configuration of Nostoc in continuous strands not only differs from that in other hornworts, but is unparalleled in land plants. To fully elucidate the development and organization of this novel symbiotic arrangement, we studied the ultrastructure and anatomy of cultured sporelings and mature field-collected plants and observed in detail both live populations from el Valle de Antón, Panamá and dry material from Ecuador and Mexico. We describe the anatomy, ultrastructure and development of cyanobacterial strands in adult gametophytes and the ultrastructure of the Nostoc filaments. Based on sporeling growth in culture, we suggest a probable mode for the establishment of the cyanobacterium in this hornwort.
MATERIALS AND METHODS
Collections
Leiosporoceros dussii (Steph.) Hässel was collected from stream banks and along roads in El Valle de Antón, province of Coclé, Panama, where it is abundant and grows on white-volcanic soil and rocks (Figs. 1, 2). A total of five living populations were sampled—Araúz et al. 791 (October 2003), Salazar Allen et al. 20932, 20940, 20942 (July 2004), and Villarreal 803a (January 2005)—and were used for light microscopy (LM), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The first collection was from the road to Río Indio and the remaining collections were made in Río el Guayabo. Locations are separated by about 1–2 km. In addition, herbarium specimens from Mexico (Eggers and Frahm MX 34/15, Eggers' personal herbarium), Ecuador (Gradstein et al. 6797; Gradstein and Frahm 6701, GOET), and Panamá (Villarreal et al. 651, November 2002, PMA) were studied to verify the presence of Nostoc strands in other locations. Voucher specimens, unless otherwise indicated, are housed in PMA.
|
). Cultures were maintained on a 14/10 h light/dark regimen at 25/21°C. Sporelings and young gametophytes were studied with LM and were fixed for SEM and TEM.
SEM
Fresh plants with Nostoc symbionts and axenically cultured gametophytes were fixed in 2.5% glutaraldehyde in 0.05 M Sorensen's buffer (pH 7.2), rinsed in buffer, postfixed for 2 h in 2% OsO4 and 1.5% potassium ferrocyanide in the same buffer, and rinsed in water. The samples were taken through an ethanol dehydration series, ending at 100% ethanol. The material was then critical point dried and sputter coated with gold palladium. Images were captured digitally with a Hitachi S570 (Hitachi, Mountain View, California, USA).
TEM
Fresh material was fixed following the same procedure as for SEM. After dehydration in a graded ethanol series, the specimens were rinsed three times in 100% ethanol for 15 min each, once in 1 : 1 ethanol to propylene oxide (PO) for 15 min, twice in 100% PO for 30 min each, and infiltrated with Spurr's resin with 3 : 1, 1 : 1, 1 : 3 proportions of PO to resin for 24 h each at –20 °C. After two changes in 100% resin, the material was cured at 65°C for 24 h. Thin sections were cut with a diamond knife and collected on copper grids, then stained for 5 min each with 2% uranyl acetate and basic lead citrate and viewed on a Hitachi H7000 TEM at 75 KV.
Light microscopy
Whole plants were examined and photographed with an Olympus BX40 light microscope equipped with digital image capture (St. Louis, Missouri, USA). Specimens fixed and embedded for TEM were thick-sectioned in vertical longitudinal, horizontal longitudinal, and transverse sections. Sections were collected on slides, stained with 1.5% aqueous toluidine blue, sprinkled with sodium borate, and examined with light microscopy. Heterocyst frequency and Nostoc cell dimensions were estimated from 80 total counts from eight longitudinal sections of thalli.
RESULTS
Sporeling development
By 45–60 d after sowing, sporelings in axenic culture are less than 1 mm long, slightly flattened, and swollen in the anterior region (Fig. 3). No apical cells are evident in sporelings, rather cell divisions are localized in the swollen anterior region (Fig. 4). In nearly transverse sections of the apical region of young sporelings, the thallus is solid with no schizogenous cavities (Fig. 4). The cells of the epidermis are smaller and contain one spherical or dumbbell-shape plastid associated with the nucleus (Figs. 4, 5). The cells of the middle region of the sporelings are highly vacuolated with occasionally one or two small chloroplasts (Fig. 4). Epidermal mucilage clefts that consist of two cells separated by a pore are present either in the ventral region near the apex or in the thickened anterior region (Figs. 4–6). Cells of the mucilage clefts typically contain one or two elliptical or dumbbell-shaped chloroplasts (Fig. 6).
