Structure and role of rhizomorphs of Armillaria luteobubalina
By
MAMTA PAREEK
A thesis submitted to the University of New South Wales in partial
fulfilment of the requirement for the degree of Doctor of
Philosophy
School of Biological, Earth and Environmental Sciences, The
University of New South Wales, Sydney, 2052, Australia
2006 Appendix: Acronyms
Carboxy-DFFDA Oregon Green® 488 carboxylic acid diacetate
CFD Computational fluid dynamics
CFDA Carboxyfluorescein diacetate
CMAC 7-amino-4-chloromethyl coumarin
CMFDA 5-chloromethyl fluorescein diacetate
DIC Differential Interference Contrast
FDA Fluorescein diacetate
HPTS# 8-hydroxypyrene-1,3,6-trisulfonic acid, trisodium salt
MM Malt marmite
PIPES Piperazine-N-N’-bis (2-ethanol sulphonic acid)
PSP Pseudosclerotial plate
PTS 8-hydroxypyrene-1,3,6-trisulphonate
RHS Right hand side
RO Reverse osmosis
ROL Radial oxygen loss
#also referred as PTS by many authors. Therefore, PTS has been used interchangeably for HPTS at many places as referred by other authors.
i List of publications
Written Publications
1. Pareek, M., Cole, L., Ashford, A.E. (2001) Variations in aerial and submerged rhizomorphs of Armillaria luteobubalina suggests that rhizomorphs are organs of absorption rather than long distance translocation. Mycological
Research 105, 1377-1387.
2. Pareek, M., Ashford, A.E., Allaway, W.G. and Pareek, V. (2005) Mass transport in small-scale biological entities: An application in plant science.
Paper published in 7th World Congress of Chemical Engineers. Paper number:
85553; ISBN number: 0-85295-494-8.
3. Pareek, M., Allaway, W.G., Ashford, A.E. (2006) Armillaria luteobubalina mycelium develops air pores that conduct oxygen to rhizomorph clusters. Mycological Research 110, 38-50.
4. Pareek, M., Ashford, A.E. (2006) Uptake of apoplastic and symplastic tracers by Armillaria luteobubalina rhizomorphs. (Manuscript prepared to be submitted in Mycological Research.)
National/International Conferences
Presenter is underlined
5. Pareek, M., Cole, L., Ashford, A.E. Are rhizomorphs of Armillaria luteobubalina are organs of absorption. Australasian Mycological Society
Conference (AMSC): Cairns, Australia. (Presented as a talk in September,
2001.)
ii 6. Pareek, M., Cole, L., Ashford, A.E. Structure and role of rhizomorphs of Armillaria luteobubalina. (Presented as a talk at Sydney Fungal Studies
Group workshop in October, 2001.)
7. Pareek, M., Ashford, A.E., Allaway, W.G. Rhizomorphs of Armillaria luteobubalina really do have role in aeration. 7th International Mycological
Conference; Oslo, Norway. (Presented as a talk in August, 2002.)
8. Pareek, M., Cole, L., Ashford, A.E. Structure, growth and permeability of rhizomorphs of Armillaria luteobubalina. 7th International Mycological
Conference; Oslo, Norway. (Presented as a poster in August, 2002.)
9. Pareek, M., Ashford, A.E., Allaway, W.G. Are rhizomorphs organs of aeration? (Talk at Sydney Fungal Studies Group workshop in October, 2002.)
10. Pareek, M., Ashford, A.E., Allaway, W.G. and Pareek, V. Mass transport in small-scale biological entities: An application in plant science.
(Presented a talk in 7th World Congress of Chemical Engineers, Glasgow; 10-
14 July, 2005).
iii Acknowledgements
First of all, I will express my deepest gratitude to my supervisor Professor A.
E. Ashford for her unequivocal support to carry out this research. Despite her busy schedule, I could see her whenever I wanted and that too without a prior appointment. Without her whole-hearted support, this work would not have been possible. I indicate my thanks and indebtedness to her for all her help.
I express my sincere gratitude to Professor W. G. Allaway for his timely support and advice on all aspects of this work, especially that on oxygen electrode experiments.
Thanks are also due to Dr. G. Hyde and A/Prof. P. Adam for their valuable advice. The help of Dr. Louise Cole, vis-à-vis microscopy and freezing techniques, is highly appreciated. The support given by Ms. Danielle Davies,
Dr. Bettye Rees, and Dr. P. Williams was also invaluable. I also thank
Professor Alan Walker FAA and Dr. Vishnu Pareek for their valuable suggestions on the mathematical modelling section.
I am greatly indebted to my parents and family for their support during this time. I especially thank my mother for visiting us in Perth for babysitting Mihir
(our 1 year old son), who otherwise was making it impossible to finish this thesis writing. Last but not least, my husband Vishnu’s support before and during this thesis writing is highly appreciated.
iv Abstract
Armillaria luteobubalina is one of the most serious pathogens in Australian ecosystems causing much damage particularly in dry sclerophyll eucalypt forests. It produces rhizomorphs like other Armillaria species but at many sites they do not to extend for long distances in soil, and disease spread is caused by mycelial systems via root contact. The aim of this research was to study the growth rate, structure, possible role(s) in absorption, aeration and transport of rhizomorphs and other spatially related structures.
Two different types of rhizomorphs were produced by A. luteobubalina in in vitro conditions - aerial and submerged. They differed in growth rate, amount of mucilage, extent of peripheral hyphae, degree of pigmentation and in the structure of inner cortex. Otherwise they had a similar internal structure comprising 4 radial zones, namely, peripheral hyphae, outer cortex, inner cortex and medulla. The central medullary space appeared to be a gas-filled cavity and a zone of inflated hyphae interspersed with narrow hyphae developed at the interface between inner cortex and medulla. This resembled higher plant aerenchyma. No vessel-hyphae equivalent to xylem vessels were found.
To examine the role of rhizomorphs in absorption, symplastic and apoplastic tracers were applied to aerial and submerged rhizomorphs. Two membrane permeant symplastic fluorescent tracers, Oregon Green® 488 carboxylic acid diacetate (carboxy-DFFDA) and 7-amino-4-chloromethylcoumarin (CMAC), which are ultimately sequestered in vacuoles, were applied to samples (whole
v rhizomorphs and/or cut sections) sectioned free hand. The apoplastic tracer 8- hydroxypyrene-1,3,6-trisulphonate (HPTS) was applied to fresh material and its localisation determined in semi-thin (dry) sections following anhydrous freeze substitution and dry sectioning. Both symplastic tracers behaved in a similar fashion in aerial and submerged rhizomorphs regardless of whether pigment was present in the outer cortical cell walls or in the extracellular material. Rhizomorphs appeared to be mostly impermeable to these probes with exception of a few fluorescent patches that potentially connected peripheral hyphae to inner cortical cells. In contrast, the apoplastic probe appeared to be impeded by the pigmentation in cell walls and/or the extracellular material in the outer cortical zone.
Structures that I identified as air pores originated in cultures on agar; they arose directly from the mycelium and grew upwards into the air. As they elongated they differentiated into a cylindrical structure with a more compact basal region from which the loose parallel intertwined hyphae emerge. A cluster of rhizomorph apices is initiated immediately beneath a group of air pores shortly after they have begun to develop. Mature air pores became pigmented as did also the surface mycelium of the colony to form a crust. The pigmented surface layer, or rind, extended into the base of air pores, where it was elevated into a mound by tissue inside the base of the air pore. Beneath the rind and pseudo parenchyma there was a region of loose hyphae with extensive gas space between them. This gas space extended into the base of the air pore and was continuous with the central gas canal of rhizomorphs. The gas space was also continuous with the internal spaces of the air pore (and atmosphere) through
vi gaps in the rind layer in its basal region. Oxygen is conducted through the air pores and their associated rhizomorph gas canals (with cut ends) into the oxygen electrode chamber with a conductivity averaging 679r68x10-12 m3s-1.
The time averaged oxygen concentration data from the oxygen electrode chamber were used to compare three different air pore diffusion models. It was found that the widely used pseudo-steady-state model overestimated the oxygen conductivity. Finally, a model developed on the basis of fundamental transport equations (widely used in computational fluid dynamics), was used to calculate oxygen diffusivities. This model gave a better comparison with the experimental data.
vii Contents
APPENDIX: ACRONYMS I
LIST OF PUBLICATIONS II
ACKNOWLEDGEMENTS IV
ABSTRACT V
CONTENTS VIII
CHAPTER 1. GENERAL INTRODUCTION 1
CHAPTER 2. LITERATURE REVIEW 7
2.1 Introduction 7
2.2 Nomenclature of Multi-hyphal Linear Aggregates 9
2.3 Armillaria species and their Rhizomorphs 12
2.4 Factors Affecting Growth and Development of Rhizomorphs 17
2.5 Organization of the Differentiated Rhizomorphs 22
2.6 Zone lines, Pseudosclerotial plates, Pseudosclerotia 25
2.7 Possible Roles of Rhizomorphs 27 2.7.1 Spread of the infection and survival of the fungus 27 2.7.2 Aeration 29 2.7.3 Uptake and translocation 32
2.8 Apoplastic and Symplastic Pathway 39 2.8.1 Apoplastic probes 43 2.8.2 Symplastic probes 47
2.9 Freeze Substitution 49
CHAPTER 3. STRUCTURE AND GROWTH OF SUBMERGED AND AERIAL RHIZOMORPHS OF ARMILLARIA LUTEOBUBALINA 59
3.1 Introduction 59
3.2 Materials and methods 62 3.2.1 Collection and culture of material 62 3.2.2 Growth Experiments 63 3.2.3 Structural Studies 63
3.3 Results 65
viii 3.3.1 Initiation and growth of rhizomorphs in culture 65 3.3.2 External morphology 67 3.3.3 Internal structure 68 3.3.3.1 The outer most zone: Peripheral hyphae 68 3.3.3.2 Outer cortex 69 3.3.3.3 Inner cortex 70 3.3.3.4 Medulla 71
3.4 Discussion 72
3.5 Legends 78
CHAPTER 4. PERMEABILITY OF AERIAL AND SUBMERGED RHIZOMORPHS TO APOPLASTIC AND SYMPLASTIC TRACERS 90
4.1 Introduction 90
4.2 Materials and methods 94 4.2.1 Fungal culture 94 4.2.2 Treatment with symplastic tracers 95 4.2.3 Treatment with apoplastic tracer 96 4.2.4 Anhydrous freeze-substitution 97 4.2.5 Epifluorescence microscopy 98
4.3 Results 99 4.3.1 Symplastic Permeability 99 4.3.1.1 Intact rhizomorphs treated with carboxy-DFFDA 99 4.3.1.2 Rhizomorph sections treated with carboxy-DFFDA 99 4.3.1.3 Intact rhizomorphs treated with CMAC 100 4.3.1.4 Rhizomorph sections treated with CMAC 101 4.3.2 Tracer localization at the extreme apex of hand-cut sections or intact rhizomorphs treated with Carboxy-DFFDA or CMAC 101 4.3.3 Apoplastic Permeability 102 4.3.3.1 Intact aerial rhizomorphs treated with HPTS 103 4.3.3.2 Intact submerged rhizomorphs treated with HPTS 103
4.4 Discussion 105
4.5 Legends 111
CHAPTER 5. CONDUCTION AND TRANSPORT OF OXYGEN TO RHIZOMORPHS THROUGH ‘AIR PORES’ 125
5.1 Introduction 125
5.2 Materials and methods 127 5.2.1 Fungal material 127 5.2.2 Air pore developmental studies 128 5.2.3 Structural studies 128 5.2.4 Oxygen conductance measurements 129
5.3 Results 131 5.3.1 Development of air pores 131 5.3.2 External Morphology of fully developed and mature air pores 134 5.3.3 The internal structure of air pores and maintenance of continuity of their gas space with the central gas space of rhizomorphs throughout development 134
ix 5.3.4 Structures similar to air pores observed on the rhizomorph surface 138 5.3.5 Oxygen conductance 139
5.4 Discussion 141
5.5 Legends 149
CHAPTER 6. NUMERICAL MODELLING OF OXYGEN TRANSPORT THROUGH AN AIR PORE 161
6.1 Introduction 161
6.2 Experimental Set-Up 162
6.3 Unsteady State Diffusion Model 162
6.4 Oxygen Diffusivity 166
6.5 Computational Fluid Dynamic (CFD) Modeling 167
6.6 Conclusions 170
CHAPTER 7. CONCLUSIONS & GENERAL DISCUSSION 176
7.1 Introduction 176
7.2 Rhizomorph Structure and Organization 176
7.3 Permeability 180 7.3.1 Radial permeability 180 7.3.2 Long Distance Transport 185
7.4 Air pores: Initiation, Development and Significance 186
7.5 Numerical Modeling of Oxygen Diffusion in Air pores 191
7.6 Comparison of Rhizomorphs with Higher Plant Roots 192
7.7 Oxygen Diffusion in Rhizomorphs 194
7.8 Concluding Remarks 196
CHAPTER 8. REFERENCES 199
x Chapter 1. General Introduction
Armillaria is one of many genera of basidiomycete fungi which can produce strand-like linear aggregates. In Armillaria they are the most complex and are usually termed rhizomorphs. Armillaria species are abundant, widespread and important fungal pathogens. They are found in a wide range of ecosystems such as natural forests, plantations, orchards, gardens and other similar situations throughout the world. They are known to cause root and butt rot diseases in a wide range of trees and shrubs (Smith, Bruhn & Anderson, 1992).
Among Armillaria species, A. mellea, A. luteobubalina and A. ostoyae are reported to be the most pathogenic (Fox, 2000). Of these, A. mellea is the most extensively studied fungus.
A. luteobubalina is indigenous to Australia. It is the most prevalent species and attacks plants in a wide range of natural ecosystems throughout Australia (Kile,
1981; Kile & Watling, 1981; Pearce, Malajczuk & Kile, 1986; Falk &
Parberry, 1995; Shearer et al., 1998). Often A. luteobubalina attacks the stump and root system of cut trees where it proliferates and survives on the killed tissues to produce a large inoculum base from which the fungus spreads to adjacent live trees (Pearce & Malajczuk, 1990b). Trees invariably die when the pathogen reaches the base of the stem.
A. luteobubalina, like other Armillaria species, produces rhizomorphs and these are implicated in spread of the fungus to infect new trees and spread the disease, but only in some field situations and in pot-grown material (Pearce et al., 1986; Falk & Parberry, 1995; Shearer et al., 1998). In eucalypt forests,
1 especially in Western Australia, rhizomorphs are rare and do not grow for long distances across the open soil; fungal spread within a disease centre occurs along root systems and by mycelial transfer across root contacts (Pearce et al.,
1986; Shearer & Tippett, 1988). The paucity of rhizomorphs in Western
Australian forests is attributed primarily to the absence of suitable conditions of temperature and moisture (Pearce & Malajczuk, 1990a) since rhizomorphs do become important in situations where there is irrigation. Pathogenicity of the fungus has been described but there is little information on the structure of rhizomorphs or their role in this species.
Several roles have been proposed for rhizomorphs. They can extend for long distances up to meters away from the inoculum site apart from fungal survival.
The most obvious role is that of propagation of infection over large distances
(Garrett, 1970). Rhizomorphs are also of interest as they bear a superficial resemblance to higher plant roots which are absorption and transport organs.
Rhizomorphs form direct connections between hyphae utilising different food resources, indicating that they also have role in uptake and translocation.
Several authors have proposed that rhizomorphs carry out translocation of water and nutrients (Granlund, Jennings & Thompson, 1985; Eamus et al.,
1985; Cairney, 1992). Rhizomorphs of A. mellea collected from the field are reported to have large diameter hyphae in the medullary region. These are reported to lack cytoplasmic contents and are thought to act as ‘vessel’ hyphae; and they have been considered as conduits for long distance translocation
(Eamus et al., 1985; Jennings, 1987; Cairney, Jennings & Veltkamp, 1988b).
2 Armillaria mellea is not found in Australia but the rhizomorphs of A. luteobubalina appear superficially similar (Smith & Griffin, 1971). In order to clarify the structure and possible role of rhizomorphs in A. luteobubalina, the first aim of my research was to compare the growth and structure of the two different types of rhizomorph produced in culture i.e. aerial and submerged rhizomorphs. This work is reported in Chapter 3.
Differences observed between aerial and submerged rhizomorphs in A. mellea support a role for rhizomorphs in absorbing nutrients; they have been shown to take up phosphate and other ions from the external environment (Schütte,
1956; Morrison, 1975; Anderson & Ullrich, 1982; Cairney et al., 1988a). These studies are largely based on nutrients labelled with radioisotopes (Morrison,
1975; Anderson & Ullrich, 1982; Cairney et al., 1988a) and the pathway(s) travelled by these absorbed nutrients are still not known. Using apoplastic tracers, Ashford and colleagues have shown that fungal structures in submerged situations are more permeable than those exposed to air. This applies to sclerotia (Young & Ashford, 1996) and the fungal mantle of ectomycorrhizas (Vesk et al., 2000). This reduction in permeability is correlated with the deposition of brown phenolic pigments (melanin) in the walls and extracellular spaces. Rhizomorphs are frequently described to be melanised. Evidence from the structural work of A. mellea suggests that there are differences in the degree of melanisation in submerged and aerial rhizomorphs. Also, more peripheral hyphae are found in submerged than in aerial rhizomorphs, indicating that in submerged conditions rhizomorphs might be involved in exploiting the available nutrients (Granlund, Jennings &
3 Veltkamp, 1984; Cairney, Jennings & Veltkamp, 1988b). Peripheral hyphae are viewed as potential absorbing structures but supporting data are scanty. I investigated the permeability of aerial and submerged rhizomorphs to both apoplastic and symplastic fluorescent probes, to determine whether permeability is correlated with the degree of melanisation and whether peripheral hyphae, which are suggested to be connected to inner cortical zone
(De Bary, 1877; Granlund et al., 1984), are permeable and potentially capable of overcoming the barrier of pigmented outer cortical zones. Anhydrous freeze substitution was used to localise the apoplastic tracer to prevent its redistribution during processing (Vesk et al., 2000). The results are reported in the Chapter 4 of the thesis.
In earlier work on A. luteobubalina (formerly called Armillariella elegans) a role for the rhizomorph in aeration was suggested (Smith & Griffin, 1971).
There are reports that rhizomorphs of Armillaria species are not produced by the submerged mycelium until this reaches the substrate-air interface (Snider,
1959). “Breathing pores’ described as “aborted side branches of the rhizomorph” are found in A. luteobubalina and are suggested to be involved in oxygen diffusion. In addition, elongate fluffy-structures (termed ‘air pores’) are reported to occur on the surface of rhizomorphs in A. mellea (Granlund et al.,
1984; Intini, 1987) and are thought to be involved in providing oxygen. In my study to investigate rhizomorph initiation and development, I noticed that elongate fluffy structures, resembling air pores, developed at the initiation sites of rhizomorphs in recently inoculated cultures. Since the only structures that had formerly been found at initiation sites have been described as
4 “microsclerotia” (Brefeld, 1877; Snider, 1959), it was thought worthwhile to investigate these structures further. I therefore investigated the initiation and development of air pores, their inter-relation to developing rhizomorphs, and their oxygen conductivity, using oxygen electrodes. This study is reported in
Chapter 5.
In order to understand the physiology of these fungi further, it is important to quantify oxygen diffusivity through air pores. Numerical modelling of oxygen diffusion in the air pore is described in Chapter 6. In oxygen electrode experiments, oxygen conductivity is normally calculated on the basis of the cumulative oxygen transferred to the chamber during a fixed time period. This approach is based on an implicit assumption that the oxygen driving force is constant during the experiment; however, in practice, the driving force decreases with the time (due to build up of oxygen in the chamber). To overcome this limitation, a new diffusion model has been developed, based on the fundamental chemical-species transport equation.
To summarise, the main objective of this research project was to examine the structure, development and permeability of A. luteobubalina rhizomorphs and associated structures when growing in two different environments (air and agar). This was done in order to evaluate their potential role as organs of translocation, absorption and aeration. The specific aims were:
1. To study the variation in growth rates and structure of rhizomorphs of A. luteobubalina when growing aerially or submerged in culture (Chapter 3).
5 2. To examine differences in permeability of submerged and aerial rhizomorphs using apoplastic (8-hydroxypyrene-1,3,6-trisulphonate) and symplastic (Oregon Green® 488 carboxylic acid diacetate and 7-amino-4- chloromethyl coumarin) tracers in order to determine the potential role of rhizomorphs as organs of absorption and/or translocation (Chapter 4).