|
Within 80–90 d, young plants are composed of numerous small erect lobes or thalli that are more or less club-shaped, i.e., they expand from base to apex (Figs. 9, 10). Mucilage clefts are randomly scattered along the flattened anterior rim of lobes where they slightly protrude from the surface (Figs. 9–11). Apical cells are first evident in this stage, albeit they are inconspicuous and difficult to discern from surrounding cells (Figs. 12, 13). Small intercellular spaces or cavities that span several cell lengths are scattered throughout the interior thallus (Fig. 12). These irregular cavities are schizogenous, i.e., they form by separation between cells at the middle lamella. Serial sections show that the cavities contain mucilage, are often interconnected, and may be in close proximity to apical cells (Figs. 12, 13).
|
|
|
Both longitudinal and transverse sections of the mature thallus show Nostoc colonies developing in the apical region (Figs. 21–26). The apical region of the thallus is massive and tilts downwards, with the apical cell and dorsal derivatives extending in front of the subtending tissue (Fig. 21). The apical cell is wedge-shaped (24–30 x 21–30 µm) and contains sparse cytoplasm and a large central vacuole (Fig. 27). The nucleus and a plastid bearing poorly developed thylakoids are located at the basal end (Figs. 22, 27, 28). Recent derivatives, unlike the apical cell, are rich in organelles, mainly mitochondria, endoplasmic reticulum, and abundant plasmodesmata (Figs. 27, 28). Nostoc canal elongation is perpetuated directly behind the apical cell by separation between dorsal and ventral derivatives (Fig. 22). A series of transverse sections from the apex backward reveals a recent dichotomy and the schizogenous Nostoc canals that can be traced from each of the two apical cells (Figs. 23–26). Compared to surrounding cells that contain abundant plasmodesmata, adjacent walls of newly formed ventral and dorsal derivatives are virtually devoid of connections (Figs. 22, 27–29). Separation of cells occurs progressively from the apex rearward by separation at the middle lamella. Thus, the resulting schizogenous strands elongate in synchrony with thallus growth (Figs. 2, 21, 29, 30). Mucilage is abundant in the schizogenous canals (Figs. 20, 31, 32). Dense, fibrous cell wall material is often visible between nearly separated cells and within the mucilage-filled matrix of the canal (Figs. 29, 30). Cyanobacteria are not found in the initial site of separation between cells; rather the symbiont is visible 2–5 cells behind (Figs. 21, 22, 25, 26). All gametophytic cells associated with Nostoc canals are thin-walled and contain sparse cytoplasm with a large central vacuole, peripheral mitochondria, abundant coated endoplasmic reticulum (ER) vesicles, and rarely, Golgi bodies (Figs. 31–34). Occasionally, pleiomorphic chloroplasts with swollen membranes are associated with the nucleus (Fig. 34).
|
|
|
|
|
DISCUSSION
Unique nature of the symbiosis
The presence of Nostoc strands in schizogenous canals is a unique feature that separates Leiosporoceros from other hornworts and represents a novel symbiotic arrangement in land plants. All other genera of hornworts develop discrete globose colonies with a continuous formation of mucilage clefts as the avenue of multiple invasions of the thallus (Renzaglia, 1978
). The innovative arrangement and development of Nostoc canals in Leiosporoceros eliminates the need for multiple invasions in a single thallus because once in Nostoc strands elongate and form a single integrated network throughout the plant. Unlike the globose colonies of other hornworts, the Nostoc filaments in Leiosporoceros are continuous along the length of the strand and thus maintain continuity in addition to increasing surface contact for exchange between both partners. This concept is consistent with the much reduced number of gametophyte cells that penetrate Nostoc canals in Leiosporoceros as compared with other hornworts (Renzaglia, 1978
).
It is reasonable to speculate that Nostoc invasion takes places at the sporeling or immature plant stage through mucilage clefts at swollen apices similar to those in figs. 9, 10, and 11. This location of clefts at the growing point is ideal for the establishment of Nostoc colonies in the center of the thallus and adjacent to the apical cell. Sporelings of hornworts typically produce mucilage clefts, but the erect club-shaped morphology that characterizes Leiosporoceros sporelings is not evident in other hornworts of similar ages (Renzaglia, 1978
). Continued elongation of the Nostoc strand follows apical growth, including bifurcations. The existence of mucilage clefts and schizogenous mucilage canals that do not harbor Nostoc in sporelings and the absence of clefts and internal mucilage canals in infected thalli support this speculated mode of invasion. Re-infection studies and greenhouse experiments are essential to determine not only the stage and mode of entry but also the possible influence of Nostoc on the elongation of plants and/or the relationship between cleft production and Nostoc presence. The absence of clefts in already infected plants suggests that the Nostoc might signal a halt in production of mucilage clefts. Alternatively, the cessation in cleft formation could simply reflect a developmental loss of cleft production in mature apices. These hypotheses remain to be tested experimentally; axenic cultures of sporelings in this study did not survive to maturity and in vitro re-infection attempts were unsuccessful.
Surprisingly, Nostoc strands in Leiosporoceros have been overlooked or reported as large globose colonies in previous studies (as Anthoceros dussii Steph., Stephani, 1893
; as Phaeoceros dussii (Steph.) J. Haseg., Hasegawa, 1986
; Hässel de Menéndez, 1986
). This misinterpretation is likely the result of poor preservation of herbarium material coupled with the unexpected nature of this symbiotic arrangement.
A key adaptation of hornworts to life out of water is the ability to sequester vulnerable organs and structures in internal spaces; the propensity to make schizogenous spaces, often filled with desiccation-retardant mucilage, is universal in hornworts (Renzaglia et al., 2000
). Virtually any hornwort cell is capable of mucilage production (Proskauer, 1969
; Renzaglia et al., 2000
), and mucilage-filled cavities are prominent features in mature thalli of Anthoceros L., Folioceros D. C. Bharadwaj, some Dendroceros Nees in Gottsche et al. (Proskauer, 1951
; Bharadwaj, 1971
; Hasegawa, 1980
), and occasionally are found in L. dussii (Fig. 18). In establishing Nostoc strands, Leiosporoceros simply may have co-opted a strategy and morphological design ubiquitous in hornworts that maximizes contact with gametophytic tissue while minimizing the effects of desiccation. In such a scenario, the mucilage canals that are widespread in hornworts may have been modified to harbor and elaborate the Nostoc symbiosis.
It is impossible to determine whether this novel arrangement evolved very early in hornwort evolution and was modified subsequently in other taxa or evolved after the divergence of Leiosporoceros. Advantages of both the globose and strand symbiosis equally support these alternate hypotheses. For example, it can be argued that a single invasion of the thallus is adaptive and eliminates the need to further attract the symbiont. Alternatively, multiple invasions would ensure successful colonization and would allow for flexibility in availability of the cyanobacterium.
Mucilage clefts and mucilage origin
The nature of the gametophytic mucilage clefts of hornworts has been a source of some debate. The contention that mucilage clefts in the gametophyte and stomata in the sporophyte are homologous has permeated the literature (Evans, 1939
; Proskauer, 1951
; Parihar, 1970
; Schuster, 1984
; Duff et al., 2004
). Guard cells of sporophytic stomata have highly thickened inner or outer wall ledges and differ markedly from cleft cells that show no specialized wall differentiation (Lucas and Renzaglia, 2002
). Wall elaborations in sporophytic guard cells and the function of clefts as portals for cyanobacterial entrance speak against homology between capsule stomata and mucilage clefts. Mucilage cleft cells typically have one spherical to dumbbell-shaped chloroplast as do other epidermal cells of the gametophyte. In contrast, all epidermal cells, except guard cells, of the sporophyte lack chloroplasts at maturity (Duff et al., 2004
; Villarreal and Renzaglia, unpublished data).