3. To demonstrate the spatial relationship between air pores developed on the mycelium and initiating rhizomorphs and to determine the possible role of these air pores as organs of aeration using oxygen electrodes (Chapter 5).
4. To develop improved numerical methods for the interpretation of the diffusion process inside the oxygen diffusivity electrode chamber (Chapter 6).
6 Chapter 2. Literature Review
2.1 Introduction
The fundamental growth or construction unit of most fungal structures is a hypha, which is a long filament usually consisting of a series of cells. Hyphae occur individually and are explorative. They also aggregate to form familiar structures such as fruit bodies. In many basidiomycetes, the hyphae also associate to form less familiar long multi-hyphal linear aggregates variously known as ‘cords’, ‘rhizomorphs’, ‘strands’, ‘syrrotia’, ‘ropes’, or ‘bundles’,
(Garrett, 1970; Watkinson, 1979; Thompson, 1984). Such multi-hyphal linear structures are found in fungi with a wide array of life styles, from wood decay fungi to ectomycorrhizal species (De Bary, 1887; Dowson, Rayner & Boddy,
1986; Bending & Read, 1995; Boddy, 1999). The structure, number and length of multi-hyphal linear aggregates is influenced both by the habitat and species involved (Rishbeth, 1982; Fox, 2000). They also vary in complexity of internal structure from simple to very complex (Townsend, 1954).
At the complex end of the scale, are the linear aggregated structures of
Armillaria species. These were first described as distinct species of the fungus by Roth as Rhizomorpha fragilis, which was further subdivided into two forms, subterranean and sub-cortical i.e. “Rhizomorpha subterranea” and
“Rhizomorpha subcorticalis” which are found in the soil and under the bark of diseased trees, respectively (see De Bary, 1887). The relationship between these two different types of rhizomorph was later elucidated by Hartig (1874) who showed that both R. subterranea and R. subcorticalis are actually
7 vegetative structures belonging to the same species as the then separately known fruit body Agaricus melleus, a synonym of Armillaria mellea (Vahl ex
Fr.). Brefeld (1877) further confirmed Hartig’s discovery by isolating cultures from single spores obtained from freshly collected mushrooms supplied by
Hartig. He demonstrated that a single spore, isolated in pure culture, grew into a mycelium which produced two different types of rhizomorphs that were similar to those found attached to the fruit body of Agaricus melleus from which the spore had been isolated.
Rhizomorphs were formerly differentiated from all other types of aggregated structure on the basis of their complexity and apical mode of growth (Garrett,
1970; Rayner et al., 1985). They are known to form large branched networks that spread from diseased tree stumps into the soil or along under the bark of diseased trees. Using techniques of molecular genetics Smith, Bruhn &
Anderson (1992) identified an individual of Armillaria bulbosa (= A. gallica
Marxm. & Romgn.), which occupied an area of at least 15 hectares in a
Michigan hardwood forest. The networks of this fungus were found to be enormous, weighing an estimated 10,000 kg, and were reported to have remained genetically stable for more than 1,500 years.
Rhizomorphs are known to be important in aiding infective potential of many
Armillaria species. They are comparable in complexity to higher plant roots and are believed to carry out a range of other functions apart from spreading infection, such as uptake and transport of water and nutrients (Anderson &
Ullrich, 1982; Clipson, Cairney & Jennings, 1987; Gray, Dighton & Jennings,
8 1996) and aeration of subterranean parts of the mycelium (Smith & Griffin,
1971). In particular it has been reported that rhizomorphs contain specialised vessel hyphae that may act as conduits for nutrient transfer (Cairney, 1992) but this is not completely conclusive. There is a need to know more about the internal structure of rhizomorphs of Armillaria species in a number of different environments to evaluate whether there are features that can be correlated with their potential roles in solute transport and gas diffusion, for example, whether there are specific potential pathways for the transport of absorbed nutrients, water, or air.
2.2 Nomenclature of Multi-hyphal Linear Aggregates
As already stated, multi-hyphal linear aggregated organs were formerly variously defined by earlier scientists by all sort of different names such as
‘strands’, ‘cords’, ‘ropes’, ‘bundles’, ‘syrrotia’ or ‘rhizomorphs’, often indiscriminately and sometimes interchangeably for the same fungus (Garrett,
1960a; Garrett, 1963; Garrett, 1970; Watkinson, 1979; Thompson, 1984;
Rayner et al., 1985). Different terms are still used by various authors, perhaps misleading the reader in interpreting these terms as representing fundamentally different structures.
De Bary (1887) appears to be the first person who proposed use of a single term ‘strand’ for all the linear aggregates for all the fungal species. He suggested abandoning the term ‘rhizomorph’, which he found to be superfluous. However, he stated that mycelium of Armillaria mellea produces the most developed ‘strand’ in the category. Garrett (1960a; 1963) found little
9 justification in calling all the multi-hyphal structures ‘strands’ and argued for two different terminologies ‘strand’ and ‘rhizomorph’, based on the fact that morphogenesis of these linear aggregates differ fundamentally; a ‘strand’ develops acropetally by the gradual build up hyphae around a pre-existing mycelial framework, while rhizomorphs grow from co-ordinated hyphal growth like an apical meristem. Butler (1966) in contrast again regrouped all multi-hyphal linear aggregates under the single term ‘strand’, on the basis that hyphal aggregates extend uni-directionally, irrespective of the degree of differentiation and thickness of the hyphae present. Butler's (1966) category
‘strand’ thus includes all the structures variously described by others as
‘mycelial strands’ and ‘rhizomorphs’ in earlier literature. However, Rayner &
Todd (1979) argued that the word ‘strand’ in a true sense describes ‘a single filament’ and therefore it is deceptive to use this term for linear aggregated structures. Moreover, it does not differentiate between truly migratory organs, capable of autonomous extension from a food-base, and those cases where there are initially diffuse mycelial extensions, and linear aggregates later develop amongst these. At one extreme of the spectrum are the true rhizomorphs of Armillaria species (Garrett, 1963) extending from a highly organised apical growing point, and at other are the cords of Merulius lacrymans (Garrett, 1963) and Phymatotrichum omnivorum (Garrett, 1970), where extension of diffuse mycelium is followed by consolidation of linear organs. Rayner et al. (1985) proposed that the best solution is to use two terms
‘cord’ and ‘rhizomorph’, and further to qualify these, where necessary, as
‘apically dominant’, ‘apically branched’, ‘apically spreading’ or ‘apically
10 diffuse’. Thereafter, in his work all linear aggregates were placed along a continuum with apically dominant forms at one extreme (i.e. ‘rhizomorph’) and apically diffuse (i.e. ‘cord’) at the other (Rayner et al., 1985). This obtained wide acceptance in the literature and was widely used in describing linear aggregates.
Initially, rhizomorphs were considered to extend by coordinated apical growth of the hyphal aggregates as in a meristem (Garrett, 1963; 1970). However, it was subsequently found that rhizomorphs do not grow by meristematic activity as thought by Garrett (1963; 1970) and Motta (1969) but grow as a co- ordinated front of apically extending hyphae (Rayner et al., 1985). Cairney,
Jennings & Veltkamp (1989) found little justification in separating
‘rhizomorphs’ and ‘mycelial cords’ because of the similarity in structural pattern found at maturity in these types of linear aggregates. Therefore, they advocated universal use of the term ‘rhizomorph’ for all linear aggregates with the qualification that they could be viewed as either simple or complex. Use of the single term ‘rhizomorph’ thereafter was proposed for all the linear aggregates and is based on the similarities between differentiation of internal structure, and the way these organs extend apically (Cairney, Jennings &
Agerer, 1991). These authors further stated that use of the term ‘rhizomorph’ is sensible for all linear aggregates as it highlights the superficial morphological resemblance to the roots of higher plants and thus helps the reader to relate the functions of rhizomorphs to those of higher plant roots. Regardless of the terminology used, rhizomorphs of A. mellea are at one end of the spectrum, being complex structures with strong apical dominance.
11 2.3 Armillaria species and their Rhizomorphs
Armillaria (Fr.:Fr) Staude (Agaricales, Tricholomataceae) is a genus of important plant pathogens causing root and butt rot of trees in natural forests, plantations, orchards, gardens, and other amenity plantings throughout the world (Shaw & Kile, 1991). The genus Armillaria encompasses approximately
50 species worldwide (Kile, McDonald & Byler, 1991; Watling, Kile &
Burdsall, 1991). They are reported from a wide range of geographic locations
(America, Africa, Europe, and Australasia), have varying host ranges
(broadleaf trees and conifers) and vary in virulence from highly pathogenic to non-pathogenic (Shaw & Kile, 1991; Fox, 2000). There are also reports of differences in wood destroying ability amongst Armillaria species (Roll-
Hansen, 1985).
Armillaria species first attack the stump and root systems of cut trees, and partly or wholly colonize these for long periods to form a large inoculum base.
The fungus may remain in these felled trees or cut stumps for many years and spread to newly planted or regenerating trees in the vicinity (Fox, 2000). The spread of the disease from one tree to another may take place by root contacts, rhizomorphs, mycelium or spores (Rishbeth, 1978; Rizzo & Harrington, 1993) though spores are not very effective in infecting live roots (Fox, 2000). The mode of disease spread in Armillaria species is reported to vary between species, within species and with change in climatic conditions, as well as with soil conditions such as soil temperature or soil pH (Redfern, 1970). It is worth emphasizing that the intra-specific variation in disease spread reported for
12 Armillaria mellea may be due to the lack of distinction made between
Armillaria species studied in the past (Rishbeth, 1982). The name A. mellea has been used to represent several inter-sterile species of Armillaria and it is possible that some of the previous work on A. mellea may in fact refer either to
Armillaria bulbosa [Armillariella bulbosa (Barla) Romagn.], or Armillaria ostoyae (Romagn) Herink, or possibly yet other species (Rishbeth, 1982). For example European A. mellea sensu lato has been split up into 5 different species, i.e. A. mellea (Vahl: Fr.) P. Kumm., A. borealis Marxm and Korhonen,
A. lutea Gillet = A. bulbosa Watling 1987, A. cepistipes Vel. and A. ostoyae
Romagn. = A. obscura (Roll-Hansen, 1985; Fox 2000).
In temperate forests, rhizomorphs provide a major means of spread and subsequent infection of many tree species by most Armillaria species
(Morrison, 1976). However, the ability to produce rhizomorphs is believed to depend on both the species and the environment. Some Armillaria species such as A. bulbosa, A. gallica, A. lutea and A. cepistipes are found to produce them extensively, others (e.g., A. mellea, A. ostoyae) more sparsely, and yet others
(e.g., A. tabescens) hardly at all, (Rhoads, 1956; Rishbeth, 1982; Morrison,
1988; Gregory, 1989). Nevertheless, almost all the Armillaria species produce rhizomorphs when cultured in vitro on agar or in liquid medium, or when mycelium is grown in pots on a sand surface during preparation of inoculum
(Guillaumin & Berthelay, 1981; Rishbeth, 1982; Fox, 2000).
Five Armillaria species are described currently from Australia, namely, A. novae-zelandiae (G. Stev.) Herink, A. luteobubalina Kile and Watl., A. fumosa
13 Kile and Watl., A. hinnulea Kile and Watl., and A. pallidula Kile and Watl. Of these A. luteobubalina is considered to be the most serious and important plant pathogen in Australia (Podger et al., 1978; Kile, 1981; Kile, 1983b; Pearce,
Malajczuk & Kile, 1986; Kile & Watling, 1988; Shearer & Tippett, 1988). A. luteobubalina is so far the only species known to behave as a primary pathogen in native forests (Kile, 1981). The evidence for this, as stated by Kile (1981), is as follows: (i) a constant association of the fungus with disease; (ii) the pattern of disease development within stands; (iii) the correlation between root infection and crown dieback; (iv) host resistance to infection; (v) pathogenicity in pot and field tests and (v) that host vigour is not reduced by the pests, pathogen or environmental stress prior to infection.
A. luteobubalina is widely distributed, having been reported from New South
Wales, Victoria, Tasmania, South Australia, Western Australia (Kile &
Watling, 1981, 1983; Kile et al., 1983). It attacks a wide range of native and introduced species in forests and other natural ecosystems, as well as plantations, orchards, vineyards, parks and gardens throughout Australia (Kile,
1981; Kile & Watling, 1981; Kile et al., 1983; Pearce et al., 1986; Shearer &
Tippett, 1988; Shearer et al., 1997a; Shearer et al., 1998). Kile & Watling
(1981) reported that A. luteobubalina caused disease in 30 species. They found that it mainly affected members of Myrtaceae and Mimosaceae but they also recorded two diseased monocotyledonous species. Currently, A. luteobubalina is viewed as highly unspecialised in host preference (Redfern & Filip, 1991) and the host range has been expanded to 125 species on the basis of disease symptoms (Shearer et al., 1998). In coastal vegetation communities, the most
14 frequently killed species were in Proteaceae followed by Myrtaceae,
Epacridaceae, Papilionaceae and Mimosaceae (Shearer et al., 1998). Three rare and endangered species - Banksia brownii Baxter ex R. Br., B. occidentalis R.
Br. subsp. formosa Hopper and B. verticillata R. Br., found in coastal vegetation of south-western Australia were also threatened by this fungus
(Shearer et al., 1997b). Moreover, A. luteobubalina is reported to be the most prevalent and widely distributed Armillaria species in eucalypt forests of south-east Australia and trees in dry eucalypt sclerophyll forests are found to be highly susceptible to the infection (Kile, 1981; Kile & Watling, 1983; 1988;
Shearer et al., 1997a). Within the forest the fungus has been reported to have a discontinuous distribution. Most of the infections were found to occur in small
(0.1-1.0 hectare) well-defined patches, but larger (up to 20-30 hectare), more diffuse infections have also been reported (Kile, 1983a). Studies of genotypes of the fungus in the locality studied have suggested that A. luteobubalina consisted of a community of genetically distinct mycelia and 44 distinct genotypes have been identified (Kile, 1983b).
Most field studies have been carried out in Western Australia. Here, the mode of spread of the disease is believed to be the root-to-root contact (Shearer &
Tippett, 1988). There is a growing notion that the fungus also spreads aerially by basidiospores, however, attempts to inoculate stumps artificially with basidiospores have failed, and how basidiospores infect woody tissue is poorly understood (Kile, 1983b). Rhizomorphs produced by the fungus are found to extend only for short distances into forest soil (Podger et al., 1978; Kile, 1981;
Pearce et al., 1986). For example, in Western Australian forests, they have
15 been reported to be 8-10 cm in length (Pearce et al., 1986). Rhizomorphs are also found on root surfaces. In one instance, they were reported to be partially embedded in the roots and extended about 3 cm into the soil (Pearce et al.,
1986). Such rhizomorphs were found 5-15 cm below ground. The paucity of rhizomorphs in Western Australia is correlated with the prevailing seasonal patterns of temperature and rainfall, which are reported to be not suitable for rhizomorph growth (Pearce & Malajczuk, 1990a). In contrast, rhizomorphs are more abundant in horticultural situations in South Australia (Falk & Parberry,
1995) and after artificial inoculation in pot-cultures (Podger et al., 1978;
Pearce et al., 1986; Morrison, 1988) and most importantly are responsible in spreading the infection in these situations (Podger et al., 1978; Pearce et al.,
1986; Falk & Parberry, 1995). The abundance of these rhizomorphs here compared with those in Western Australian natural forests is reported to be due to “openness” and use of irrigation (Pearce & Malajczuk, 1990a).
Pearce & Malajczuk (1990a), while experimenting on the 'factors affecting the growth of rhizomorphs in soil', found that incorporation of organic matter into the soil by ploughing tended to increase rhizomorph growth. Morrison (1988), while studying pathogenicity of Armillaria species, found that A. luteobubalina produces rhizomorphs profusely and he reported it to be the second most pathogenic fungus among Europe and Australasian Armillaria species after A. mellea, on the basis of the rhizomorph growth pattern. This indicates that A. luteobubalina has the capacity to produce rhizomorphs profusely, and that local conditions are limiting their production in particular situations.
16 Rhizomorphs are superficially very similar to the roots of higher plants and in some cases they are so similar that it is very hard to differentiate between them.
There are reports of rhizomorphs being found in close association to the roots
(Pearce et al., 1986). It seems possible that they might have been overlooked in natural situations and probably this explains the belief that A. luteobubalina does not produce rhizomorphs. It is also possible that some of the differences described above may be attributed to the strain difference (pers. comm. Mr.
Jack Simpson). For example A. luteobubalina found in Western Australia may belong to a particular strain which does not produce long rhizomorphs.
Despite the importance of rhizomorphs in spreading and causing infection in some situations such as irrigated areas, there is little information on either the structure or function of A. luteobubalina rhizomorphs. Their paucity in the soil in natural ecosystems appears to be different from the situation with A. mellea and many typical European species, which apparently produce them profusely.
There is also confusion and lack of evidence about the abundance of rhizomorphs under different conditions. Further studies on growth are needed to clarify some of these issues. A more detailed study of rhizomorph structure and function in A. luteobubalina could also make a valuable contribution in understanding this difference in behaviour.
2.4 Factors Affecting Growth and Development of Rhizomorphs
Growth of rhizomorphs is reported to vary with species and environmental conditions (Morrison, 1976; Morrison, 1982; Rishbeth, 1982; Fox, 2000).
Several environmental factors (such as temperature, pH, light, organic matter,
17 moisture, aeration, etc.) have been demonstrated to have an effect on rhizomorph growth and development. The most extensively studied of these factors are the moisture level (in particular for A. luteobubalina) and temperature of the substrate, as these are found to be critical for growth of rhizomorphs. In addition to these two factors, the role of aeration and effect of organic matter on rhizomorph growth have also received recent attention. All these four factors are discussed below.
The growth of rhizomorphs of A. mellea, A. gallica and A. luteobubalina is reported to be influenced by differences in water potential in the culture medium (Pearce & Malajczuk, 1990a; Whiting & Rizzo, 1999). Rhizomorphs of all three species were found to grow well between –0.5 and –1.5 MPa. In further detailed experiments Pearce & Malajczuk (1990a) found that A. luteobubalina rhizomorphs grew profusely at –0.6 MPa, and did not grow at –
0.001 MPa. Smith & Griffin (1971) had earlier reported that the apex of rhizomorphs of Armillariella elegans Heim = A. luteobubalina Kile and Watl. needed to be covered by water film for growth. Incidence of Armillaria disease in field situations also supports the view that rhizomorph growth may be limited by the lack of moisture, it being greater in sprinkle-irrigated areas than elsewhere (Adaskaveg & Ogawa, 1990). It also explains the frequent association of the Armillaria root disease with wet soil conditions as found by
Garrett (1944). Dry conditions are also found especially to suppress the more saprotrophic Armillaria species, which grow through the soil (Cruickshank,
Morrison & Punja, 1997).
18 The prevailing temperature in the environment has been found to be important for the initiation and growth of rhizomorphs of Armillaria species. The temperature range for the growth of rhizomorphs has been thought to vary with species (Rishbeth, 1978), but in general, rhizomorphs are not formed above
28 oC. Moreover, rhizomorph initiation has been found to have a narrower temperature range than rhizomorph growth (Rishbeth, 1968). The lack of rhizomorphs of Armillaria species in tropical Africa at low elevations (Swift,
1968) has been explained by prevailing high soil temperatures, which are too high for rhizomorph growth (Rishbeth, 1968; Rishbeth, 1978). Pearce &
Malajczuk (1990a) ascribed the paucity of A. luteobubalina in eucalypt forests in south-western Australia to unfavourably high temperatures and low soil moisture. Rhizomorph initiation and growth of A. luteobubalina in cultures were optimum between 20-26 oC and rhizomorphs failed to form at 30 oC
(Pearce & Malajczuk, 1990a).
Aeration is also thought to be one of the important factors necessary for the growth of Armillaria rhizomorphs (Smith & Griffin, 1971; Morrison, 1976;
Rishbeth, 1978). Many observations indicate that rhizomorphs of Armillaria species need a good supply of O2 for their initiation and sustained growth. The basic similarity in the development of rhizomorphs in cultures and the field is undoubtedly the initiation of rhizomorphs at the substrate-air interface (Snider,
1959). Snider has argued that the requirement for ready access to the atmosphere may be related to an O2 requirement for the initiation of rhizomorphs. He further indicated that rhizomorph apices grow in length much faster than the hyphal tips, and probably have a higher O2 requirement per unit
19 weight. It has also been reported that whenever growing rhizomorphs meet a new interface, they establish a new, direct connection between the lumen and the atmosphere (Intini, 1987) via an aborted side branch (Smith & Griffin,
1971). There is also evidence that in A. mellea, submerged inoculum produces mycelium, but no rhizomorphs are initiated until the mycelium reaches the substrate-air interface (Snider, 1959; Worrall, Chet & Hüttermann, 1986).
There are also reports that rhizomorphs of A. mellea do not grow in O2 deficient situations (Reitsma, 1932). Jacques-Félix (1968) found that rhizomorphs of Armillariella species do not grow in a limited quantity of O2 or in evacuated tubes free of O2. Rhizomorph growth is shown to be affected by the varying amount of oxygen concentration at the surface or the point of origin of rhizomorphs (Smith & Griffin, 1971; Rishbeth, 1978). Rhizomorphs of A. mellea in laboratory experiments were found to grow preferentially towards increasing oxygen levels and decreasing carbon dioxide levels
(Morrison, 1976). Rhizomorphs initiated on inoculum segments buried at 30 or
60 cm depth frequently grew toward the soil surface (Morrison, 1976).
Rhizomorph production in various Armillaria species (dry weight) in the soil is less if the O2 content of the atmosphere above the soil is reduced (Rishbeth,
1978). In Armillariella elegans Heim (now A. luteobubalina Kile and Watl.) higher O2 concentrations at the point of origin of rhizomorphs are found to support greater growth of the rhizomorphs (Smith & Griffin, 1971). Soil disturbances during timber extraction were found to stimulate the growth of rhizomorphs of A. mellea further indicating some role in aeration (Redfern,
1973; Morrison & Mallett, 1996).
20 Both production and growth of rhizomorphs increased after clear cutting
(Stanosz & Patton, 1991). Twery et al. (1990) compared and quantified the abundance and distribution of rhizomorphs of Armillaria species in two different types of stands; one where the soil was unamended and other where the stands had been defoliated 1 and 5 years previously by insects. The authors found that total rhizomorph abundance was greatest on the plot defoliated 5 years before sampling, less on the plot defoliated 1 year earlier, and least on un-defoliated plots.
In A. mellea (= A. ostoyae), rhizomorph production from wood inocula was found to be proportional to the amount of organic matter in the surrounding soil, suggesting the importance of organic matter in the initiation or growth of rhizomorphs (Morrison, 1982). Morrison also suggested that substances acting like growth factors might be absorbed from organic matter by these growing rhizomorphs. Incorporation of organic matter was also found to increase the rhizomorph production in A. luteobubalina (Pearce & Malajczuk, 1990a).
There are several studies which report that the initiation and growth of rhizomorphs are affected by various substances such as ethanol, tannic acid, gallic acid, vitamins, enzymes etc (Garrett, 1953; Moody & Weinhold, 1972;
Shaw III, 1985; Garraway, Hütterman & Wargo, 1991). It is important to emphasise that separation of single factor affecting hyphal extension or rhizomorph growth is difficult as many factors affect the hyphal extension and/or rhizomorph growth and also limit the growth. Additionally, factors may
21 be interrelated in their actions or act synergetically. The concept of ‘limiting factors’ may operate as in photosynthesis.
2.5 Organization of the Differentiated Rhizomorphs
In Armillaria species, two different types of rhizomorphs are recognised from nature, one which is hard and red to black and grows through the soil, termed subterranean, and the other which is pale and soft and grows beneath the bark, termed subcortical (Hartig, 1874). The morphological differences between them may be due to the different environmental conditions prevailing in each of these situations (Hartig, 1874). Both types of rhizomorph may be produced in vitro (Brefeld, 1877; Lopez-Real, 1975). The only substantive difference reported between these two rhizomorph types is due to variation of pigments in the rind region (Hartig, 1870; Townsend, 1954). There are also reports that variation of colour from red to black may reflect the age of the rhizomorph; black rhizomorphs are thought to be more mature (Redfern, 1973; Morrison,
1976).
Rhizomorphs of Armillaria species are complex structures that are 1-5 mm.
Mature rhizomorphs of Armillaria species are reported to possess a common, gross structural pattern comprising a variously melanised, densely packed outer cortex surrounded by a loose network of hyphae associated with mucilage
(Hartig, 1870; De Bary, 1887). The melanised outer cortical zone is termed
‘rind’ in the literature (De Bary, 1887; Townsend, 1954; Motta, 1969). The sub cortical layer beneath the rind shows a continuous transition from the melanised cortex to the medulla. The medullary region is variously developed,
22 but often contains a large central space. This central space in the mature region may be filled either with fine hyphae interspersed in all directions (Stanosz,
Patton & Spear, 1987) or with both narrow and wide hyphae also interspersed
(Stanosz et al., 1987).
It is important to emphasize that most of the information described here is based on study of a mature region close to the apex of rhizomorphs and that the terminology used to describe various regions of the rhizomorph does vary with authors (see table 2.1). There may be variation in overall structural dimensions and hyphal diameter, while overall zonation of the rhizomorph remains the same (Cairney et al., 1991).
Armillaria mellea is the most extensively studied species of Armillaria and its rhizomorphs have been studied from more than 150 years by various researchers (De Bary, 1887; Hartig, 1874; Brefeld, 1877; Townsend, 1954;
Motta 1969; Wolkinger, Plank & Brunegger, 1975; Granlund, Jennings &
Veltkamp, 1984; Cairney, Jennings & Veltkamp, 1988b). In all studies authors reached one conclusion that a rhizomorph may be differentiated into 4 zones, namely an outer zone of free hyphae associated with mucilage, a compact outer cortex (which may or may not be melanized), a sub-cortex and a medullary region. Nevertheless, some variation is observed in describing the medullary region.
De Bary (1887), Redfern (1973), Wolkinger et al., (1975) and Granlund et al.,
(1984) reported that the central medullary region is filled with fine hyphae of uniform diameter. In contrast, Townsend (1954), Motta (1969) and Smith &
23 Griffin (1971) all have proposed that the mature rhizomorph has a hollow centre. They proposed that young rhizomorphs near the apex were composed of wide, thin walled hyphae, which later due to “wear and tear” resulted in the hollow centre (Townsend, 1954; Motta, 1969). Granlund et al. (1984) described the central space as empty or filled to varying degrees with both narrow and wide diameter hyphae. They further emphasised that the wide diameter hyphal structures in medullary zone may be due to the mechanical stress brought about by the growth of the rhizomorph. More recently Cairney et al. (1988b) reported large diameter hyphae in the medullary zone of rhizomorphs collected from field and termed them ‘vessel’ hyphae. They described the rhizomorphs collected from field as being ‘mature’, and also stated that mature rhizomorphs have large diameter hyphae in the medullary zone. Granlund et al. (1984) reported similar large diameter hyphae in the medullary zone of rhizomorphs cultured in vitro, contradicting the view that large diameter hyphae in the medullary zone are only found in mature rhizomorphs. It is clear that the literature as yet does not fully clarify this issue.
Some of the confusion regarding the structure of A. mellea rhizomorphs may be explained by the variability found within rhizomorphs including that between rhizomorphs produced in liquid culture, agar culture or in the field. It is also possible that rhizomorphs described as belonging to A. mellea may represent more than one species (see section 2.3) and that variability within the medullary region may reflect particular role of the rhizomorph within the species or conditions under which it has grown.
24 The rhizomorphs of A. luteobubalina appear superficially similar to those of A. mellea but there has been no detailed investigation of their structure. The only structural study done so far on these rhizomorphs is that by Smith & Griffin
(1971), where they showed transmission electron micrographs of rind cells and reported that these cells are thin walled with melanized intercellular spaces. In addition to this, they also observed some emergent apices of rhizomorph branches developed into what they called ‘breathing pores’. These ‘breathing pores’ occur on what they described as aborted branches of rhizomorphs and their central cavity is demonstrated to be connected to the central hollow canal of the main rhizomorph, which is supposed to be involved in oxygen diffusion.
2.6 Zone lines, Pseudosclerotial plates, Pseudosclerotia
Since Hartig’s (1874) first description, almost every paper on wood destroying fungi or wood decay mentions or discusses the dark lines which characterize wood degraded by fungi (Rayner & Todd 1979). Campbell (1934) conducted the first systematic study on these structures called “zone lines” and showed that they can also form in sterile wood blocks inoculated with fungus. Later,
Lopez-Real (1975) carried out morphological and developmental studies on these zone lines and showed similarities between the pseudosclerotial plate
(PSP) formed on cellophane, the cells that constituted the pigmented band in agar, the black crust on sawdust and the bladder cells of the PSP formed in wood. This similarity in structure had also been noted by Campbell (1934).
The brittle PSP of Armillaria consists of a layer of bladder-like melanized cells
25 (often described as pseudo-parenchyma, see Hartig, 1874) formed in three distinct phases (Campbell, 1934; Lopez-Real, 1975):
1. proliferation of hyaline hyphae 3-3.5 μm in diameter,
2. hyphal swelling and aggregation (bladder like cells arise as a chain
branching out from a septum of a hyaline hypha),
3. pigmentation and melanization of hyphae.
Campbell (1934) termed the complete structure enclosed by zone lines a pseudosclerotium, not based on its structure but based on its position. He also proposed that the zone lines prevent the underlying tissues from drying out and the pseudosclerotium thus withstands desiccation in a similar way to a sclerotium. Lopez-Real (1975) considered that “the term pseudosclerotium for the complete structure is highly appropriate whereas that of ‘zone line’ for the sectional view of the outer rind is felt to be inappropriate and misleading”. This author suggested the use of term pseudosclerotial plate as an alternative.
Garraway et al., (1991) list three distinct mechanisms that can stimulate PSP or zone line formation:
1. mechanical and physical factors. These include:-
a. fluctuating moisture content (Campbell, 1934; Lopez-Real & Swift,
1975).
b. gas phase composition (Hartig, 1874; Lopez-Real & Swift, 1977).
c. wounding respiration-induced damage to hyphae (Lopez-Real &
Swift, 1977).
26 Rayner et al. (1985), however, commented that experimental studies done by
Lopez-Real & Swift (1975 and 1977), failed to confirm a role for desiccation or oxygenation as stimuli for PSP formation.
2. Antagonistic interaction of different mycelia (incompatibility reactions).
3. Genetic factors within species.
2.7 Possible Roles of Rhizomorphs
The development of rhizomorphs represents a departure from the normal form of a typical fungal vegetative colony and often involves transition through non- nutrient patches. Rhizomorphs extend for long distances in the soil and several roles have been proposed for them, such as, spread of infection, aiding prolonged survival of the fungus, aeration and finally uptake and translocation of water and nutrients.
2.7.1 Spread of the infection and survival of the fungus
The most obvious and established role of rhizomorphs is that of propagating infection over large distances, (Thomas, 1934; Redfern, 1968; 1973; Reaves,
Shaw III & Roth, 1993). The role of A. mellea rhizomorphs in causing root infection is well established (Garrett, 1960b) and has been fully described by
Garrett (1970). The results from his experiments suggest that rhizomorphs of
A. mellea can colonize dead woody tissues and have the capacity to decompose cellulose and lignin-like substrates, a capability generally restricted to a minority of soil fungi (Garrett, 1960b). Garrett (1970) proposed that the key to the importance of rhizomorphs in the infection process is inoculum potential.
27 This is an intrinsic property of the rhizomorph as shown by Redfern (1973) who observed cut rhizomorph pieces to develop several new branches thus increasing the inoculum density dramatically.
Additionally, it has been suggested that rhizomorphs are less susceptible to damage in fluctuating conditions of external environment (Thompson, 1984) as aggregation enables fungi to conserve and recycle nutrients and resist drying out. There is convincing evidence that aggregation may amplify hyphal sensitivity to external stimuli resulting in directed growth towards new carbon rich food/host bases (Thompson, 1984; Dowson et al., 1986). Pigmentation
(deposition of melanin) in the outer cortical zone is found in subterranean rhizomorphs and is thought to confer an anti-desiccant property that helps rhizomorphs to survive in drier environmental conditions (Townsend, 1954). In general, melanin in fungi confers resistance to many types of environmental stress. It protects the hyphae from physical and biological stresses, may exclude toxic compounds and helps limit loss of useful compounds (Butler et al., 2001). There is also evidence that melanin protects the hyphae from microbial antagonists (Bloomfield & Alexander, 1967; Kuo & Alexander,
1967). Melanization also suggests longevity beyond that of mycelium, which would be expected to be more ephemeral. In rhizomorphs, the melanin layer is highly persistent and can be found intact even when other parts such as the medulla have already decomposed (Reaves et al., 1993). Rhizomorphs are highly persistent and their survival has been reported to be as long as 12 years
(Reaves et al., 1993).
28 2.7.2 Aeration
The role of rhizomorphs in aeration has been studied for more than a century.
Several observations (mostly circumstantial) indicate that rhizomorphs of
Armillaria species need a good supply of oxygen for their initiation and growth. In natural environments and plantations more rhizomorphs have been found in the upper 10 cm of the soil than in deeper horizons (Ono, 1965;
Redfern, 1973). There are also examples where the incidence of disease was greater and more severe in logged sites with fresh moist or dry, well-drained soils than those with heavier, poorly drained and wet soils (Singh, 1975) and a greater abundance of rhizomorphs in the upper layers has been related to greater aeration, as well as higher moisture (Singh, 1981). Morrison (1976) also found that rhizomorphs were concentrated in the upper 10 cm of soil on moist sites, although on dry sites they occurred deeper in the profile. They were rarely found below 30 cm. In vitro when inoculum with rhizomorphs was buried at 30 or 60 cm, the rhizomorphs grew towards increasing oxygen level and decreasing CO2 level, leading Morrison (1976) to postulate that the upward growth of rhizomorphs is associated with these concentrations in soils.
Rhizomorph production or growth is known to be affected by varying external oxygen concentration. In Armillaria species rhizomorph production in the soil was less when the oxygen content of the atmosphere above the soil was reduced (Rishbeth, 1978). In Armillaria luteobubalina, higher oxygen concentrations at the point of origin of rhizomorphs supported greater rhizomorph growth (Smith & Griffin, 1971). Soil disturbances during timber extraction also stimulated the growth of rhizomorphs in A. mellea (Redfern,
29 1973; Morrison & Mallett, 1996). Production and growth of rhizomorphs was observed to increase after clear cutting (Stanosz & Patton, 1991). There is evidence that in A. mellea (Snider, 1959; Worrall, Chet & Hüttermann, 1986) submerged inoculum produces mycelium but no rhizomorphs form until the mycelium reaches the substrate-air interface. Reitsma (1932) reports that rhizomorphs of A. mellea did not grow in oxygen deficient situations. Jacques-
Félix (1968) also found that rhizomorphs of Armillariella species did not grow at low oxygen levels, or in evacuated tubes free of oxygen.
The above studies (Snider, 1959; Jacques-Félix, 1968; Redfern, 1973;
Rishbeth, 1978; Worrall et al., 1986; Stanosz & Patton, 1991; Morrison &
Mallett, 1996) indicate that rhizomorph initiation and growth are affected by oxygen, and suggest that rhizomorphs need to be well aerated. Several authors have suggested that the central space of rhizomorphs is an air-channel and may be involved in O2 translocation (De Bary, 1887; Snider, 1959; Smith & Griffin,
1971). Schmid & Liese (1970) reported that mitochondria are concentrated in the inner medullary region, surrounding the central medulla. Münch (1909) and
Reitsma (1932) have shown that the oxygen travels through the central canal of the rhizomorphs in A. mellea.
Smith & Griffin (1971) also observed ‘tufts’ of hyphae what they described as aborted rhizomorph branches that arose behind emergent rhizomorph tips.
They observed that the rhizomorph tip grew in the air up-to 1.0 cm and darkened in colour, after which it stopped growing until new tips developed beneath the agar. The small lateral ‘tufts’ of free hyphae were shown to
30 develop laterally from a region 0.5 cm beneath the apex. These ‘tufts’, borne on aborted branches, were called ‘breathing pores’. The authors for the first time showed that the central space of rhizomorph and aborted branches bearing
‘breathing pores’ are directly connected. They proposed that air containing O2 diffuses through these ‘tufts’ of hyphae into the central canal of rhizomorphs.
They also observed flattened ribbon like rhizomorphs at the base of the test tube and proposed that this flattening might be due to O2 deficiency. Later on,
Granlund et al. (1984) and Intini (1987) noted the constant presence of ‘air pores’ on the surface of the rhizomorphs of A. mellea and A. obscura, respectively, when in contact with air. They proposed that these air pores were there probably there to overcome the depleting amount of oxygen within the rhizomorph.
There is an increasing notion that air pores may be conducting O2 and rhizomorphs may be involved in translocating O2, but direct evidence for O2 conductance through air pores and rhizomorphs is lacking and information is largely based on the rhizomorph response to different partial pressures of O2 supplied to the surface of the medium or growing mycelium. In Smith &
Griffin’s (1971) experiments, particular atmospheric oxygen levels were found to be necessary for initiation of rhizomorphs while CO2 hindered the growth.
There is no direct evidence that O2 is conducted through the central canal and it is not known how much O2 is conducted.
31 2.7.3 Uptake and translocation
There is no doubt that rhizomorphs are capable of uptake and translocation. A simple demonstration of this is the ability of rhizomorphs to grow out for considerable distances from the woody food base through non-nutrient patches
(Jennings, 1987; Fox, 2000). However, previously rhizomorphs of saprotrophic and ectomycorrhizal fungi have been, in general, interpreted as being different in their roles, the former being thought to be solely explorative (Cooke &
Rayner, 1984; Dighton & Boddy, 1989) in comparison to the latter which were viewed as exploitative (Harley & Smith, 1983; see Cairney, 1991). Recent evidence, however, proves unambiguously that “rhizomorphs” sensu lato of saprotrophic fungi can also absorb nutrients from the external environment
(Anderson & Ullrich, 1982; Clipson et al., 1987; Cairney et al., 1988a; Gray et al., 1996) and may have a role in transport. The majority of studies done so far to show uptake and translocation by rhizomorphs of Armillaria species have used radiotracers (Anderson & Ullrich, 1982; Granlund, Jennings &
Thompson, 1985) with an exception of Schütte (1956) who demonstrated that injected fluorescein moves to growing tips of rhizomorphs. Studies tracing movement of radiotracers give valuable information indicating that nutrients are absorbed and translocated, but are less informative on the specific pathway(s) taken by the absorbed nutrients. There is paucity of information not only on radial pathway(s) of transport across the cortex of rhizomorphs but also on whether any specific hyphae acts as conduits for long distance translocation.
Anderson & Ullrich (1982) and Morrison (1975) showed that nutrients applied basipetally were transferred to an actively growing rhizomorph tip, but not vice
32 versa. Morrison (1975) also showed that the apical region of rhizomorphs of A.
36 32 3- mellea (now A. ostoyae) absorbed chloride Cl and ortho-phosphate PO4 .
However, there was little or no movement of the tracer towards the food base when apices were placed in solutions containing labelled ions. He suggested that the ions absorbed were utilised by the growing tips due to their high metabolic activity. He also observed lower radioactivity in mature regions where rind development was advanced in comparison to the region with a less developed rind nearer the apex and suggested that uptake is influenced by the rind development. He proposed that nutrients available from a food base may be supplemented through the uptake from soil. More recently, Anderson &
14 32 2- Ullrich (1982) observed uptake of C glucose and PO4 and transport from the base to tip in rhizomorphs of A. mellea, growing aerobically in culture across an air gap. The concentration of label was observed to decrease with increasing distance from the base. Rhizomorphs in anaerobic conditions (i.e under an atmosphere of nitrogen) absorbed but did not transport ions, as no movement from base to tip was seen. They believed that if the base were to be converted to a sink for nutrients, rhizomorphs might transport from tip to base.
They ruled out diffusion as a mechanism of translocation and showed that translocated isotopes do not diffuse on the surface or between the cells of the rhizomorphs but instead are carried within the cells.
In contrast Granlund et al. (1985) and Gray et al. (1996) demonstrated that translocation of nutrients in rhizomorphs can be bidirectional. Granlund et al.