The presence of Nostoc in other hornworts stimulates enlargement of the mucilage cavity that houses the Nostoc filaments (Rodgers and Stewarts, 1977
; Renzaglia, 1978
); without Nostoc, cavities form but do not continue to develop as the thallus elongates. This also appears to be true in Leiosporoceros where mucilage cavities that lack Nostoc are rarely seen and Nostoc colony development follows thallus growth. As the schizogenous space forms, mucilage is produced by the hornwort. Mucilage of similar density is abundant in canals with Nostoc and most likely derives from surrounding hornwort cells. Mucilage contributed by Nostoc is visible as electron lucent halos surrounding cells (Figs. 40, 41). Duckett et al. (1977)
suggested that the mucilage present in Nostoc cavities of Anthoceros and Phaeoceros Prosk. is produced by the cyanobacterium because the gametophytic cells of the hornwort lack specialized secretory cells. Indeed, gametophytic cells associated with Nostoc colonies and any mucilage-producing cavities in hornworts are strikingly devoid of organelles that are typically associated with secretion (Fahn, 1988
; Gunning and Steer, 1996
). Mitochondria and ER vesicles are occasional evident but Golgi bodies are rare; thus, ER vesicles are the probable means of mucilage secretion and not Golgi vesicles as in other bryophytes (Ligrone, 1986
; Duckett et al., 1990
). Mucilage production without the presence of Golgi bodies has been reported in the ligule of Selaginella kraussiana (Kunze) A. Braun (Bilderback and Slone, 1987
). In symbioses, the host provides protection and thus may reduce the need for mucilage production, which plays a role in desiccation resistance and UV-protection in free-living Nostoc (Bergman, 2002
).
Symbiosis in land plants
Among embryophytes, there are no exact parallels with the cyanobacterial symbiosis in Leiosporoceros. Concentric bands of Nostoc present in cycad "coralloid" roots superficially resemble Leiosporoceros strands in the avenue of entry and establishment (Lindblad et al., 1985
; Adams, 2002
). In these gymnosperms, the invasion of cyanobacteria also occurs at a young stage, presumably through injured epidermal cells (Nathanielsz and Staff, 1975
). However, specialized pores such as the clefts in hornworts do not exist in cycads. Cyanobacteria make a lysigenous (cell lysis)-schizogenous (cell separation) path toward the defined phycobiont zone, which develops between the inner and outer cortex (Nathanielsz and Staff, 1975
; Lindblad et al., 1985
; Adams, 2000
). No anatomical similarities exist between the Nostoc symbiosis of hornworts and that of the angiosperm Gunnera, in which the cyanobacteria reside within cells (Bergman et al., 1992
). Cells surrounding Nostoc colonies in Leiosporoceros and other hornworts (except for one isolated report from Anthoceros punctatus L., Duckett et al., 1977
) lack wall ingrowths and differ from the cells associated with the cyanobacterium in Azolla and Blasia, which exhibit wall labyrinths (Duckett et al., 1975
, 1977
; Renzaglia, unpublished data).
Identity and specificity of the cyanobiont
With regard to the identity of the cyanobiont, symbiotic cyanobacteria of plants are typically assigned to either Nostoc or Anabaena (Rippka, 1988
; Rippka et al., 1979
). The taxonomy of Nostoc and related genera is still controversial (Baker et al., 2003
; Rajaniemi et al., 2005
). Although this study does not focus on the strain identity of the Nostoc, there are morphological and ultrastructural differences between the cyanobacteria collected at different times of the year, suggesting lack of host-cyanobiont specificity.
Based on morphological and ultrastructural features, Nostoc 1 (collected in January and October) and Nostoc 2 (collected in July) either represent two strains or morphological plasticity, as has previously been reported in cycads (Grilli Caiola, 1980
; Obukowicz et al., 1981
; Baulina and Lobakova, 2003
). Nostoc 2 has slightly elongated cells, less dense content and more numerous cyanophycin granules (Table 1). Greater production of cyanophycin granules has been correlated with stress conditions, phosphate dearth in the environment and sulfurous pollution (Obukowicz et al., 1981
; Sharma et al., 1982
; Allen, 1984
; Simon, 1987
). Obukowicz et al. (1981)
also reported a seasonal appearance of cyanophycin granules and carboxysomes in Anabaena of Cycas revoluta Thumb. The effects of dry and wet seasons in the tropics have not been investigated in plant–cyanobacteria symbioses, and this clearly deserves attention.