(1985) demonstrated that translocation of nutrients in rhizomorphs of
Armillaria mellea (Vahl.: Fr.) can be bidirectional. They demonstrated
33 movement of 3H, 14C, and 32P from the base to tip and also from the tip to base when 3H and 14C were supplied in glucose. They concluded that there is an irreversible movement of radioactivity into a lateral compartment. They also calculated the velocity of translocation of radioactivity, which they estimated to be 0.55-10.8 cm h-1. Gray et al. (1996) also demonstrated bidirectional translocation while measuring 137Cs distribution and translocation through mycelia of A. gallica Marxm. & Romagn. (= A. lutea Gillet) and A. ostoyae
(Romagn.) Herink, growing in small microcosms in the laboratory together with the rhizomorphs of A. gallica, labelled with 134Cs in the field. They concluded that the direction of translocation in rhizomorphs in soil was dependent upon the location of sources and sinks. These studies indicate that most probably nutrient translocation rate and direction depend on the concentration gradient but again the requirement for the nutrients during rhizomorph growth had not been quantified. It is worthwhile considering that, if rhizomorphs of Armillaria withdraw nutrients only from the wood, a shortage of nitrogen and phosphorus is likely to occur, and thus would limit growth. Some researchers believe that water and possibly nutrients are absorbed along the entire length of the rhizomorph (Morrison, 1975). Control of uptake along rhizomorphs is attributed to the development of various potential permeability barriers such as the rind (Jennings, 1987; Gray et al.,
1995; 1996). However, the effect of the rind on rhizomorph permeability is largely unknown. Work done on similar tissues in other fungal structures may be relevant to rhizomorphs. In the elegant studies on permeability of sclerotia using fluorescent tracers, Young & Ashford (1992; 1995; 1996) have shown
34 that the pigment in the cell walls and intercellular spaces impedes tracer movement in to the apoplast. These authors correlated deposition of brown pigment (presumed to be melanin) with the loss of permeability to the fluorescent tracer Sulforhodamine G (mol. wt. 552.59 Da). The fringe of peripheral hyphae produced by some rhizomorphs may circumvent the rind barrier. It is very likely that nutrient uptake takes place into fringing peripheral hyphae, which appear to be connected to the medullary region via hyphae radiating through the rind and cortex. Hyphae emerging from the surface were described by De Bary (1887) and Cairney et al. (1988b). These are thought to increase the surface area significantly for absorption. However, the function of these structures (hyphae) has not been studied in detail and a lot still remains to be understood about these fine radially extending hyphae, not least the role of the copious mucilage that surrounds them.
The route of translocation within rhizomorphs has seen a great deal of speculation. The mechanism of translocation in rhizomorphs also remains obscure.
Translocation in specialised ‘vessel’ hyphae has been proposed by a number of authors. Eamus et al. (1985) and Cairney et al. (1988a) described vessel hyphae in “mature” Armillaria mellea rhizomorphs either collected directly from the field or grown from field-collected logs and compared them to xylem vessels of plants. These vessel-hyphae are thought to be empty and without septa. The vessel-hyphae concept agrees with observations of Duddridge,
Malibari & Read (1980) who described “vessel” hyphae in “rhizomorphs” of
35 an ectomycorrhizal fungus. They found that flow rates of tritiated water through these mycorrhizal rhizomorphs were similar to those reported for higher plant xylem. They suggested that there must be a high conductance pathway and this is most likely to be the central vessel hyphae. Eamus et al.
(1985) measured the hydraulic conductivity of A. mellea rhizomorphs and compared these values with those calculated on the assumption that the vessel hyphae were the major channels of water flow. The calculated values were slightly lower than the experimental values. However, a careful analysis of the values reveals inconsistencies in experimental values presented for particular vessel hyphae of a given mean radius. For example, 1120 vessel hyphae with mean radius of 9.0×10-4 cm show 0.5×10-2 cm2 bar-1 s-1 hydraulic conductivity whereas 980 vessel hyphae with the same radius having 0.6×10-2 cm2 bar-1 s-1 hydraulic conductivity, making it difficult to assume that the vessel hyphae are the sole hyphae responsible for long distance translocation.
It is important to note that the vessel hyphae concept is largely based on calculated fluxes and there is no direct evidence of any empty, thick walled hyphae without septa in functional rhizomorphs. The structural evidence for vessel-like hyphae in the medullary region of mature rhizomorphs of A. mellea is largely based on SEM’s (Cairney et al., 1988b). These do not clearly show whether the vessel hyphae have contents, whether they contain liquid or air, or whether there are cross walls and what their frequency is. Indeed, not all vessel hyphae in saprotrophic rhizomorphs lack cytoplasmic contents (Cairney et al.,
1989). Furthermore there is also no direct evidence of ‘vessel hyphae’ being involved in translocation along mature rhizomorphs (Cairney, 1992).
36 The mechanism of translocation in rhizomorphs is also not clear and few studies have been done to explore this. Anderson & Ullrich (1982) demonstrated that translocation in rhizomorphs is an active phenomenon and it requires living tissues and aerobic conditions. Granlund et al. (1985) demonstrated that translocation occurred simultaneously in both directions.
Jennings (1987) suggested that Anderson and Ullrich’s investigation gave no indication of the mechanism involved; however, Granlund’s bidirectional flow in the rhizomorphs might be accommodated if water and inorganic solutes travel apoplastically via vessel hyphae and bulk flow of assimilates occurs via cytoplasmic cortical hyphae, a situation analogous to transport in xylem and phloem in higher plants. He acknowledged that further investigations are required to ascertain the actual pathways involved but proposed vessel hyphae to be the most probable route. Other scientists believe that translocation in rhizomorphs from source to sink by mass flow is theoretically possible (see
Cairney et al., 1991) as proposed for transport of sugars in sieve tubes of higher plants but the resistance would be high. Cairney (1992) proposed mass flow in vessel hyphae as a mode for translocation and suggested a mechanism for this in both ectomycorrhizal and saprotrophic rhizomorphs (see Figure 2.1).
He suggested that an essentially bidirectional flow occurs in the spatially separated hyphal elements and that the acropetal translocation of carbon from the source supplying the rhizomorph is by mass flow of solution along the vessel hyphae (with contents) of the medulla. Nutrients thus absorbed from the soil are believed to be translocated along hyphae that contain fully functional cytoplasm by contractile processes. Water movement in the reverse direction to
37 that carrying carbon is thought to occur through the walls of the functional hyphae (Cairney, 1992).
Though studies of translocation in Serpula lacrymans (Brownlee & Jennings,
1982a; 1982b) suggest that mass flow of solution can occur in fully functional hyphae, Cairney (1992) believes that this does not occur in rhizomorphs because flow would be impeded by dolipore septa. The preferred mechanism in
Serpula lacrymans is by pressure driven mass flow of solution, as suggested for translocation in the phloem of higher plants (Brownlee & Jennings, 1982a;
Brownlee & Jennings, 1982b; Jennings, 1995). However, there is no certainty that translocation in rhizomorphs of Armillaria species or indeed any other fungus is also brought about by mass flow. As stated by Watkinson et al.
(2005) for cords “we know little about the cellular and subcellular anatomy of the pathway, the mechanism of transport and its driving forces”. Recent evidence on long distance translocation in Phanerochaete velutina indicates pulsatile flow in the cords along the route and does not support mass flow as a mechanism of translocation (Tlalka et al., 2002). However, the authors here admitted that such pulsatile behaviour is not a universal feature for all the foraging saprotrophs as they found no similar pulsing in S. lacrymans under similar conditions. Furthermore, cords of either S. lacrymans or P. velutina are relatively undifferentiated in comparison to highly differentiated rhizomorphs of Armillaria species. Also, there is variation in the velocity of translocation in rhizomorphs of A. mellea, which is 5-10 times less then that in S. lacrymans
(see Jennings, 1987) and P. velutina (Tlalka et al., 2002).
38 Zimmermann (1971) proposed that three criteria need to be fulfilled for pressure driven flow to be accepted as a translocation mechanism in plants.
These are - (1) the conducting channel must be impermeable to water in a lateral direction; (2) it must be very permeable to solutes and water in longitudinal direction and (3) turgor gradients must exist between source and sink. It may be interesting to evaluate rhizomorphs in the light of
Zimmermann’s three criteria. More work is needed to clarify whether vessel hyphae or any other specific hyphae together with vessel hyphae are responsible for long distance translocation, and whether the mechanism is by mass flow.
2.8 Apoplastic and Symplastic Pathway
The routes and the mechanisms of translocation along fungal hyphae remain obscure. However, in higher plants it is widely accepted that the space within the plant body can be divided into a symplast and an apoplast. Both are available as potential pathways for water and nutrient transport. It was Münch who first separated the plant body into the two principal compartments apoplast and symplast (see Canny, 1995). Münch originally described the apoplast as the continuum of wet non-living cell walls surrounding the living protoplasts, and the symplast (confined by the plasmalemma) as the continuum of cells in contact via plasmodesmata. He demonstrated the relationship between these two in his famous mass-flow model by showing two osmotic cells connected by a tube representing a symplast immersed in a water bath, the apoplast. Other groups of scientists working on the uptake of molecules into
39 tissue introduced the concept of 'apparent free space' (Briggs & Robertson,
1957) or 'outer space' (Conway & Downey, 1950) into which solutes move relatively freely, to distinguish it from the ‘apparent osmotic volume’ where solutes, but not the solvent, penetrate relatively slowly (Briggs, Hope &
Robertson, 1961). In the 1960’s, it became clear that apparent free space was equivalent to Münch’s apoplast. The term ‘apparent free space’ got wide acceptance and was commonly used thereafter as it stressed the point that ion movement is influenced by interaction with the non-diffusible anions of the cell wall (see Sattelmacher, 2001). Apparent free space is subdivided into
‘water free space’ and ‘Donnan free space’. Water free space contains only water and diffusible positive and negative ions. Donnan free space contains a fraction with fixed net negative charges (cell wall) so that uptake of positive and negative ions is equal (Briggs & Robertson, 1957; Briggs et al., 1961;
Epstein, 1972). Currently, the word ‘apoplast’ is broadly used to describe cell walls and intercellular spaces irrespective of whether these spaces might contain gas rather than liquid (Canny, 1995). Canny found little justification for the word ‘apoplast’ as used by the scientific community. He states that
“apoplast” was adopted by plant anatomists, apparently unaware of its special meaning, and the term was broadly used for walls and intercellular spaces even when these spaces contained gas, not liquid. He further justifies his comments by explaining that spaces containing water were part of Münch's apoplast and
Briggs's apparent free space, but Gunning & Steer's (1975) free space meant gas. Therefore, to avoid the confusion, Canny (1995) suggested that the term apoplast should not be used alone, but with qualifying adjectives. He proposed
40 that it should be considered separately in sections as xylem-lumen apoplast, cell wall apoplast and intercellular space apoplast, the latter further subdivided into liquid and gas space. In spite of this still all the compartments beyond the plasmalemma are still recognized to constitute the apoplast, i.e. the interfibrillar and intermicellar space of cell walls, the xylem as well as the gas and water filled intercellular spaces in their entirety (Sattelmacher, 2001). The border of the apoplast is formed by the outer surface of plants i.e. the rhizoplane or the cuticle.
Within the apoplast/symplast concept, Steudle (1997) has distinguished three main pathways of water flow, “apoplastic”, “symplastic” and “transcellular”
(vacuolar) paths. However, the symplastic and vacuolar paths cannot be distinguished experimentally and so are summarized as the cell-to-cell pathway
(Drew, 1987; Steudle, 1997). Therefore, in the current context potential pathways will be subdivided into apoplastic and symplastic pathways.
In this work Canny's (1986) definition will be used according to which, the symplast in plants is the continuum of living protoplasts connected through the cell walls by plasmodesmata and the apoplast is the continuum of wet dead cell wall matrices and water in the lumens of vessels. The boundary between these two is cell membrane or plasmalemma (Canny, 1986). The pathway of water and solutes in the plant can be either apoplastic or symplastic, or a combination of the two. Ashford et al. (1989) suggested that the same concept may be used for fungal cells. They proposed that within fungi, as in plants, there is a fundamental division of tissue spaces into the apoplast and the symplast
41 separated from each other by the cell membrane or plasmalemma. The apoplast here represents the continuum of hyphal walls and inter-hyphal spaces/matrices outside the plasma membrane, while the symplast comprises the continuum of cell protoplasts interconnected by septal pores. This is especially relevant to multi-hyphal aggregates such as rhizomorphs. Any internal spaces filled with gas will not by this definition be included in the apoplast. It is obvious that they will not be conduits for solutes dissolved in aqueous solution.
It is valid to view an apoplastic pathway in fungi as similar to that in higher plants. The primary cell walls of plants are essentially a hydrated polysaccharide gel, comprising water-filled micropores between cellulose microfibrils and cross-linked molecules of the pectic and hemicellulosic fractions (Läuchli, 1976). Although the polysaccharides involved are generally different, the walls of fungi are structurally similar to those of plants, consisting of crystalline microfibrils embedded in an amorphous matrix
(Garraway & Evans, 1985; Kuhn & Trinci, 1990). The primary wall comprises water filled pores between chitin microfibrils that are cross-linked by smaller, hydrated poysaccharides and glycoproteins (Peberdy, 1990). The effective pore size of the interfibrillar spaces in both plants and fungi is estimated to be between 3.5-5 nm (Carpita et al., 1979; Papendick & Mulla, 1986). This is much larger than the dimensions of water molecules, hydrated ions or small organic solutes (Läuchli, 1976) and implies that plant and fungal walls should be freely permeable to these substances. However, accessibility of these pores is modified by the ionized groups associated with the matrix, which give rise to fixed charges in the cell wall and bind ions from solution. Consequently,
42 transport through the apoplast will be determined by the ability of the cell wall spaces (water free space) to accommodate ions in solution, plus the nature and density of ion exchange sites (Donnan free space) in the matrix (Lüttge &
Higinbotham, 1979). In higher plant roots there is discontinuity in passage through this system provided by the Casparian band in exodermis where the wall micropores are blocked by the deposition of lignin-like and suberin materials (Clarkson, 1991; 1993). It would appear that the melanin found in fungal tissues plays a similar role and blocks the apoplastic pathway. Melanin which is deposited in both fungal walls and extracellular spaces has been shown to hinder the uptake of fluorescent probes in sclerotia (Young &
Ashford, 1992; 1995; 1996). The rind that forms on the outer surface of rhizomorphs is melanised and very similar to that found in sclerotia. However, it is not known what role it may play in rhizomorphs. To investigate this and elucidate potential pathways of uptake and transport into rhizomorphs whole rhizomorphs will be immersed in apoplastic and symplastic probes and their localisation will be investigated following anhydrous freeze substitution.
2.8.1 Apoplastic probes
Here, PTS (8-hydroxypyrene-1,3,6-trisulfonic acid, trisodium salt) with mol. wt. of 524.37 has been chosen as an apoplastic tracer (Fitzgerald & Allaway,
1991; Vesk et al., 2000) as it is a water soluble and membrane impermeant fluorescent dye. The properties of PTS were first described by Strugger (1939) who used it in a study of the transpiration stream in wheat and primrose. He found that the probe did not move into the protoplasts of living cells except
43 under highly acid conditions. Schlafke (1959) reported that it moved as fast as water and Peterson & Edington (1976) confirmed that the dye is confined to apoplast and that it does not bind to the cell walls. This dye is pH sensitive and found to fluoresce blue in acidic solutions and organelles and green in basic solutions and organelles (Haugland, 2002).
Coloured probes, whether fluorescent or non-fluorescent, have been used to study processes in living plant tissues and have a long history. Probes, which are dissociated at the pH of the plant apoplast, remain in the apoplast and can be used to mark the path of materials that move there (Canny, 1986; Canny &
McCully, 1986). Both positively or negatively charged fluorescent probes are available to trace the apoplastic pathway. Cationic tracers move sluggishly, being attracted to the prevailing negative charges of the cell walls. These subsequently bind to various cell wall polysaccharides and stain the walls, often impeding the further progress of other unbound dye molecules through the walls, so that these accumulate the intermicrofibrillar space (Ashford et al.,
1988). The results thus obtained from cationic tracers are dependent on dye binding as well as permeability (Canny, 1986; Canny & McCully, 1986; Moon et al., 1986; Ashford et al., 1988) and therefore, they are not good tracers of apoplastic pathways. Calcofluor is a strongly fluorescent tracer that has been widely used to trace apoplastic permeability (Peterson, Emanuel &
Humphreys, 1981). Calcofluor binds to E-1,4 glycans or their derivatives in fungal walls and produces intense blue fluorescence (Maeda & Ishida, 1967).
However, there are problems with calcofluor as the result obtained by it depends on dye binding as well as the permeability. Furthermore, the unbound
44 dye might be lost or redistributed and bound dye might further impede the progress of other dye molecules through the walls (Canny & McCully, 1986;
Ashford et al., 1988). Although calcofluor has in the past been used to trace apoplastic permeability barriers in roots (Moon et al., 1986) and in mycorrhizas (Ashford et al., 1988), it has now been replaced by more suitable tracers.
Anionic tracers do not bind to cell walls and are reported to move as fast as water through the apoplast (Canny, 1986; Canny & McCully, 1986). They have been used to study the transpiration pathway in leaves and pathways of movement of water and ions through roots (Canny, 1986; Canny & McCully,
1986). Sulphorhodamine G (SR) and Trisodium 3-hydroxy-5,8,10- pyrenetrisulphonate (PTS) are the probes of choice to study preferential pathways of water and solute flow in leaves and roots (Strugger, 1939; Canny,
1986; Canny & McCully, 1986; Moon et al., 1986; Canny, 1988; Fitzgerald &
Allaway, 1991; Varney, McCully & Canny, 1993). However, it is worth emphasising that the movement of apoplastic probes cannot quantify movement of water in the apoplast (Steudle & Peterson, 1998; Zimmermann &
Steudle, 1998). Zimmermann & Steudle (1998), while testing the permeability of young maize roots to water and ionic compounds using apoplastic tracer
PTS, observed that PTS passed across the roots at different rates from the water, and the permeability of the root to PTS was the same regardless of whether the root had a mature exodermis or not. During its passage across the root, PTS was diluted to a concentration as low as 0.2% of that in the medium.
They proposed that if PTS was a good tracer for water, the apoplastic
45 component of water flow should be only 0.2%. However, they found that this was not true because the comparison between root Lpr and literature data for root cell Lp showed that there must have been a considerable flow of water around protoplasts (Radin & Mathews, 1989; Zimmermann & Steudle, 1998).
Therefore, Zimmermann and colleagues questioned the use of PTS as a tracer for apoplastic water and proposed that it should be abandoned (Steudle &
Peterson, 1998; Zimmermann & Steudle, 1998).
It is important to note that an apoplastic tracer cannot be used to visualize the movement of water, a point well recognised by Canny and co-workers. In contrast to water, such a tracer does not penetrate cell membranes and, even in the apoplastic compartment, its rate of diffusion will be independent of and considerably slower than that of water (see Canny, 1990). However, it should be possible to use it to obtain indirect evidence for the bulk flow of water in the walls by observing apoplastic tracer movement resulting from solvent drag
(Aloni, Enstone & Petertson, 1998), and these tracers are useful to test available apoplastic pathway(s). If a pathway allows solute (ie. apoplastic tracer) to move through it then it must be open to water, and it is therefore very likely that water travels through this pathway too. If a tracer, however, is blocked at a particular point in the apoplast then it cannot be said that it is also blocked to water, since the tracer has a larger molecule with different properties from water. So tracers determine pathways that must be water permeable but not those that are less permeable, or those further on from a block in the apoplast. It is therefore of value to use them but it is necessary to be very cautious with interpretations.
46 2.8.2 Symplastic probes
Undissociated lipid-soluble probes may rapidly cross the plasmalemma and spread in the protoplasm, selectively staining the structures they encounter there (Canny & McCully, 1986). Fluorescein, a symplastic tracer, has traditionally been used to study translocation in phloem (see Canny &
McCully, 1986). Currently, a large range of vital fluorescent probes, most of which are fluorescein analogues, are available to trace symplastic pathways, for example, 5-chloromethyl fluorescein diacetate (CMFDA), 7-amino-4- chloromethyl coumarin (CMAC), fluorescein diacetate (FDA), carboxyfluorescein diacetate (CFDA) and 2ƍ,7ƍ difluorescein diacetate
(carboxy-DFFDA), patented as Oregon Green® 488 carboxylic acid diacetae
(Haughland, 2002).
FDA is a vital stain which is hydrolysed within living cells to release fluorescein, which accumulates in the cytoplasm of cells and produces yellow- green fluorescence when illuminated with blue light (Rotman & Papermaster,
1966). However, FDA was reported to fade rapidly during excitation, making it impractical for general use (Stewart & Deacon, 1995). CFDA is a carboxylated analogue of FDA, which similarly undergoes hydrolysis in living cells to yield carboxyfluorescein, a more hydrophilic compound. In plant cells it prefentially accumulates in vacuoles and is reported to be retained better than FDA (Baron-
Epel, Gharyal & Schindler, 1988). CFDA also accumulates in vacuoles in fungi
(Shepherd, Orlovich & Ashford, 1993). Carboxy-DFFDA is taken up in a similar way to CFDA. It freely permeates through the plasma-membrane and is hydrolysed by esterases in cytoplasm to release fluorescein DFF-, which then
47 accumulates across the tonoplast by a mechanism that is inhibited by the anion transport inhibitor, probencid (Cole, Hyde & Ashford, 1997). There is no similar anion transporter operating in the reverse direction and the probe remains in the lumen of the vacuoles, most likely trapped as the ‘membrane impermeant’ anionic form. CMFDA and CMAC are both membrane permeant probes with reactive chloromethyl groups. Once inside the cell, they undergo what is believed to be a glutathione S-transferase-mediated reaction to produce a cell impermeable fluorochrome, a thioether complex (Haugland, 2002).
Chloromethyl fluorescein diacetate requires enzymatic cleavage to release the green fluorescence product, 5-chloromethyl-fluorescein.
Here, we have chosen CMAC (mol. wt. 209.63) and carboxy-DFFDA (mol. wt.
496.38) to trace potential symplastic pathways in rhizomorphs of Armillaria luteobubalina. Both probes have been selected, as they are membrane permeant and freely pass through the cell membrane and they work well in basidiomycete fungi (Cole, Orlovich & Ashford, 1998). The obvious advantage of using both probes is that the mechanisms by which they are accumulated inside the cell are different and most importantly they have different molecular weights. Carboxy-DFFDA needs esterases for hydrolysis into its fluorescent product while CMAC requires a glutathione S-transferase mediated reaction.
Glutathione levels in most cells are high and glutathione transferases are ubiquitous. CMAC is also described to be very photo-stable during microscopic examinations and is also supposedly aldehyde fixable (Haugland,
2002). Carboxy-DFFDA fluoresces green and is very bright compared with
48 CMAC, which fluoresces a pale blue. The proposed mechanisms of uptake and fluorescence and accumulation of the two probes are given in Figure 2.2.
2.9 Freeze Substitution
The requirement to localise probes that are water soluble at least at the cellular level requires a special approach to specimen preparation. Freeze substitution is a technique which is capable of life-like ultrastructural preservation and is superior to conventional chemical fixing methods (Hoch, 1986; Kellenberger,
1991). Superior quality of preservation is achieved by snap freezing of the specimen; this results in ultra-rapid immobilisation of the cellular contents and avoids chemical fixative-induced artefacts (Howard & Aist, 1979; Roberson &
Fuller, 1988; Kellenberger, 1987; 1991). Rapid freezing supposedly arrests the movement and interaction of cell constituents at the molecular level so that biological structure is trapped in an essentially native state and ice crystals, which would disrupt cellular organisation, have little time to grow (Harvey,
1982; McIntosh, 2001). Freeze substitution permits further stabilisation when proteins (enzymes) are denatured by the substitution solvent at low temperatures (Harvey, 1982).
In spite of their benefits, freezing techniques are not trouble free and failure to freeze the specimen rapidly enough may result crystallisation of tissue water in the specimen, deforming the structural organisation of the material and causing redistribution of soluble substances, leading to artefacts. To avoid this, freezing methods should be very rapid (Howard & O'Donnell, 1987). Ideally the specimen should be rapidly frozen at a very low temperature, so that tissue
49 water is converted into vitreous ice, which is then slowly substituted by an anhydrous solvent at -70 or -80 oC (Feder & Sidman, 1958; Morgan, Davies &
Erasmus, 1978). The problem is that biological specimens are low in thermal conductivity and the cooling rate decreases exponentially with distance from the surface (Morgan et al., 1978) resulting in tissue damage and poor structural fixation in all but a narrow surface region. Plunge freezing of tissue samples into a very cold quenching fluid maximises the cooling rate by maintaining good thermal contact between the cryogen and the entire surface of the specimen (Harvey, 1982). In samples thus frozen in direct contact with a quenching agent damage and element displacement are minimised. Freezing small sized samples is also helpful for better structure and retention of contents
(Feder & Sidman, 1958).
Following freezing, the frozen water must be removed without any melting to avoid any displacement of the soluble cellular elements, or redistribution of fluorescent probes. Use of super-dry solvents and maintenance of a dry environment at all stages avoids water contamination and any consequent redistribution of soluble cellular elements or the supplied tracer (Canny &
McCully, 1986; Orlovich & Ashford, 1995).
Freeze substitution has proved invaluable for, and has been widely used in, studying the distribution of water-soluble metabolites, (Fisher, 1972; Fisher &
Housley, 1972; Altus & Canny, 1985) ions (Harvey, Hall & Flowers, 1976) and exogenously applied water soluble fluorochromes (Canny, 1986; Canny &
McCully, 1986; Canny, 1988; Fitzgerald & Allaway, 1991) in plant tissues and
50 fungal tissues (Vesk et al., 2000; Young & Ashford, 1992; 1995; 1996). Fisher
(1972) in his pioneering work on freeze substitution showed that endogenous water soluble anthocyanin in the beet root was retained in the vacuoles during processing through to embedding and sectioning in Spurr’s resin. Canny &
McCully (1986) demonstrated that freeze substitution enables retention of labile fluorescent probes in cell walls with very great precision if dryness of the solvent is rigorously maintained during all processing steps. It has been reported that mobile ions can be also retained in sub-cellular compartments and frozen droplets during freeze substitution if these rigorous anhydrous conditions are followed (Orlovich & Ashford, 1995; Cole, et al., 1998).
Nevertheless, the technique is not without controversy. There have been reports of overall loss of ions from the tissues during freeze substitution (Edelmann,
1991; Harvey, 1982; Stelzer & Lehmann, 1993). However, measurements of total or relative loss of ions into the substitution fluid give no indication of where the lost ions have come from (see Orlovich & Ashford, 1995). To evaluate freeze substitution technique for ion localization, Orlovich & Ashford
(1995) adopted a novel approach by testing the frozen droplets containing salts and dextran at known concentration as a model system and demonstrated that ions of biological importance could be retained in situ within individual drops after freeze substitution, Spurr’s embedding and dry sectioning. They showed unambiguously that freeze-substitution is a reliable technique and can be used successfully for ion localization at high spatial resolution, provided extreme care is taken to use ultra-dry solvents and resins and dry environment conditions are maintained throughout the procedures.
51 A variety of organic solvents, for example, acetone, di-ethyl ether, ethanol, tetrahydrofuran, have been employed to remove water from frozen samples at low temperatures. There are many claims in the literature regarding the improved morphological qualities of tissue exposed to particular substituting fluids. Rebhun (1972) concluded while reviewing his own experiments that no significant morphological improvements could be detected as attributable to a given fluid. Monaghan, Perusinghe & Müller (1998) also did not find much difference in morphology when using 3 different solvents. However, they mentioned that it might be critical for immunocytochemistry work. The choice of substitution fluid is known to be critical for retention of ions and water soluble fluorescent probes, diethyl ether and tetrahydrofuran being the preferred substituents (Steinbrecht & Müller, 1987; Pålsgård et al., 1994;
Ashford et al., 1988) An important point which emerged from several experiments by various authors is that freeze substitution should be for period not less than several days at a temperature of –70 oC to -80 oC, the time depending on the substitution fluid used (Van Harreveld, Crowell & Malhotra,
1965; Rebhun, 1965; Pålsgård et al., 1994). Ice crystal damage and cracks in the sample are found to be less prominent in slowly warmed samples
(Monaghan et al., 1998). Following freeze substitution, specimens are brought to room temperature for resin infiltration and embedding. The choice of an embedding medium is determined by the nature of the specimen and whether structural work, immunocytochemistry or localisation of water soluble components is the goal. Low viscosity resins are particularly suitable for freeze substitution of plant and fungal tissues, where wall components may hinder the
52 penetration of harder resin (Howard & O'Donnell, 1987). All manipulations including infiltration and embedding must be performed in a dry box to eliminate any artefacts (Fisher, 1972; Morgan et al., 1978; Marshall, 1980;
Canny & McCully, 1986). Harvey et al. (1976) noted that substantial ion losses occurred during sectioning on to aqueous media and recommended the maintenance of anhydrous conditions during sectioning. Altus & Canny (1985) subsequently developed a technique for dry sectioning, which retained labile elements and this method has been successfully used by several authors in tracing experiments with water soluble probes (Canny, 1988; Fitzgerald and
Allaway, 1991; Young & Ashford, 1992; 1995; 1996; Varney et al., 1993).
To conclude, freezing may be extremely quick and arrest vital processes, and minimize autolytic and other changes. Results obtained using this method with fungi have been consistently superior to those attained by conventional fixation. Several filamentous species have been re-examined in detail using freeze substitution. A number of new structures have been described (Howard
& Aist, 1979; Hoch & Howard, 1980; Roberson & Fuller, 1988; Howard &
Ferrari, 1989; Shepherd et al., 1993; Bohrmann & Kellenberger, 2001). This technique has also been proven suitable for retaining tracer molecules in situ in fungal sclerotia and mycorrhizal roots. Therefore, it would not be unreasonable to expect this to hold for rhizomorphs and freeze substitution will be used in experiments to trace apoplastic pathways.
53 Figure 2.1. Schematic diagram of a rhizomorph, growing from a unit of carbon resource showing putative pathways of translocation, taken from Cairney (1992). Shaded hyphae are septate and retain their cytoplasmic contents. Unshaded lengths of vessel hyphae (defined apoplast of vessel hyphae) are without cytoplasm and have undergone septal dissolution. Carbon absorbed by the loading hyphae at the carbon resource is envisaged as being translocated acropetally by cytoplasmic streaming in the hyphal symplast to the sink at the apoplastic junction (a). Turgor pressure created by carbon efflux would generate a mass flow of solution in the apoplastic phase of ‘vessel’ hyphae, maintained by absorption of carbon by hyphae at the symplastic junction of the growing front (b). Carbon compounds could then move by cytoplasmic streaming to sinks at the apices of extending hyphae. Nutrients absorbed from soil by hyphae at the growing front are suggested to be translocated basipetally by streaming in the cytoplasm along the entire rhizomorph length. Water may flow along the rhizomorph apoplast (which excludes the ‘defined apoplast’ of ‘vessel’ hyphae) by mass flow along a gradient of water potential generated at the base. Removal of the water at the carbon resource may increase the rates of cytoplasmic streaming, permitting basipetal movement in the symplast. Some acropetal movement of carbon may also occur in the symplast of these hyphae. The apoplastic phase of ‘vessel’ hyphae may be isolated from the remainder of the apoplast, and so the cytoplasmic translocatory stream, by the presence of hydrophobic ‘vessel’ hyphal walls and deposition of poorly-permeable extracellular material. Hyphal branches connecting cytoplasmic hyphae with apoplastic regions of ‘vessel’ hyphae may occur at points within the differentiated rhizomorph.
54 Figure 2.2 Scheme of carboxy-DFFDA and CMAC uptake into cells.
Carboxy-DFFDA is a colourless and non-fluorescent probe. It enters the cell freely, being membrane permeant. Once the probe is inside the cell, cytosolic esterases present in the cytoplasm cleave off the acetate and release a hydrolysis product, a bright fluorescent (DFF-) and this is transported to vacuoles by an ATP-dependent transporter, where it remains trapped.
CMAC is also a colourless and non-fluorescent probe and, being membrane permeant it enters freely into the cell. Once the probe is inside the cell, it undergoes what is believed to be a glutathione S-transferase-mediated reaction, producing a membrane impermeant reaction product, thiol-ether adduct, which is fluorescent and this is transported to vacuoles by an ATP- dependent transporter, which recognizes glutathione conjugates as substrates.
55 Table 2.1 Rhizomorph structure of Armillaria species as described by researchers since 1870’s
Author Year Species studied Culture/field Type of Zonation within rhizomorph as described during structural studies material rhizomorph studied Cairney et al. 1988 Armillaria mellea Field collected Below apical Below Apex: Mature: & infected region & mature Mucilage Rind logs incubated region Cortex Mid cortex in vitro Medulla Inner cortex Central space Medulla (vessel hyphae)
* fine hyphae protruding from external surface observed on rhizomorphs Intini 1987 Armillaria obscura Culture Aerial & Peripheral hyphae (2-3 Pm) (Persoon 1800 submerged Cortex (2-3 Pm) – protective Herink in Hasek) Sub-cortex (3-8 Pm) - transport nutrients [Armillaria ostoyae Medulla (5-15Pm) – O. layer of medulla- transport nutrients (Romagnesi) Central space (16-23 Pm) - Balloon like hyphae- reserve function Herink] Stanosz et al. 1987 Armillaria species Field Two types Thin rhizomorph Thick rhizomorph Thin Thick walled (small lumen) 1st and 2nd zone less differentiated Thick >2mm Thin walled (wide lumen) (3)Fibre hyphae Loose fibre hyphae (filled or sparse) (4) Thin (1-2 Pm) and thick - Thin (1-2 Pm) and thick - swollen (2-8 swollen (20 Pm) hyphae Pm) hyphae * surface or peripheral hyphae absent Eamus et al. 1985 Armillaria species Field Rind Cortex - similar dimensions as mentioned in Wolkinger et al. Outer Medulla – loose inter-woven narrow diameter (1-2 Pm) Inner Medulla – wide diameter (10-20 Pm) interspersed with narrow diameter hyphae (thick walled with small lumen)
56 Author Year Species studied Culture/field Type of Zonation within rhizomorph as described during structural studies material rhizomorph studied Granlund et al. 1984 Armillaria mellea Culture Aerial (1)Peripheral hyphae (2.2 Pm) (Vahl ex Fr.) rhizomorphs (2) Cortex (2.3 Pm) Submerged (3) Medulla (13.9 Pm) *(subcortex- 4.7 Pm) rhizomorphs (4) Cental space (cottony pith) Shown diameter of hyphae in table contradicting zone names* Powell & Rayner 1983 Armillaria bulbosa Logs Tips of Mucilage producing region (cells - 2-6 Pm) incubated in rhizomorphs Densely packed small cells (1-5 Pm) vitro Densely packed large cells (8-20 Pm) Loosely packed cells (8-20 Pm) Young rhizomorphs- both fine and wide diameter hyphae (large vacuolate cells unpublished) Motta & 1982 Armillaria Culture Apical region Shown apical hyphae and apical cell (Zone terminology different) Peabody tabescens (scop. Ex Lateral meristem differentiate distally into 3 cortical layers Fr.) Peripheral hyphae, centrifugally Subcortex, ill defined intermediate area (may proliferate inward or outward Secondary medulla, thin walled Wolkinger et al 1975 Armillaria mellea Field Subcortical Outer cortex (5-8 Pm) (Vahl ex Fr.) Karst rhizomorph- Mid-cortex – may act as vessel hyphae = Agaricus malleus beneath bark Medulla- filled with fine hyphae (1-2 Pm) Vahl Redfern 1973 Armillaria mellea Field Red to brown to Young rhizomorph - Hollow centre black Old rhizomorph – filled with compact mass of hyphae (as seen by Brefeld, rhizomorphs 1877)
Smith & Griffin 1971 Armillaria elegans culture Brown Rind - thin walled cells surrounded by large intercellular spaces. Pigment Heim (=Armilaria rhizomorphs located in intercellular spaces. luteobubalina) from nutrient Central space hollow and connected to intertwining hyphae borne on emergent broth apices of rhizomorph
57 Author Year Species Culture/field Type of Zones rhizomorph Motta 1969 Armillaria mellea Culture Apical region Apical region consists of 3 concentric layers from which all the tissues of mature rhizomorph derive Apical centre extends laterally to give rise to parallel iso-diamatric thickened walls which later differentiates into subcortex laterally and secondary medulla medially Posterior to apical region- inflated designated as primary medulla Anteriorly apical initials give rise to thin layers which later become laterally oriented from loose networks- peripheral hyphae Differentiated rhizomorph at maturity: Outer cortex (rind- may get pigmented) Sub cortex – give rise to lateral branches (pH) & centrally to form a cottony pith Central medulla (hollow, loosely woven, inflated or collapsed hyphae) Townsend 1954 Armillaria mellea Probably in Described Subterranean Sub-corticalis culture subterranean and Filamentous thin walled hyphae - 2-3 Pm Irregular layer of cells with few subcorticalis (absent in old) projecting hyphae rhizomorphs Rind – thick walled fused cells (4-7 Pm) Slightly thickened cells but not Central Medulla – Broad thin walled dark in colour cells. Divided into two parts (a) outer- 5- Broad thin walled cells 7 Pm (b) inner- 12-15 Pm Central region hollow with a few In mature inner medullary cells tear apart scattered hyphae to leave a hollow centre Goffart 1902 Armillaria mellea Rind Vahl Cortex Medulla (vessel hyphae) De Bary 1887 Agaricus malleus Not known Subterranean Subterranean rhizomorphs Gelatinous felt with numerous spreading hair-like branches Rind – have brown membrane (12 rows without intercellular spaces) Medullary hyphae in connection with the inner rind cells – large celled tissue, firm and septate
58 Chapter 3. Structure and Growth of Submerged and Aerial
Rhizomorphs of Armillaria luteobubalina
3.1 Introduction
Armillaria luteobubalina Watling and Kile is an important pathogen that attacks a wide range of native and introduced species in forests and other natural ecosystems, plantations, orchards, vineyards, parks and gardens throughout Australia (Kile, 1981; Kile & Watling, 1981; Kile et al., 1983;
Pearce, Malajczuk & Kile, 1986; Shearer & Tippett, 1988; Falk & Parberry,
1995; Shearer et al., 1997a; 1998). Of the five Armillaria species described from Australia, namely, A. novae-zelandiae (G. Stev.) Herink, A. luteobubalina
Kile and Watl., A. fumosa Kile and Watl., A. hinnulea Kile and Watl., and A. pallidula Kile and Watl., A. luteobubalina is considered to be the most serious pathogen and it is the prevalent species in dry sclerophyll eucalypt forests (Kile
& Watling, 1983; 1988). Trees in dry eucalypt sclerophyll forests are highly susceptible to infection and infected areas are characterised by a discontinuous distribution of discrete disease patches amongst a group of trees (Podger et al.,
1978; Pearce et al., 1986; Shearer & Tippett 1988; Shearer et al., 1997a).
These contain distinct fungal genotypes consistent with the concept that an initial basidiospore infection is followed by local vegetative spread (Kile,
1983). Often A. luteobubalina attacks the stump and root system of cut trees and may totally colonise this over a long period, to produce a large inoculum base from which the fungus spreads to adjacent live trees where it proliferates
59 and survives on the killed tissues (Pearce & Malajczuk, 1990b). Trees invariably die when the pathogen reaches the base of the stem.
A. luteobubalina, like other Armillaria species, produces highly differentiated rhizomorphs and these are implicated in spread of the fungus to infect new trees and spread the disease, but only in some field situations and in pot-grown material (Podger et al., 1978; Pearce et al., 1986; Falk & Parberry 1995). In eucalypt forests, especially in Western Australia, rhizomorphs are rare and do not grow for long distances across the open soil; fungal spread within a disease centre occurs along root systems and by mycelial transfer across root contacts
(Pearce et al., 1986; Shearer et al., 1997a; Shearer & Tippett 1988). The paucity of rhizomorphs in Western Australian forests is attributed primarily to absence of suitable conditions of temperature and moisture (Pearce &
Malajczuk, 1990a). However this contrasts with the situation for other linear aggregated mycelial structures of saprotrophs and mycorrhizal fungi in the same location (Pearce et al., 1986) and suggests differences in growth requirements and function. Apart from the above work, there is little information on the structure and growth of rhizomorphs of A. luteobubalina, despite its economic significance as a serious root pathogen.