Multiple cyanobacterial strains within an individual hornwort thallus have been reported and confirmed using molecular markers (West and Adams, 1997
; Costa et al., 2001
). Because of the continuous production of mucilage clefts and the independent origin of Nostoc colonies in most hornworts, it is possible for multiple strains to invade a single thallus (Renzaglia, 1978
; Adams, 2002
). With the potential for a single Nostoc strain to enter and proliferate in each Leiosporoceros thallus/lobe, it is less likely that more than one type of Nostoc will invade a single thallus, a situation that mirrors that in cycads (Costa et al., 2004
). However, within the rosette that constitutes each young plant, there are numerous lobes, each with mucilage clefts. Nostoc enters each lobe separately, and this may indeed be a source of variability in the type of cyanobacterium.
Heterocysts are the site of nitrogen fixation in symbiotic filamentous cyanobacteria (Rodgers and Stewart, 1977
; Rippka, 1988
; Adams and Duggan, 1999
; Meeks and Elhai, 2002
). The frequency of these cells in Leiosporoceros is similar to those documented for other plants but higher than those in lichens (Grilli Caiola, 1980
; Honegger, 1980
; Lindblad et al., 1985
; Adams, 2002
). A single akinete was observed in one population (Villarreal 803a), and this occurrence may be related to the unusual habitat of this collection on rocks in a stream, similar to the habitat of Megaceros Campb. Leiosporoceros usually grows on moist volcanic soil. The atypical and unstable environment might have induced akinete production, which in turn could be dispersed with thallus fragments.
Conspectus
The arrangement and growth of Nostoc strands in Leiosporoceros are strikingly different from other hornworts and, indeed, are unique among all land plants. Most notable is the development of Nostoc colonies in synchrony with apical growth through elongation of ramified schizogenous canals. Contemporary studies have now identified a suite of intriguing vegetative, reproductive, and genetic features that support the isolated phylogenetic position of Leiosporoceros among hornworts (Duff et al., 2004
; Shaw and Renzaglia, 2004
; Cargill et al., 2005
; Villarreal et al., 2005
). Further examination of the details of the life history of this plant, coupled with controlled culture experiments, will undoubtedly reveal other novel and unpredicted features of both the hornwort and its cyanobiont.
FOOTNOTES
1 The authors thank N. Salazar Allen, C. Chung, B. M. Araúz, I. Ramírez, J. De Gracia, and M. Chong for collections of Leiosporoceros from Panama. J. Heinrichs (GOET), S. R. Gradstein (GOET), and J. Eggers provided Ecuadorian and Mexican collections, respectively. Habitat pictures were provided by N. Salazar Allen and Z. Li, and the recipe for the culture medium was provided by K. Yoshinaga. The authors are grateful for comments and technical assistance from the Renzaglia "Plant Morphology" Lab and the staff at the IMAGE Center, SIUC. This work was supported by the Smithsonian Tropical Research Institute through N. Salazar Allen, a graduate research award from the Botanical Society of America, and research grants (DEB-0235919, DEB-0235985 and DEB-0228679) from the National Science Foundation as part of the Systematic Biology and Assembling the Tree of Life programs and the Green Tree of Life Project. J. G. Duckett and one anonymous reviewer provided excellent comments to improve the manuscript. 
2 Current address: Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, Connecticut 06269-3043 USA.
3 Author for correspondence (juan.villarreal{at}uconn.edu
)
LITERATURE CITED
Adams D. G. Duggan P. S.. 1999. Heterocyst and akinete differentiation in cyanobacteria. New Phytologist 144: 3-33.[CrossRef][ISI]
Adams D. G.. 2000. Symbiotic interactions. In B. A. Whitton and M. Potts [eds.] Ecology of cyanobacteria: their diversity in time and space 523-561 Kluwer Academic, Dordrecht, Netherlands.
Adams D. G.. 2002. Cyanobacteria in symbiosis with hornworts and liverworts. In A. N. Rai, B. Bergman, and U. Rasmussen [eds.] Cyanobacteria in symbiosis 117-152 Kluwer Academic Publishers, Dordrecht, Netherlands.