Several other roles have been ascribed to Armillaria rhizomorphs in addition to propagating infection effectively over long distances (Smith, Bruhn &
Anderson, 1992). Early work suggested that A. luteobubalina (formerly called
A. elegans) rhizomorphs play a role in aeration (Smith & Griffin, 1971). There is evidence that rhizomorphs absorb nutrients; those in Armillaria mellea have
60 been shown to take up phosphate and other ions from the external environment
(Morrison, 1975; Cairney et al., 1988a). In addition it has been proposed that rhizomorphs carry out translocation of water and nutrients (Granlund, Jennings
& Veltkamp, 1984; Eamus et al., 1985; Cairney 1992). Many rhizomorphs including those of A. mellea collected from the field contain large diameter hyphae in the medullary region. These are reported to lack cytoplasmic contents and are thought to act as ‘vessel’ hyphae, conduits for long distance translocation (Eamus et al., 1985; Jennings, 1987; Cairney, Jennings &
Veltkamp, 1988b). However some vessel hyphae are known to have contents and not all the rhizomorphs have vessel hyphae (Cairney, Jennings &
Veltkamp, 1989). There are differences not only in describing the structure of rhizomorphs but also in the nomenclature of their radial zonation. Clearly some of this confusion is attributable to variation in the structure of rhizomorphs growing under different conditions. However, in many cases observations have been based primarily on scanning electron micrographs which are excellent for gross structure but give little information on cell structure and do not clarify whether particular cell profiles had contents and were alive at the time of sacrifice.
Therefore, to compare their structure and growth, rhizomorphs of A. luteobubalina have been grown submerged in liquid culture and on an agar surface, and examined, using a range of techniques that discriminate better between cell contents and extracellular material. This was done to see whether structural differences under these two sets of growth conditions could shed
61 light on which the above proposed functions applies to A. luteobubalina. It also allows comparison with rhizomorphs of the better-studied A. mellea.
3.2 Materials and methods
3.2.1 Collection and culture of material
The isolate (930199KGS) of A. luteobubalina Kile & Watl. used in all experiments was obtained from the culture collection of the University of
Western Sydney. This culture was originally isolated from fruit bodies collected from Gore Hill Park, (33.5o S, 151.15o E) North Sydney in June 1993.
Stock cultures were maintained in Petri dishes on malt marmite (MM) agar (2
% malt, 2 % agar and 0.1 % marmite w/v) at 23 oC and sub-cultured every 4-8 wk. To produce rhizomorphs, plugs of inoculum (10 mm diam.) were cut from the edge of actively growing colonies and transferred to one of the following:
(i) 2 % MM agar in Petri dishes, (ii) 2 % MM agar in Petri dishes covered with a layer of cellophane, inoculated and then covered with second cellophane layer (Cole, Hyde & Ashford, 1997), and (iii) MM liquid medium (without agar). Test tubes (150 mm X 12 mm internal diam.) were filled to within 20 mm of the top with liquid MM medium and capped with aluminium foil.
Media were autoclaved for 15 min at 1.5 MPa. A plug of inoculum (10 mm diam.) was then placed on the surface of each medium and incubated in the dark at 23 oC. Liquid culture tubes were kept upright.
62 3.2.2 Growth Experiments
Rhizomorphs were cultured on MM agar in Petri dishes at a constant temperature of 23 oC. Five rhizomorphs were selected randomly as soon as they appeared on each of four Petri plates and their length was measured using a cotton thread at approximately the same time every day for 14 days.
Rhizomorphs which emerged from the agar medium and grew either on its surface or out into the air are termed “aerial rhizomorphs” even though they were initially submerged.
3.2.3 Structural Studies
All rhizomorphs were examined 3-4 wk after initiation. Aerial and submerged rhizomorphs of similar length were compared. The gross external morphology of live rhizomorphs was examined with a M400 Wild dissecting microscope and recorded on Ektachrome 64T professional colour reversal film. Surface structure was also examined using a scanning electron microscope (SEM).
Rhizomorphs (1-4 cm in length) were excised from the mycelium and either kept intact or cut into smaller segments approx 3-5 mm in length. They were then washed in distilled water, dehydrated through a graded ethanol-water series and critical point dried in dry acetone (Eamus et al., 1985). Transverse or longitudinal sections were then cut, mounted on stubs, sputter-coated with gold-palladium for 4 min and examined with a Cambridge Stereoscan S360 at
20kV. Images were recorded with a digital camera.
For light microscopy of sections, rhizomorphs were either fixed in situ by flooding the Petri dish with 5 % (v/v) glutaraldehyde solution in 0.08 M
63 piperazine-N-N’-bis (2-ethanol sulphonic acid) or PIPES buffer (pH 7.0) for 3 h at room temperature. In the case of liquid cultures, mycelium and attached rhizomorphs were removed from the test-tube with minimum disturbance, placed in a clean Petri dish and immediately flooded with fixative for 3 h at room temperature. After fixation, rhizomorphs were rinsed 3-4 times with
0.3M PIPES buffer at pH 7.2 (Salema & Brandão, 1973) at 4 oC, dehydrated through a graded ethanol-water series (4 oC) and gradually infiltrated with medium grade LR White resin (London Resin Co.) over 5 d also at 4 oC.
Samples were polymerised at 60 oC for 24 h in gelatin capsules. Semi thin sections (1.5-2 Pm) were cut on a Reichert Ultracut microtome using glass knives, collected on to a drop of water on slides and dried on a hot plate at
35oC for 12 h. Sections were either stained with 0.05 % toluidine blue (pH 4.4), and observed with bright-field optics or stained with 0.1 % Calcofluor White
M2R (O'Brien & McCully, 1981) and observed by epifluorescence microscopy with blue excitation (EX 365, DIC 395, EM 420). Digital images were captured using a Zeiss Axiophot microscope equipped with an image point cooled CCD video camera (Photometrics, Tucson, AZ, U.S.A.) via a PCI- compatible LG3 framegrabber (Scion corp., Fredrick, MD, U.S.A.), using
Scion version of NIH image software installed on a Power Macintosh 9500/132 computer. The digital images obtained were processed using Adobe Photoshop
6.0 software and plates were prepared using Adobe Illustrator 10 software.
64 3.3 Results
3.3.1 Initiation and growth of rhizomorphs in culture
Rhizomorphs were readily produced by A. luteobubalina both on agar and in liquid medium, but they only appeared after some growth of the mycelium had first taken place. On agar, a narrow zone of fluffy hyphae developed from the inoculum plug and from this rhizomorphs emerged and grew radially across the plate (Figure 3.1).
No rhizomorphs were produced from cultures inoculated between the two cellophane layers unless the cellophane was wrinkled to leave a small gap. In this case rhizomorphs emerged radially through this gap (Figure 3.2).
In liquid culture a fluffy mycelium was first produced at the liquid-air interface and then rhizomorphs were initiated and grew downward into the medium
(Figure 3.3). When the inoculum was submerged in the liquid medium no rhizomorphs were produced until the mycelium had reached the air-liquid interface. If the fungus was prevented from reaching the surface they did not develop.
Interestingly, many rhizomorphs growing submerged in the agar medium in the
Petri-plate were seen to emerge out of the medium randomly on various days and at various points (Figure 3.4). Nevertheless, the majority of the rhizomorphs were found to remain submerged in the agar medium throughout the experimental procedure and thereafter (Figure 3.5). The emerged rhizomorphs subsequently became pigmented and the growth of these was
65 observed to slow down. This is demonstrated below where initiation and growth of these rhizomorphs occurred in agar-grown cultures.
The initiation of rhizomorphs in the agar-grown cultures began after about 6 d
(Figure 3.6). They began to grow immediately and those that remained submerged in the agar continued to extend for the duration of the experiment.
Rhizomorphs emerged from the agar at various times during the experiment and grew either on its surface or in the air above (Figures 3.4 and 3.6). At emergence these became strongly pigmented and their growth declined rapidly to values close to zero (Figure 3.7). Since it was not known at the onset of the experiment which rhizomorphs would emerge, all rhizomorphs selected were measured individually throughout the experiment. The data were then pooled into two batches separating rhizomorphs that remained submerged for the duration of the experiment from those that emerged at some stage, termed the
"aerial rhizomorphs". The data were averaged and are presented as cumulative growth in Figure 3.8 and as growth rates in Figure 3.9. Initially average growth rate progressively increased (Figures 3.8 & 3.9) to reach a maximum at about day 13 (Figure 3.9). During this phase of the experiment there was no significant difference between growth rates of rhizomorphs that were to remain submerged and those that would later emerge. The first rhizomorphs emerged at day 14. The progressive decline in average growth rate of aerial rhizomorphs shown in Figure 3.9 from day 14 to day 20 reflects the increasing number of rhizomorphs that became aerial during this time, rather than a slow decline in the growth rate of individual rhizomorphs. The t-test on the growth rates after
66 day 14 yielded a value of 1.95898, indicating a significant difference between the two growth patterns at confidence level of 95%.
3.3.2 External morphology
In addition to the differences in pigmentation observed, submerged and aerial rhizomorphs also showed differences in gross morphology. This is most obvious in fresh material (compare Figures 3.10 & 3.11), where a fringe of radial hyphae, embedded in a copious mucilaginous gel formed a sheath-like structure (Figure 3.11) around the submerged rhizomorphs. This sheath gradually increased in width away from the tip, measuring up to 0.5 mm in thickness around the more mature regions of the rhizomorph. In contrast, aerial rhizomorphs were not surrounded by such a thick mucilaginous sheath and appeared much smoother (Figure 3.10); the mucilage was so much reduced in volume that the outer cortical hyphae were frequently clearly visible in face view at the tip region (Figure 3.12). However, these differences were less obvious in scanning electron micrographs (compare Figures 3.13 & 3.14) where peripheral hyphae were obscured by mucilage that had dried down on to the rhizomorph surface, presumably during critical point drying. Nevertheless, tips of both aerial and submerged rhizomorphs were seen to be much smoother and had fewer peripheral hyphae.
Interestingly, two different types of exudate were seen associated with the aerial rhizomorphs. Clear, transparent, small droplets were seen on the surface of peripheral hyphae (Figure 3.15) and large droplets varying in colour from yellowish to dark brown were on the main body of the rhizomorph (Figure
67 3.16). Small transparent droplets similar to those observed on surface of peripheral hyphae were also seen on the surface of the radially growing mycelium (data not shown).
3.3.3 Internal structure
Despite differences in surface morphology both aerial and submerged rhizomorphs had a similar internal structure comprising four distinct regions; an outer zone of thin-walled peripheral hyphae, two cortical regions, and a central medulla of fine loosely woven hyphae that were connected with the inner cortex (Figures 3.17 & 3.18). Differences were observed within the regions when aerial and submerged rhizomorphs were compared. Also, just behind the tip as the rhizomorph increased in diameter, the medulla became mostly filled with intercellular space.
3.3.3.1 The outer most zone: Peripheral hyphae
The peripheral region in both submerged and aerial rhizomorphs comprised an outer zone of loosely arranged hyphae, which formed a distinct surface layer, varying in width from one to several hyphae (Figures 3.19 & 3.20). The hyphae in this layer were narrow (4-5 Pm in diameter) and thin walled. Hyphal branches emerged from the surface of this layer and grew perpendicular to the long axis of the rhizomorph to form the characteristic fringe noted in the literature (Figure 3.11). Submerged rhizomorphs developed more of these hyphal branches (Figures 3.18 & 3.20) than aerial rhizomorphs (Figures 3.17 &
3.19). In both submerged and aerial rhizomorphs, the number and length of
68 these radiating hyphae increased with distance from the apex. This is obvious in scanning electron micrographs where the extreme tip of rhizomorph appears smooth (Figures 3.13 & 3.14) but progressively becomes “hairy” with distance from the tip. As already mentioned dried mucilage obscured the surface features in scanning electron micrographs especially of submerged rhizomorphs.
All peripheral hyphae had well preserved cell contents that stained with toluidine blue indicating that they were living at the time of fixation (Figure
3.20). Most hyphae were vacuolate and some contained metachromatic granules. The surrounding mucilage stained pink to light purple with toluidine blue at pH 4.4, indicating that it carries a net negative charge and is probably rich in carboxyl groups. Septa and clamp connections were also seen (Figure
3.20). Calcofluor stained the walls of the peripheral hyphae intensely but the associated mucilage did not react (data not shown).
3.3.3.2 Outer cortex
Inside to peripheral hyphae was the cortical zone, which was further divided into two cortical regions – outer and inner cortex. The outer cortex was more distinct in aerial rhizomorphs due to the pigmentation of the cell walls and extracellular material to form a rind-like layer (Figure 3.17). This region was
3-4 hyphae thick and hyphae were generally of wider diameter (5-10 Pm) than those in the peripheral zone (Figures 3.17, 3.19 & 3.21). The outer cortex appeared cellular because of the frequent septa and the compact arrangement of hyphal segments, with few intercellular spaces (Figures 3.21-3.24). The cell
69 walls appeared thicker than those of either the peripheral or inner cortical hyphae. In submerged rhizomorphs the cell walls of the outer cortex were much less pigmented and less obviously different from those of the inner cortex, even though they still appeared somewhat thicker and stained more with toluidine blue (Figure 3.18). Outer cortical cells were highly vacuolate and their contents stained with toluidine blue (Figure 3.22). Mucilage was visible between the cortical cell walls of submerged rhizomorphs, identified by its appearance and staining reaction. In non-pigmented regions the mucilage stained pink, and in pigmented regions it stained dark blue with toluidine blue at pH 4.4, indicating incorporation of phenolics (O'Brien, Feder & McCully,
1964). Mucilage was often difficult to visualise between the cortical cell walls of aerial rhizomorphs perhaps due to heavy pigmentation, but it may be seen in
Figure 3.21. Outer cortical cell walls of aerial rhizomorphs did not stain well with Calcofluor (Figure 3.23), whereas those of submerged rhizomorphs stained brightly (Figure 3.24). In general, outer cortical cells were more elongated in aerial (Figures 3.21 & 3.23) than in submerged rhizomorphs where they were tightly packed and isodiametric (Figures 3.22, 3.24). It is important to note that hyphae in the outer cortical zone in both aerial and submerged rhizomorphs were compactly arranged with few or no intercellular spaces.
3.3.3.3 Inner cortex
Internal to the outer cortex was an inner cortex (Figures 3.22-3.28). This layer varied in thickness (3-5 hyphae) and contained elongated thin-walled hyphae of
70 varying width; hyphae measuring 20-30 Pm in diameter were commonly interspersed with narrow hyphae, approximately 5 Pm in diam. (Figures 3.25 &
3.27). Cells of the inner cortex were thin-walled, non pigmented and their walls stained light blue with toluidine blue (Figure 3.25). In both aerial and submerged rhizomorphs the inner cortical cell walls stained strongly with
Calcofluor (Figures 3.23 & 3.24). There were differences in the inner cortex of aerial and submerged rhizomorphs (compare Figures 3.26 & 3.27). In the aerial rhizomorphs the inner cortex appeared to be tightly cemented with less intercellular space and no cellular detail was apparent (Figures 3.25 & 3.26).
Also, the hyphae were not swollen. There was a distinct boundary between the inner face and the medullary space. In submerged rhizomorphs, however, there was more intercellular space in the inner cortex (Figure 3.24) and the boundary with the medullary space was less sharp; this can be clearly seen in scanning electron micrographs showing the medullary/inner cortex boundary in face view (Figure 3.27). The majority of the cells were swollen, appearing large and ovoid, and there were fine (narrow) hyphae interspersed among them, some clearly arising from the swollen hyphae (Figure 3.28). In both aerial and submerged rhizomorphs fine hyphae spread out from the inner cortex into the medullary space and formed a loosely interwoven weft in the central medulla
(Figures 3.26 & 3.27).
3.3.3.4 Medulla
Hyphae in the medullary region of both aerial and submerged rhizomorphs were loosely arranged, with a large amount of intervening intercellular space
71 (Figures 3.17, 3.18). These hyphae were oriented in all directions and were of similar appearance to the peripheral hyphae. In both aerial and submerged rhizomorphs the number of hyphae forming the weft varied relative to medullary space. In general, the medullary region near the tip is relatively empty (Figure 3.29) in contrast to further away from the tip, where it is filled with fine hyphae interspersed in all directions with lots of intercellular spaces
(Figure 3.31). The walls of these medullary hyphae stained a pale blue with toluidine blue and they were also Calcofluor positive. All medullary hyphae showed well-preserved cell contents indicating that they were living at the time of fixation (data not shown).
No enlarged, elongate, empty hyphae with dissolved septa that could be equated with ‘vessel’ hyphae were found in any cell layer in either aerial or submerged rhizomorphs.
3.4 Discussion
There were several differences in the structure of submerged and aerial rhizomorphs of A. luteobubalina. For example, there was less pigmentation in the outer cortical cell walls, better development of peripheral hyphae and mucilage, differences in cell shape and more intercellular spaces within the inner cortex in submerged rhizomorphs. Many of these differences have also been found in A. mellea (Granlund et al., 1984) and they indicate that there is considerable plasticity superimposed on a basic structure according to differences in external conditions, for example moisture levels and oxygen availability.
72 Several of the observations support the earlier literature that A. luteobubalina rhizomorphs might be organs of aeration. These are the failure of rhizomorphs to form from mycelium that is not in contact with air, the presence of the gas- filled cavity throughout the length of the rhizomorph and the faster growth of submerged rhizomorphs. The most likely role for this central cavity is to facilitate oxygenation of any regions of submerged rhizomorphs subjected to low O2 levels, as proposed by Smith & Griffin (1971). In addition, our growth data clearly indicate the preference of A. luteobubalina rhizomorphs to grow submerged, not aerially. While it must be remembered that our rhizomorphs are axenically grown, this is nevertheless a characteristic that would not be expected if these rhizomorphs were to play a significant role in exploration across open soil, either to initiate new infections or to translocate nutrients.
This agrees with the field observations that rhizomorphs of A. luteobubalina are not frequently found in open soil surrounding centres of infection in many forest sites, and those rhizomorphs that are present are relatively short (Podger et al., 1978; Pearce et al., 1986; Shearer & Tippett, 1988).
There was no structural evidence for specialised translocatory hyphae with the appearance of xylem vessels or the ‘vessel hyphae’ reported in field grown basidiomycete rhizomorphs (Eamus et al., 1985; Cairney et al., 1988b). In A. luteobubalina the central cavity was occupied by wefts of loosely arranged fine hyphae which developed to varying extents under different conditions. Hyphae, including those that were oriented more or less longitudinally within this central space, were narrow, septate, thin walled, and filled with cytoplasm.
Similarly, there was no evidence of loss of contents or dissolution of septal
73 walls that would indicate development of vessel hyphae in the cortex as
Wolkinger, Planck & Bruneggar (1975) had formerly suggested for A. mellea.
The investigations reported here are based on young rhizomorphs examined a few cm from the growing apex. Older regions, several cm from the tip in field- collected A. mellea rhizomorphs are reported to have a densely packed medulla with large diameter empty vessel hyphae (Cairney et al., 1988b). Field-grown rhizomorphs of A luteobubalina have not been examined in this study.
Nevertheless, 10-14 wk old rhizomorphs of A. luteobubalina growing from inoculated autoclaved wood blocks in jars containing moist sterile sand resembled the aerial rhizomorphs seen on agar plates. Both their morphology and anatomy was similar, their being a pigmented cortex and large central medulla which was mostly intercellular space, again containing wefts of fine hyphae but lacking any vessel-like hyphae in any regions of the rhizomorph
(D. Stevenson, pers. comm.).
The tissue at the interface between the inner cortex and the central medulla containing inflated hyphae also did not have characteristics of specialised translocatory tissue. In A luteobubalina it was well-developed in submerged rhizomorphs but not in aerial rhizomorphs. Similar tissue has been described in
A. mellea (Granlund et al., 1984). Therefore, it may be suggested that, with its large surface area and low volume relative to intercellular space, this tissue could be the fungal equivalent to higher plant aerenchyma. If so, we would expect its development to vary according to the degree of anaerobiosis, which it did.
74 Only aerial rhizomorphs were strongly pigmented. In this respect they show many parallels with other structures such as sclerotia and the sheath of some ectomycorrhizae. Onset of pigmentation in both these cases has been correlated with reduced permeability to fluorochromes (Ashford et al., 1988; Young &
Ashford, 1992; 1995 and 1996). This is essential to prevent excessive loss of water by evaporation and promote longevity, but it has the potential to reduce the capacity for uptake from the environment (Young & Ashford, 1995; Vesk et al., 2000). This may also apply to pigmentation in emergent rhizomorphs, a point recognised by Morrison (1975). In contrast, submerged rhizomorphs with their extensive development of peripheral hyphae surrounded by mucilage have an increased surface area and do not appear to be waterproofed. Their outer hyphae contain cytoplasm, are thin walled and have the characteristics of absorbing structures. The surface mucilage may also facilitate this role by forming bridges to maintain contact with the substrate under varying conditions of moisture, as proposed for root cap mucilage (Guinel & McCully, 1986;
McCully & Boyer, 1997; Read, Gregory & Bell, 1999). However, much more information about its composition and properties are required to substantiate this.
The A. luteobubalina rhizomorph appeared to be a well-aerated tube strengthened around the perimeter. The overall structure and preference for submerged growth suggests that rhizomorphs of A. luteobubalina are specialised for fungal spread and substrate exploitation in moist potentially anaerobic conditions, rather than for water and nutrient transportation. This agrees with the field observations that A. luteobubalina rhizomorphs are scarce
75 in many forest soils (Pearce & Malajczuk, 1990a; Shearer et al., 1997a). This is not to say that these rhizomorphs do not carry out longitudinal transport, indeed for A. mellea there is much evidence that they do (see Jennings, 1987;
Cairney, 1992), but rather that there are no obvious specialised translocatory hyphae in A. luteobubalina, at least under our conditions.
Small droplets observed on peripheral hyphae of rhizomorphs in other
Armillaria species have been viewed as an expression of mass flow of water through the mycelium (Brownlee & Jennings, 1981) as in the cords of Serpula lacrymans, where droplets are also associated with the mass flow (Jennings,
1987). The possibility that droplets to serve as a reservoir of water for the hyphae to allow them to continue growth, when subjected to unfavourable water potential, as suggested by Jennings (1991). Unestam & Sun (1995) made a similar suggestion that drop exudate represents a kidney like function and bridge for drier particles to initiate nutrient uptake by the hyphae.
Recent findings with ectomycorrhizal fungi have emphasised the plasticity of soil mycelium, and of particular interest is the stimulation of branching to produce specific absorbing regions when mycelium contacts nutrient- containing patches (Bending & Read, 1995). Rhizomorphs similarly appear to exhibit a plasticity in amplifying their absorptive surface of peripheral hyphae when growing through moist nutrient-rich substrates.
Dark brown exudates on the main body of the rhizomorph have been shown to occur in A. mellea and have been proposed to have a possible role in pathogenesis (Mallett & Colotelo, 1984). However, this is not conclusive.
76 Further investigations are required to find the role of these droplets and exudates found associated with the rhizomorph.
77 3.5 Legends
Figures 3.13.3. Appearance of rhizomorphs under different culture conditions. Figure 3.1. Rhizomorphs (r) grow radially from the white fluffy mycelium that develops around the inoculum plug on agar in Petri plates. 9 days after inoculation. Bar = 10 mm. Figure 3.2. Rhizomorphs develop and grow through an air gap where the cellophane is ruckled (arrows) when inoculum is placed between two layers of cellophane; 21 days after inoculation. Note the extensive growth of pigmented mycelium. Bar = 10 mm. Figure 3.3. In liquid culture a fluffy mycelium with many aerial hyphae (a) first grows on the surface followed by rhizomorphs (r) that grow down in to the medium. 21 days after inoculation. Bar = 2.5 mm.
Figures 3.43.5. Appearance of aerial and submerged rhizomorphs in the same agar-grown culture. Figure 3.4. Aerial rhizomorphs emerging at various points (see arrow). Note the random appearance. Figure 3.5. Culture viewed from underside showing submerged rhizomorphs growing in the agar. Bars = 10 mm.
78 79 Figure 3.6. Histogram showing change in the ratio of aerial and submerged rhizomorphs as more rhizomorphs emerge from the agar during the experimental period in agar plate cultures.
Figure 3.7. Cumulative growth of the 5 individual rhizomorphs to demonstrate the rapid decline in growth as each rhizomorph emerges. Rhizomorphs 1 & 2 remained submerged while 3, 4 & 5 emerged. Arrows indicate the point at which rhizomorphs became aerial.
80 Figure 3.6
Figure 3.7
81 Figure 3.8. Average cumulative growth of rhizomorphs that remain submerged in the agar medium throughout the experiment compared with those that emerged at some stage. Measurements were made every 2 days on the same set of rhizomorphs. Values are means ± standard errors. Arrow indicates the point at which the first rhizomorph became aerial. Cultures were grown in Petri plates in the dark at 25 oC.
Figure 3.9. Change in growth rate of rhizomorphs expressed as an average for the 48 h previous to each measurement point over the experimental period from the data in Figure 3.7. Values are means ± standard errors.
82 Figure 3.8 50 Aerial Submerged 40
30
20
10
0 0 2 4 6 8 10 12 14 16 18 20 22 Days after inoculation
Figure 3.9
10 Submerged Aerial 8
6
4
2
0 0 2 4 6 8 10 12 14 16 18 20 22 Days after inoculation
83 Figures 3.103.14. Comparison of surface features of intact aerial (Figures 3.10, 3.12, 3.13) and submerged (Figures 3.11, 3.14) rhizomorphs to show differences in the extent of surface hyphae and mucilage; light micrographs (Figures 3.10-3.12) and scanning electron micrographs (Figures 3.13-3.14) Figure 3.10. There is a relatively narrow zone of fringing hyphae (arrow) which extends to the tip, with little mucilage in aerial rhizomorphs. Focus in median plane. Bar = 250Pm. Figure 3.11. Submerged rhizomorphs show a dense zone of peripheral hyphae surrounded by copious mucilage (arrow) and no pigmentation. Figure 3.12. There is so little mucilage in aerial rhizomorphs that cells of outer cortex with their brown pigmented walls are visible in surface view. Bar = 5 Pm. Figures 3.13-3.14. Peripheral hyphae (p) develop a short distance from the rhizomorph tip of both aerial (Figure 3.13) and submerged (Figure 3.14) rhizomorphs such that there is a smooth sub-apical zone (arrow). Peripheral hyphae (p) are partially obscured by dried mucilage in the submerged rhizomorphs. Bars = 500 Pm.
Figures 3.153.16. Droplets of exudate associated with the rhizomorph. Figure 3.15 Tiny transparent droplets viewed on peripheral hyphae of aerial rhizomorph (arrow). Note the random arrangement and the small size of the droplet. Figure 3.16. Exudates observed on a pigmented rhizomorph. Note the variation in the colour and size of the droplets. Bars = 500 Pm.
Figures 3.173.18. Longitudinal sections of aerial (Figure 3.17) and submerged (Figure 3.18) rhizomorphs stained with toluidine blue (pH 4.4); light micrographs. Figure 3.17. There are only a few peripheral hyphae (p). The outer cortex (o) is pigmented and contrasts with the non-pigmented inner cortex (i) of longitudinally aligned thin-walled hyphae. The central medullary (m) region comprises mostly gas space containing a few fine hyphae of narrow diameter. Figure 3.18. Peripheral hyphae (p) are abundant. The outer cortex (o) consists of isodiametric cells with little intercellular space and no pigmentation. This grades into an inner inner cortex (i) of similar hyphae. There is a central medulla (m) mostly of intercellular space but containing narrow diameter hyphae. Bars = 20 Pm.
84 85 Figures 3.193.20. Detail of the peripheral zone in of aerial (Figure 3.19) and submerged rhizomorphs (Figure 3.20) showing the differences in number of peripheral hyphae. Figure 3.19. Peripheral hyphae (p) oriented parallel to the rhizomorph long axis are septate and branch (arrow) to give rise to the few radiating hyphae. Figure 3.20. The numerous radiating hyphae in submerged rhizomorphs arise by branching of the longitudinal peripheral hyphae (p) overlying the cortex. Radiating hyphae have abundant cytoplasm and are septate, with clamp connections (arrows). Bars = 5 Pm.
Figures 3.213.25. Longitudinal sections of aerial (Figures 3.21, 3.23, 3.25) and submerged rhizomorphs (Figures 3.22, 3.24) stained with Toluidine blue pH 4.4 (Figures 3.21-3.22, 3.25) or Calcofluor White (Figures 3.23-3.24) showing differences in the cortex; bright field microscopy (Figures 3.21-3.22, 3.25); epifluorescence (Figures 3.23-3.24). Figure 3.21. Outer cortical cells (o) showing pigmentation in walls and extracellular material. Note transition to peripheral zone (p) with thin walled non-pigmented hyphae. Figure 3.22. Outer cortical cells (o) are tightly packed and not pigmented in submerged rhizomorphs. They are much more similar to the peripheral hyphae (p). Vacuoles (v) are apparent, surrounded by cytoplasm. Bars = 5 Pm. Figure 3.23. In aerial rhizomorphs walls peripheral hyphae (p) and inner cortex (i) stain with Calcofluor white, those of the outer cortex (o) do not. Figure 3.24. The outer cortex (o) and inner cortex (i) both stain with Calcofluor white in submerged rhizomorphs. Note the large intercellular space (asterisks) in the inner cortex. Bars = 10 Pm. Figure 3.25. The elongated hyphae of the inner cortex (i) are more tightly packed. A few narrow hyphae (asterisk) are interspersed amongst hyphae of wider diameter. Bar = 20 Pm.
Figure 3.26. The median longitudinal section showing the internal structure of an aerial rhizomorph, as seen by SEM. Note the tightly packed elongated inner cortical cells (i) with a sheet-like inner surface and the narrow-diameter hyphae surrounded by a large amount of intercellular space throughout the medulla (m). Bar = 10 Pm.
86 87 Figures 3.273.31. The internal structure of submerged rhizomorphs (Figures 3.27, 3.28, 3.29 & 3.31) and aerial rhizomorphs (Figure 3.30) as seen using SEM. Note: outer cortex (o), peripheral hyphae (p). Figure 3.27. Tangential LS showing the inner cortex (i) in face view. In submerged rhizomorphs this contains fine hyphae interspersed between ovoid swollen cells with large intercellular spaces. Bar = 50 Pm. Figure 3.28. Detail of ovoid swollen cells (arrow) in the inner cortex region and their connections fine hyphae (arrowhead). Bar = 10 Pm. Figure 3.29. Transverse section of submerged rhizomorph (in liquid medium) near the tip which, being distant from the source of inoculation, was flattened. Note the medullary region (m), empty with few or no hyphae. Figure 3.30. Transverse section of the aerial rhizomorph 5-7 mm away from the tip showing medullary region (m) containing fine hyphae interspersed with many intercellular air spaces. Figure 3.31. Transverse section of submerged (in agar medium) rhizomorph 3-4 cm away from the tip showing medullary region (m) filled with hyphae. Bars = 100 Pm.
88 89 Chapter 4. Permeability of Aerial and Submerged Rhizomorphs to
Apoplastic and Symplastic Tracers
4.1 Introduction
Transport of nutrients in plant tissues has been the subject of many investigations. The uptake and movement of apoplastic and symplastic fluorescent tracers have been used successfully to investigate permeability in many plant and fungal tissues (see for example Canny, 1988; Fitzgerald &
Allaway, 1991; Barnabas, 1994; Young & Ashford, 1992; 1995; 1996; Vesk et al., 2000). Abundant information is available on root permeability from these techniques in flowering plants (see for example Canny, 1986; Canny &
McCully, 1986; Moon et al., 1986; McCully & Canny, 1989; Moreshet, Huang
& Huck, 1991; Varney, McCully & Canny, 1993; Barnabas, 1994;
Zimmermann & Steudle, 1998) but they have been rarely used with fungal tissue.
Data suggest that rhizomorphs in Armillaria species play a role in nutrient uptake and are more abundant where nutrient levels are relatively high. For example, it has been found that the production and growth of rhizomorphs
(Garrett, 1956; Redfern, 1973; Rykowski, 1984) is lower in pure sand than in organic matter like peat (Redfern, 1973; Rykowski, 1984, or pine bark compost
(Rykowski, 1984) which encourages rhizomorph growth and branching
(Redfern, 1973). Rhizomorphs are reported to be more abundant where there is a humus layer than in the mineral soil below (Singh, 1981). There is also
90 evidence that A. luteobubalina rhizomorphs increase in number with the incorporation of organic matter in soil (Pearce & Malajczuk, 1990a) .
In a previous study, radioactive nutrients supplied to rhizomorphs of Armillaria ostoyae were absorbed but not translocated (Morrison, 1976), and this observation led Morrison to hypothesize that rhizomorphs absorb and utilise nutrients from the soil. To confirm this, Morrison tested the influence of nutrient solution, soil humus, on rhizomorph growth in a sandy soil; all stimulated growth and branching, supporting his view that rhizomorphs absorb and utilise nutrients from the soil (Morrison, 1982). Clipson, Cairney &
Jennings (1987) in an examination of phosphate uptake reached the conclusion that cords of species other than Armillaria spp. also absorb nutrients directly from the soil and are not fully reliant on the import of nutrients from the food base. Cairney et al. (1988a) also found phosphate to be immobilised in the excised Armillaria mellea segments which had taken it up, by its accumulation into vacuoles as polyphosphate.
There are many reports that long distance translocation of nutrients and water takes place in rhizomorphs from the tip to base or vice versa (Anderson &
Ullrich, 1982; Granlund, Jennings & Thompson, 1985; Gray, Dighton &
Jennings, 1996) in an analogous manner to xylem and or phloem transport in vascular plants (Jennings, 1976; Jennings, 1987; Cairney, 1992). However, underlying mechanisms of translocation are not yet clear and several alternative mechanisms that may function simultaneously are currently under discussion. Moreover, there are also uncertainties regarding the actual pathway
91 involved in the long distance translocation (Cairney, 2005). There is increasing notion that peripheral hyphae which protrude from the surface of rhizomorphs might be involved in the uptake of nutrients or water (Hartig, 1874; Granlund,
Jennings & Veltkamp, 1984; Clipson et al., 1987; Jennings, 1987; Cairney,
Jennings & Veltkamp, 1988b). The view that peripheral hyphae may be absorptive is based on structural evidence. These hyphae have been found to be transversely connected to the other tissues of the rhizomorph when the majority of hyphal elements are parallel with the direction of growth of the rhizomorph; rhizomorphs have a system of connecting hyphae which arise from the sub- cortex and travel outwards to form the fringe of peripheral hyphae and inwards to connect with the medullary hyphae (Granlund, Jennings & Veltkamp, 1984;
Cairney et al., 1988b). That the cortex layer contains short hyphae extending into the soil had already been observed by Hartig (1874) and De Bary (1887).
Hartig (1874) further showed these hyphae to grow in the interior of the host cells, boring through the cell walls of the host plant, passing from one cell to another, and withdrawing nutrients from the wall. These hyphae usually penetrate through the starch-bearing medullary ray cells of the wood (Hartig,
1874). Cairney et al. (1988) have shown simple pores in the thickened rind cell walls of mature rhizomorphs and suggest that these may represent points of interconnection with neighbouring hyphae. They suggest that these pores facilitate transfer of nutrients from cell to cell, in an interconnecting system where it is anticipated that lateral movement of solutes would be a common feature, presumably by symplastic flow. However, this information does not confirm that peripheral hyphae are absorptive or that rhizomorphs are
92 permeable across their radius. Therefore, there is a need to obtain information on the radial permeability of rhizomorphs. This can be done by using both apoplastic and symplastic tracers, as shown by the work on roots of flowering plants and fungal sclerotia cited above.
There are two forms of rhizomorphs found in the field or in vitro. Rhizomorphs in the field situation were originally distinguished as (i) subterranea and (ii) sub-corticalis (Hartig, 1874). ‘Subterranea’ rhizomorphs in A. mellea are described as, “hard, brown or black and may be cylindrical or flattened”, while
‘sub-corticalis’ type are “softer, white, flattened strands formed in wood”
(Townsend, 1954). However, the same two kinds can be produced in culture in the absence of plants and soil (Townsend, 1954; Snider, 1959) and for this reason Snider (1959) called them ‘round’ and ‘flat’ rhizomorphs. The two types may be produced by changing nutrients in the medium (Townsend,
1954). In A. luteobubalina grown on malt agar the corresponding entities are (i) aerial rhizomorphs which are pigmented and (ii) submerged rhizomorphs which are less or non-pigmented and more flattened.
It has been demonstrated in the sclerotia of Sclerotinia minor (Young &
Ashford, 1992; 1996), Sclerotium rolfsii and Sclerotium cepivorum (Young &
Ashford, 1995) that deposition of brown pigments in walls and extracellular matrix of the rind acts as a barrier to the apoplastic tracer Sulphorhodamine G.
Young & Ashford (1992) showed that, as sclerotia matured and their outer surface became differentiated into a melanised rind, they became impermeable to this tracer. If, however, they were cut to remove the rind, the underlying
93 areas were found to be permeable and the structure once again became impermeable as the rind regenerated (Young & Ashford, 1996). Reduction in permeability appeared to be correlated with deposition of brown pigment.
Armillaria rhizomorphs have a consolidated cortex like that in many sclerotia and they may also develop a rind. It would be interesting to determine whether the pigmented cortical zone of aerial rhizomorphs similarly acts as a barrier to apoplastic fluorescent probes, and whether the peripheral hyphae which appear to be connected to inner cortical cells, may circumvent this barrier by uptake and symplastic transport. In order to obtain insights into pathways available for uptake or transport of water or nutrients, and determine any barriers which might impede tracer movement, the apoplastic fluorescent probe, 8- hydroxypyrene-1,3,6-trisulphonate (HPTS), and symplastic fluorescent tracers
Oregon Green 488 carboxylic acid diacetate (carboxy-DFFDA) and 7-amino-4- chloromethylcoumarin (CMAC) were applied to the rhizomorphs of Armillaria luteobubalina. Techniques of anhydrous freeze substitution were used
(Orlovich & Ashford, 1995), with the apoplastic probe to minimise artefacts due to diffusion of the probe during tissue processing and observation.
4.2 Materials and methods
4.2.1 Fungal culture
A. luteobubalina Kile & Watl. from the culture collection of the University of
Western Sydney was used in all tracing experiments. Stock cultures were maintained as described in Chapter 3.1. Plugs of inoculum (10 mm diam.)
94 were cut from the edge of actively growing colonies and transferred to either
2% MM agar in Petri dishes, or liquid MM medium (without agar) in the test tubes (150 mm × 12 mm internal diam.), to produce aerial or submerged rhizomorphs, respectively. Submerged rhizomorphs produced in liquid medium were used because in those produced in Petri dishes the agar was impossible to remove from the rhizomorph surface. All media were autoclaved for 15 min at
1.5 MPa. Both types of culture were incubated in the dark at 23 oC. Liquid culture tubes were kept upright.
4.2.2 Treatment with symplastic tracers
Aerial and submerged rhizomorphs were excised from the cultures in Petri-dish and test-tubes respectively. They were rinsed thoroughly with reverse osmosis
(RO) water and incubated, either intact or as hand-cut sections, for 0.53.0 h in a solution containing either 0.08 Pg/ml carboxy-DFFDA (Molecular Probes,
Eugene, OR. U.S.A.) or 0.16 Pg/ml CMAC (Molecular Probes, Eugene, OR.
U.S.A.). Specimens were then rinsed again with RO water for 10-30 mins.
Hand-cut sections of rhizomorphs were mounted directly on a slide in aqueous medium, while intact rhizomorphs were first hand-sectioned, and then mounted in a similar manner and examined immediately under UV excitation. All the procedures - the treatment, rinsing, and epifluorescence microscopy- were carried out at room temperature and in a dark room, with dim illumination only when necessary. For each investigation more than 5 replicates were used.
95 4.2.3 Treatment with apoplastic tracer
Aerial and submerged rhizomorphs were treated with 0.1% w/v 8- hydroxypyrene-1,3,6-trisulphonate (HPTS) (Molecular Probes, Eugene, OR.
U.S.A.) in RO water. For treatment of aerial rhizomorphs, intact Petri-dish cultures on agar were flooded with HPTS solution. For treatment of submerged rhizomorphs, whole test-tube cultures on liquid medium were gently transferred to plastic Petri-dishes with extreme care to avoid damage. The liquid medium was pipetted out and the culture was rinsed with RO water and were immediately flooded with the HPTS solution. Controls of both aerial and submerged rhizomorphs were kept in RO water instead of tracer. All procedures were done at room temperature, and specimens were kept in dark throughout by covering the dishes with aluminium foil.
To determine the appropriate length of treatment, preliminary experiments were carried out by treating rhizomorphs with HPTS solution for 5, 10, 15, 30,
60, 90 and 180 min. Hand-cut sections of these rhizomorphs were mounted in immersion oil and immediately examined with the epifluorescence microscope.