Allen M. M.. 1984. Cyanobacterial cell inclusions. Annual Review of Microbiology 38: 1-25.[CrossRef][ISI][Medline]
Baker J. A. Entsch B. Mckay D. B.. 2003. The cyanobiont in an Azolla fern is neither Anabaena nor Nostoc. FEMS Microbiology Letters 229: 43-47.[CrossRef][ISI][Medline]
Baulina O. I. Lobakova E. S.. 2003. Atypical cell forms overproducing extracellular substances in populations of cycad cyanobionts. Microbiology 72: 701-712.[CrossRef][ISI]
Bergman B.. 2002. The Nostoc-Gunnera symbiosis. In A. N. Rai, B. Bergman, and U. Rasmussen [eds.] Cyanobacteria in symbiosis 207-232 Kluwer Academic, Dordrecht, Netherlands.
Bergman B. Johannson C. Söderbäck E.. 1992. The Nostoc-Gunnera symbiosis. New Phytologist 112: 379-400.
Bharadwaj D. C.. 1971. On Folioceros, a new genus of Anthocerotales. Geophytology 1: 6-15.
Bilderback D. E. Slone J. H.. 1987. The ultrastructure of the secreting ligule of Selaginella kraussiana. Botanical Gazette 148: 413-419.[CrossRef]
Campbell D. H.. 1895. The structure and development of the mosses and ferns (Archegoniate) McMillan, London, UK.
Cargill D. C. Duff R. J. Villarreal J. C. Renzaglia K. S.. 2005. Generic concepts in hornworts: historical review, contemporary insights and future directions. Australian Systematic Botany 18: 7-16.[CrossRef][ISI]
Costa J.-L. Martínez Romero E. Lindblad P.. 2004. Sequence based data supports a single Nostoc in individual coralloid roots of cycads. FEMS Microbiology Ecology 49: 481-487.[CrossRef]
Costa J.-L. Paulsrud P. Rikkinen J. Lindblad P.. 2001. Genetic diversity of Nostoc symbionts endophytically associated with two bryophyte species. Applied and Environmental Microbiology 67: 4393-4396.
Dalton D. A. Chatfield J. M.. 1985. A new nitrogen-fixing cyanophyte-hepatic association: Nostoc and Porella. American Journal of Botany 72: 781-784.[CrossRef][ISI]
Duckett J. G. Prasad A. K. S. K. Davies D. A. Walker S.. 1977. A cytological analysis of the Nostoc–bryophyte relationship. New Phytologist 79: 349-362.[CrossRef][ISI]
Duckett J. G. Renzaglia K. S. Pell K.. 1990. Desiccation causes the proliferation of multicellular hairs, but not mucilage papillae, in Cryptothallus mirabilis (Hepatophyta): a correlated light and electron microscopy study. Canadian Journal of Botany 68: 697-706.[CrossRef]
Duckett J. G. Toth R. Soni S. L.. 1975. An ultrastructural study of the Azolla–Anabaena azollae relationship. New Phytologist 75: 111-118.[CrossRef][ISI]
Duff R. J Cargill D. C. Villarreal J. C. Renzaglia K. S.. 2004. Phylogenetic relationships of the hornworts based on rbcL sequence data: novel relationships and new insights. In B. Goffinet, V. Hollowell, and R. Magill [eds.] Molecular systematics of bryophytes. Monographs in Systematic Botany from the Missouri Botanical Garden 98: 41-58.
Evans A. W.. 1939. The classification of the Hepaticae. Botanical Review 5: 49-96.
Fahn A.. 1988. Secretory tissues in vascular plants. New Phytologist 108: 229-257.[CrossRef][ISI]
Granhall U. Hofsten A.. 1976. Nitrogenase activity in relation to intracellular organisms in Sphagnum mosses. Physiologia Plantarum 36: 88-94.[CrossRef]
Grilli Caiola M.. 1980. On the phycobionts of the cycad coralloid roots. New Phytologist 85: 537-544.[CrossRef][ISI]
Gunning B. E. S. Steer M. W.. 1996. Plant cell biology: structure and function Jones and Bartlett, Sudbury, Massachusetts, USA.
Hasegawa J.. 1980. Taxonomical studies on Asian Anthocerotae. III. Some Asian species of Dendroceros. Journal of the Hattori Botanical Laboratory 47: 287-309.
Hasegawa J.. 1986. Anthoceros dussii Steph. (Anthocerotae) and its isobilateral spore tetrads. Hikobia 9: 357-360.
Hässel de Menéndez G. G.. 1986. Leiosporoceros Hässel n. gen. and Leiosporocerotaceae Hässel n. fam. of Anthocerotopsida. Journal of Bryology 14: 255-259.[ISI]
Honegger R.. 1980. Zytologie der blaualgen-hornmoos-symbiose bei Anthoceros laevis aus Island. Flora 170: 290-302.[ISI]
Lechno-Yossef S. Nierzwicki-Bauer S. A.. 2002. Azolla–Anabaena symbiosis. In A. N. Rai, B. Bergman, and U. Rasmussen [eds.] Cyanobacteria in symbiosis 153-178 Kluwer Academic, Dordrecht, Netherlands.
Leitgeb H.. 1878. Die Nostoccolonien im thallus der Anthoceroteen. Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften 77: 411-418.
Lindblad P. Bergman B. Hofsten A. V. Hällbom L. Nylund J. E.. 1985. The cyanobacterium–Zamia symbiosis: an ultrastructural study. New Phytologist 101: 707-716.[CrossRef][ISI]
Ligrone R.. 1986. Structure, development and cytochemistry of mucilage-secreting hairs in the moss Timmiella barbuloides (Brid.) Moenk. Annals of Botany 58: 859-868.
Lucas J. R. Renzaglia K. S.. 2002. Structure and function of hornwort stomata. Proceedings Microscopy and Microanalysis 8: 1090-1091.
Meeks J. C.. 1998. Symbiosis between nitrogen-fixing cyanobacteria and plants. Bioscience 48: 266-276.[CrossRef][ISI]
Meeks J. C. Elhai J.. 2002. Regulation of cellular differentiation in filamentous cyanobacteria in free-living and plant-associated symbiotic growth states. Microbiology and Molecular Biology Reviews 66: 94-121.
Naidu T.. 1977. On the occurrence of a new symbiont in Anthoceros. Current Science 46: 87-88.
Nathanielsz C. P. Staff I. A.. 1975. A mode of entry of blue-green algae into the apogeotropic roots of Macrozamia comunis. American Journal of Botany 62: 232-235.[CrossRef][ISI]
Obukowicz M. Schaller M. Kennedy G. S.. 1981. Ultrastructure and phenolic histochemistry of the Cycas revoluta–Anabaena symbiosis. New Phytologist 87: 751-759.[CrossRef][ISI]
Parihar N. S.. 1970. An introduction to Embryophyta, volume 1, Bryophyta Central Book Depot, Allahabad, India.
Potts M.. 2000. Nostoc. In B. A. Whitton and M. Potts [eds.] The ecology of cyanobacteria; their diversity in time and space 465-504 Kluwer Academic, Dordrecht, Netherlands.
Proskauer J.. 1951. Studies on Anthocerotales III. The genera Anthoceros and Phaeoceros. Bulletin of the Torrey Botanical Club 78: 331-349.[CrossRef]
Proskauer J.. 1969. Studies on Anthocerotales VIII. Phytomorphology 19: 52-66.[ISI]
Rai A.N. Bergman B. Rasmussen U.. 2002. Cyanobacteria in symbiosis Kluwer Academic, Dordrecht, Netherlands.
Rajaniemi P. Hrouzek P. Kastovská K. Willame R. Rantala A. Hoffmann L. Komárek J. Sivonen K.. 2005. Phylogenetic and morphological evaluation of the genera Anabaena, Aphanizomenon, Trichormus and Nostoc (Nostocales, Cyanobacteria). International Journal of Systematic and Evolutionary Microbiology 55: 11-26.
Raven J. A.. 2002a. The evolution of cyanobacterial symbioses. Biology and Environment Proceedings of the Royal Irish Academy 102: B 3-6.
Raven J. A.. 2002b. Evolution of cyanobacterial symbioses. In A. N. Rai, B. Bergman, and U. Rasmussen [eds.] Cyanobacteria in symbiosis 329-346 Kluwer Academic, Dordrecht, Netherlands.
Raven P. H. Evert R. F. Eichorn S. E.. 1999. Biology of plants W. H. Freeman, New York, New York, USA.
Renzaglia K. S.. 1978. A comparative morphology and developmental anatomy of the Anthocerotophyta. Journal of the Hattori Botanical Laboratory 44: 31-90.
Renzaglia K. S.. 1982. A comparative developmental investigation of the gametophyte generation in the Metzgeriales (Hepatophyta). Bryophytorum Bibliotheca 24: 1-253.
Renzaglia K. S. Duff R. J. Nickrent D. L. Garbary D.. 2000. Vegetative and reproductive innovations of early land plants: implications for a unified phylogeny. Transactions of the Royal Society, London 355: 769-793.[CrossRef]
Rippka R.. 1988. Recognition and identification of cyanobacteria. Methods in Enzymology 167: 28-67.[ISI]
Rippka R. Deruelles J. Waterbury J. B. Herdman M. Stainer R. Y.. 1979. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology 111: 1-61.
Rodgers G. A. Stewart W. D. P.. 1977. The cyanophyte-hepatic symbiosis. I. Morphology and physiology. New Phytologist 78: 441-458.[CrossRef][ISI]
Schopf J. W.. 2000. The fossil record: tracing the roots of the cyanobacterial lineage. In B. A. Whitton and M. Potts [eds.] The ecology of cyanobacteria; their diversity in time and space 13-35 Kluwer Academic, Dordrecht, Netherlands.
Schuster R. M.. 1984. Evolution, phylogeny and classification of the Anthocerotae. In R. M. Schuster [ed.] New manual of bryology, volume 2 1070-1092 Hattori Botanical Laboratory, Nichinan, Japan.
Sharma P. Bergman B. Hällbom L. Hofsten A. V.. 1982. Ultrastructural changes of Nostoc of Peltigera canina in presence of SO2. New Phytologist 92: 573-579.[CrossRef][ISI]
Shaw J. Renzaglia K. S.. 2004. Phylogeny and diversification of bryophytes. American Journal of Botany 91: 1557-1581.
Simon R. D.. 1987. Inclusion bodies in the cyanobacteria: cyanophycin, polyphosphate, polyhedral bodies. In P. Fay and C. Van Baalen [eds.] The cyanobacteria 199-225 Elsevier, Amsterdam, Netherlands.
Solheim B. Zielke M.. 2002. Associations between cyanobacteria and mosses. In A. N. Rai, B. Bergman, and U. Rasmussen [eds.] Cyanobacteria in symbiosis 137-152 Kluwer Academic, Dordrecht, Netherlands.
Stephani F.. 1893. Hepaticarum species novae II. Hedwigia 32: 137-147.
Villarreal J. C. Duff R. J. Renzaglia K. S.. 2005. Anatomical and ultrastructural innovations in Leiosporoceros dussii (Steph.) Hässel. XVII International Botanical Congress, Vienna, Austria 450 (Abstract).
West N. J. Adams D. G.. 1997. Phenotypic and genotypic comparison of symbiotic and free-living cyanobacteria from a single field site. Applied and Environmental Microbiology 63: 4479-4484.[Abstract]
Whitton B. A. Potts M.. 2000. The ecology of cyanobacteria; their diversity in time and space Kluwer Academic, Dordrecht, Netherlands.
Yoshinaga K. Linuma H. Masuzawa T. Uedal K.. 1996. Extensive RNA editing of U to C in addition to C to U substitution in the rbcL transcripts of hornwort chloroplasts and the origin of RNA editing in green plants. Nucleic Acids Research 24: 1008-1014.
This article has been cited by other articles:
|
|
 |
|
  D. G. Adams and P. S. Duggan Cyanobacteria-bryophyte symbioses J. Exp. Bot., March 1, 2008; 59(5): 1047 - 1058. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