From these preliminary experiments it appeared that 90 min was long enough for the probe to diffuse across the entire rhizomorph. Therefore, for all subsequent experiments a 90 min treatment period was used.
Apart from these preliminary experiments, all observations on HPTS-treated rhizomorphs and their controls were made on specimens prepared by anhydrous freeze-substitution. For each investigation more than 5 replicates were used.
96 4.2.4 Anhydrous freeze-substitution
After 90 min HPTS treatment, the tracer solution was pipetted out.
Rhizomorph segments were rapidly cut from a specified part of each rhizomorph and immediately frozen by plunging into liquid propane [High grade (99.5%), Linde Gas Pty Ltd., NSW] that had been pre-cooled with liquid nitrogen to –187 oC. Five replicates each of aerial and submerged rhizomorphs were tested. Rhizomorph segments 5-10 mm in length were taken from three different positions: the tip, 5 cm away from the tip, and the basal region 7-10 cm proximal to the tip (or very close to the inoculum plug if the rhizomorph was <10 cm long). For each segment a separate rhizomorph was used, to avoid any re-distribution of tracer that might occur after cutting the rhizomorph.
After freezing, specimens were substituted in anhydrous tetrahydrofuran
(previously dried over molecular sieve) for 3 days at –80 oC (Vesk et al.,
2000), and then gradually brought to room temperature in a Leica AFS freeze substitution apparatus as follows. Samples were warmed gradually to –30 oC, kept at –30 oC for 24 h, warmed to 6 oC and kept at 6 oC for 1 h. They were then transferred to a sealed perspex box flushed with dry nitrogen gas, and infiltrated with a graded series of anhydrous Spurr’s resin (see Canny &
McCully, 1986) in tetrahydrofuran (30%, 60%, 90% & 100% v/v for one day each) at 23 oC. Samples at all stages were kept in the dark by covering with aluminium foil.
Samples were then either placed in flat embedding rubber moulds (Proscitech,
Australia) or flat embedded between two frosted end glass slides (Superfrost,
97 manufacturer Menzel-Glaser), previously sprayed with polytetrafluoroethylene
(The Chemilube Company, Victoria, Australia), to make the surface non- sticky. Samples were polymerised at 60 oC for 48 h. Polymerised samples were always kept in a desiccator over dry molecular sieve and covered by aluminium foil, until sectioning. Samples were sectioned dry at 2 μm using a dry glass knife, collected dry, and flattened dry on electrostatically charged Star Frost adhesive microscope slides (Proscitech, Australia) using a pencil eraser and cellophane. Sections were stored dry, or mounted in oil that had been dried overnight on 5Å molecular sieve (Ajax), in a desiccator prior to the fluorescence microscopy.
4.2.5 Epifluorescence microscopy
All specimens were examined with a Zeiss Axiophot microscope equipped with bright field, Nomarski differential interference contrast (DIC) optics, and epifluorescence optics. Blue excitation (BP 450-490, FT 510 and LP 515-585) was used for carboxy-DFFDA, and UV excitation with the filter set BP 365,
DIC 395, and LP 420 for CMAC and HPTS. Individual images were captured with an image point cooled CCD video camera (Photometrics, Tucson, AZ,
U.S.A.) or Zeiss Axiocam digital camera and processed using Adobe
Photoshop 6.0 software. Images were arranged with Adobe Illustrator version
10 and printed with a Hewlett Packard business inkjet 1100 printer to produce the plates.
98 4.3 Results
With the symplastic tracers, results for aerial and submerged rhizomorphs were similar, so the types of rhizomorph are not distinguished in the subsections below. With the apoplastic tracer, however, aerial and submerged rhizomorphs behaved differently and therefore they are described separately.
4.3.1 Symplastic Permeability
4.3.1.1 Intact rhizomorphs treated with carboxy-DFFDA
In both aerial and submerged rhizomorphs treated intact with carboxy-DFFDA and subsequently sectioned, strong fluorescence was observed in most of the peripheral hyphae (Figure 4.1). The fluorescent material (presumably the hydrolysis product of carboxy-DFFDA) accumulated in the vacuoles of the peripheral hyphae (Figure 4.2) and the vacuoles showed motility. A DIC image of the same hypha taken later is shown in Figure 4.3; all the vacuoles in focus are fluorescent. No fluorescence was seen in any part of the cortical cells of either aerial or submerged rhizomorphs but a diffuse fluorescence was observed in the medullary region adjacent to the inner cortical zone (Figure
4.4).
4.3.1.2 Rhizomorph sections treated with carboxy-DFFDA
In contrast to intact rhizomorphs, in sections of the aerial and submerged rhizomorphs which were first cut and then immersed in carboxy-DFFDA, the peripheral hyphae and many inner cortical cells showed strong fluorescence
99 (Figure 4.5). (A DIC image of the same section is shown in Figure 4.6 for comparison; note lack of fluorescence in out of focus peripheral hyphae). The labeling was observed in the vacuoles of the peripheral hyphae and vacuoles were motile, as when intact rhizomorphs are treated. There was also strong fluorescence in the majority of the inner cortical cells. In this case fluorescence occupied the full cell profile; in a few cells labeling was found in sub cellular structures that were more obviously vacuoles (Figure 4.7). The majority of the outer cortical cells did not show any fluorescence, with the exception of a few patches of cells (Figure 4.5), which showed fluorescence in the whole cell profile. Sometimes these appeared to form a continuum between peripheral hyphae and inner cortical cells. Many medullary hyphae fluoresced strongly, showing labeling in the vacuoles as in peripheral hyphae (Figure 4.8). A DIC image (Figure 4.9) for comparison. There was no difference in the permeability of aerial and submerged rhizomorphs regardless of the pigmentation present in the aerial rhizomorphs, and the localization of carboxy-DFFDA was very similar in both types of rhizomorphs.
4.3.1.3 Intact rhizomorphs treated with CMAC
When intact rhizomorphs were treated with CMAC and subsequently sectioned, the labeling was again seen to have accumulated in the vacuoles of many peripheral hyphae (Figure 4.10). A DIC image of the same hypha as
Figure 4.10 is shown in Figure 4.11 for comparison. The vacuoles showed motility. Bright fluorescence was also observed in most of the inner cortical cells but absent from most of the outer cortex except for a few cells in patches
100 that sometimes formed a bridge between the inner cortical zone and peripheral hyphae (Figure 4.12). The fluorescence appeared across the whole cell profile, rather than in the cytoplasm or what were obviously vacuoles. The corresponding DIC image is shown in Figure 4.13 for comparison. Again, in a few cells, labeling was also clearly shown to be sequestered in the vacuoles
(Figure 4.14). Medullary hyphae also fluoresced brightly and showed labeling in the vacuoles (not shown). Both aerial and submerged rhizomorphs showed similar patterns of fluorescence with CMAC regardless of the pigmentation.
4.3.1.4 Rhizomorph sections treated with CMAC
In contrast, in sections of the aerial and submerged rhizomorphs cut and then immersed in CMAC, the peripheral hyphae and most of the cortical cells showed strong fluorescence (Figure 4.15). (A DIC image of the same section is shown in Figure 4.16 for comparison). Labeling was observed in the vacuoles of the peripheral hyphae (Compare Figure 4.17 with 4.18), a situation comparable to that when intact rhizomorphs are treated. Most outer cortical cells showed strong fluorescence. Again fluorescence mostly occupied the full cell profile but in a few of the inner cortical cells labeling was found in sub cellular structures that were more obviously vacuoles (Figure 4.15).
4.3.2 Tracer localization at the extreme apex of hand-cut sections or intact
rhizomorphs treated with Carboxy-DFFDA or CMAC
With both carboxy-DFFDA and CMAC, fluorescent labeling was seen in cells throughout the inner cortex at the extreme apex of the rhizomorph. Results for
101 hand-cut sections treated with carboxy-DFFDA are shown in Figure 4.19 with
DIC image (Figure 4.20) for comparison. The results were similar in intact rhizomorphs treated with CMAC and then sectioned. Figure 4.21 shows result with CMAC, and a DIC image (Figure 4.22) for comparison. In this case even more cells were labeled. These results contrast with more basal regions, where fewer inner cortical cells were fluorescently labeled (compare Figures 4.19 and
4.21 with Figures 4.5 and 4.12).
4.3.3 Apoplastic Permeability
The apoplastic permeability of aerial and submerged rhizomorphs was compared in the region near the tip. Controls, from a similar region to that viewed in HPTS-treated rhizomorphs, examined under UV excitation, using similar filter combinations and photographic exposures to those used for HPTS treatment, showed no auto-fluorescence (compare Figures 4.23 and 4.24). With long photographic exposures all the cell walls of peripheral, cortical and medullary hyphae of the rhizomorph appeared yellowish-white (data not shown). In contrast, HPTS-treated rhizomorphs showed the bright blue fluorescence characteristic of this probe in some cell walls of both aerial and submerged rhizomorphs. There were differences in distribution of the probe in aerial and submerged rhizomorphs, and within submerged rhizomorphs depending on whether they were pigmented or not.
102 4.3.3.1 Intact aerial rhizomorphs treated with HPTS
Aerial rhizomorphs sectioned near the tip showed blue fluorescence in the peripheral hyphae but fluorescence was not seen throughout the cortex or in any part of the medulla (Figures 4.25 and 4.26). The bright blue fluorescence characteristic of HPTS was located in the cell walls of the peripheral hyphae, including the septa, while the protoplast did not fluoresce (Figures 4.27 and
4.28). Mucilage around the peripheral hyphae also did not fluoresce. To confirm low fluorescence in regions inside the pigmented cortex another section with a well pigmented cortex was viewed at high magnification and photographed at long exposures, but no HPTS fluorescence was seen in any part of the cortex or medulla and peripheral hyphae also showed no fluorescence (Figure 4.29). The same area under DIC illumination shows pigmentation in extra cellular matrix and cell walls of the outer cortical zone
(Figure 4.30). In contrast to those shown in Figures 25 and 27 the peripheral hyphae are oriented parallel to the long axis of the rhizomorph and are appressed to its surface; there are no sectioned free radiating hyphae shown in this section.
4.3.3.2 Intact submerged rhizomorphs treated with HPTS
Submerged rhizomorphs sectioned in the same region (i.e. at the tip) showed blue fluorescence in the cell walls of free peripheral hyphae and at the surface of the rhizomorph. HPTS fluorescence was also present throughout the cortex and medulla, specifically located in the apoplast (compare Figure 4.31 with
4.32). The outer cortical cell walls were fluorescent but the intensity was much
103 lower and more obviously variable with occasional bright patches (Figure
4.33). The tracer was not only located in the cell walls but was also in the extra cellular matrix (Figure 4.33). A clear picture of this using DIC illumination reveals the bright patches in the intercellular spaces at the cell corners
(compare Figures 4.33 and 4.34). In contrast to this, inner cortical cells showed bright HPTS fluorescence neatly restricted to the apoplast or the cell wall, with no diffuse fluorescence in any intercellular spaces, as was seen in the outer cortical region (Figure 4.35 and 4.36).
To obtain insights into the permeability of submerged rhizomorphs in their entirety, HPTS permeation at different distances (i.e. 5 cm and 7-10 cm away from the tip) was also examined. At 5 cm distant from the tip, all the tissue zones of the rhizomorph were similar to the apical region (not shown).
Rhizomorph samples from the basal region (i.e. 7-10 cm from the tip) became fragmented during processing and were difficult to orient for sectioning.
Information gathered from the fragmented pieces indicated that results are similar to those in aerial rhizomorphs but not completely. Most of the peripheral hyphae fluoresced brightly and the probe was restricted to the cell walls including the septa (Figures 4.37 and 4.38). A few peripheral hyphae also showed green fluorescence (data not shown). This indicates that there may have been some moisture introduced into this section (Fitzgerald and Allaway,
1991). There were no traces of any fluorescence seen in outer cortical cell walls (Figures 4.39 and 4.40). Figure 4.40 shows brown pigmentation not only in the cell walls but also in the extracellular material. In contrast to aerial rhizomorphs, blue HPTS fluorescence was seen in some walls of the inner
104 cortical zone (recognised on the basis of its morphology) at high magnification
(Figures 4.41 and 4.42). However, the intensity of the fluorescence was very low.
4.4 Discussion
The results reported in this chapter show that peripheral hyphae are permeable to the two symplastic tracers and the apoplastic probe HPTS. Both of the symplastic tracers are taken up across the plasma membrane and accumulated in the vacuoles while the apoplastic tracer permeates the cell walls but is excluded from the protoplasts.
The possibility of leakage as a result of wounding during experimentation is minimal. The two symplastic tracers applied are membrane permeant until the fluorescent product is released at which time it should be taken up into the vacuoles and trapped there. Cole, Hyde & Ashford (1997) found that in
Pisolithus tinctorius probenecid inhibits vacuolar uptake of the hydrolysed
(non fluorescent) probe and that fluorescence in this case accumulates in the cytoplasm. On removal of the probenecid, the cytoplasm clears and fluorescence accumulates in the vacuole, within minutes. If mechanisms are similar in Armillaria luteobubalina, as one would expect them to be, the levels of non-hydrolysed, non-fluorescent probe would not be expected to build up in the cytosol of exposed cells, and therefore would not be expected to move for significant distances by cell to cell transport. This tends to indicate that fluorescent cells are labelled with probe because they have come into contact with it via the apoplast immediately surrounding them, and that patterns of
105 labelled cells obtained with these symplastic tracers are not a result of cell to cell transport. The apoplastic tracer permeates non-modified cell walls but not walls impregnated with impermeable substances. It travels in the apoplast or the free space of the cell wall and does not bind to the cell walls, being anionic, and there is thus a high risk of redistribution of the tracer. However, it was applied in conjunction with rapid freezing and anhydrous freeze substitution, which not only arrests the movement of the dye but also prevents the loss or redistribution of water soluble fluorochromes including HPTS (Canny &
McCully, 1986; Fitzgerald & Allaway, 1991; Vesk et al., 2000). There is a body of evidence indicating that careful anhydrous technique prevents loss or redistribution of the tracer during processing and sectioning (Orlovich &
Ashford, 1995). Segments were excised from separate rhizomorphs to minimise redistribution of the tracer, after cutting.
The data indicate that peripheral hyphae are able to absorb molecules from the external solution and that their walls are permeable to quite large molecules
(the molecular weight of Na3 HPTS is 524.37). The outer cortex, irrespective of the degree of pigmentation, was mostly impermeable to the symplastic tracer, except that the tracer accumulated in a few cell patches that bridged peripheral hyphae to the inner cortex and medulla of the rhizomorph where most cells contained the tracer. In contrast, the pigmented cell walls and extracellular space of the outer cortex hindered the movement of the apoplastic tracer to the inner regions. There was no barrier to movement of the apoplastic tracer into the rhizomorphs where there was no pigment in the outer cortex of the rhizomorph.
106 The two symplastic tracers showed similar results irrespective of the fact that they differ in the mechanism that releases the fluorescent product. Carboxy-
DFFDA is hydrolysed by esterases which cleave off the acetate to release the green fluorescent product DFF-, while CMAC is modified by a glutathione S- transferase-mediated reaction to produce a membrane permeant fluorescent thiol-ether adduct that is transported to vacuoles by an ATP-dependent transporter (Haugland, 2002; Cole et al., 1997). Carboxy-DFFDA did not permeate whole rhizomorphs and this may be related to the molecular weight being 496.38 in comparison to 209.63 for CMAC; it may be too large to pass through the walls of outer cortical rhizomorph hyphae. The diffuse fluorescence seen at the inner surface of the inner cortex and medulla may be interpreted in relation to the protocol used for the symplastic tracer experiments. It could be caused either by leakage of probe from cut cells at the edge of the gas canal, or from movement of the probe along the interface between the gas canal and the inner cortex surface from the cut end.
A few patches of cells containing the symplastic tracer in the relatively impermeable outer cortical zone may be interpreted as a pathway from peripheral hyphae to the inner cortical cells across the less permeable compact outer cortical zone where outer cortical cells are reported to be inter-connected via pores to each other (Cairney et al., 1988b). Comparison with hand-cut sections treated with the tracer supports this view in that it appears that more cells are labelled because they have more access to the probe. We know that outer cortical cells have the potential to take up the tracer if they are completely exposed, as for example in cut longitudinal sections treated with
107 CMAC, where most cells of the outer cortex became labelled, in contrast to whole rhizomorphs where only a few patches of CMAC fluorescence were seen.
Similar results with symplastic tracers in both aerial and submerged rhizomorphs are consistent with the findings of Cairney et al. (1988a), who observed identical patterns in the uptake of radiolabelled phosphate by the apices of rhizomorphs of Armillaria mellea in culture growing from wood logs and in vitro cultured rhizomorphs (liquid medium).
Our observations of there being more cells containing symplastic tracers at the apex support the view that the apoplast is more permeable at the apex. This is in accord with experiments by Cairney et al. (1988a) and Morrison (1975), where radioactive labelling is recorded as being more at the apex than distant from it. A lower accumulation of radioactive tracers in the sub-apical region of the rhizomorph (Cairney et al., 1988a; Morrison, 1975) was correlated to either difference in pigmentation (Morrison, 1975) and/or more differentiation
(Cairney et al., 1988a).
The pigmentation in walls or extracellular spaces of the outer cortical zone hindered the movement of the apoplastic tracer. In aerial rhizomorphs the apoplastic probe is not seen in any part of the inner regions as the pigmented cortical region formed a barrier to the movement of the tracer. While we can not infer from these results that pigmented cells are also impermeable to water and nutrients we can infer that there is a substantial drop in apoplastic permeability in the region where walls and extracellular spaces are pigmented.
108 There is a body of evidence from other fungal structures that pigmentation is correlated with a fall in apoplastic permeability (Young & Ashford, 1992;
Young & Ashford, 1995; Young & Ashford, 1996) and is likely to reduce the capacity for uptake from the external environment (Ashford et al., 1988; Vesk et al., 2000).
Though they still leave many questions unanswered, the results presented here show that rhizomorphs are sufficiently radially permeable to take up water and ions from their surroundings. A permeable layer of peripheral hyphae allows uptake of nutrients that then can travel symplastically across the rhizomorph radius and circumvent the barrier of a variously pigmented outer cortex. It cannot be excluded that there could have been uptake of the tracer at the tip followed by longitudinal transport to the base, and then from inside to outside.
However, the circumstantial evidence does not support this explanation, as more fluorescence was observed in peripheral hyphal cells (their walls) than inner cortical cells (or their walls). Nevertheless, the results with the apoplastic tracer in whole (non-excised) rhizomorphs indicate that there is a potential pathway for longitudinal transport in the apoplast of the inner cortex; this would be limited by the outer cortex (if pigmented) on the outside and the air spaces of the gas canal on the inside.
The work presented here shows that rhizomorphs are radially permeable to both apoplastic and symplastic tracer and that pigmentation in cortical cell walls or in extracellular material hinders the further transport of apoplastic tracer. It is worth emphasizing that impermeability to the apoplastic tracer in
109 no way implies impermeability to water which is much smaller molecule (18
Da compared with 524.37 Da for HPTS).
110 4.5 Legends
Figures 4.14.9. Aerial and submerged rhizomorphs treated with carboxy- DFFDA to demonstrate which hyphae are most permeable; epifluorescence (Figures 4.1, 4.2, 4.4, 4.5, 4.7 and 4.8) and DIC optics (Figures 4.3, 4.6 and 4.9). Figure 4.14.4. Intact submerged rhizomorphs exposed to carboxy- DFFDA for 3 h and then sectioned. Figure 4.1. Peripheral hyphae show strong labeling. Bar = 100 Pm. Figure 4.2. Single peripheral hypha shows accumulated fluorescence in the vacuoles (v). Note: no fluorescence was observed in the cytoplasm. Figure 4.3. The same region as in Figure 4.2 at lower magnification shows relationship of labeled vacuoles to hyphal tip. Bars = 10 Pm. Figure 4.4 showing strong fluorescence in almost all the peripheral hyphae and diffuse fluorescence at the edge of inner cortex and extending into the medullary region of the rhizomorph. No fluorescence was seen in the cortical region of the rhizomorph. Bar = 100 Pm. Figures 4.54.9. Sections of rhizomorphs treated with carboxy-DFFDA for 0.5-1.5 h after cutting. Figure 4.5. Transverse section of a submerged rhizomorph showing fluorescence in peripheral hyphae (p) and in both cortical cells, i.e. inner cortex (i) and outer cortex (o). The labeling mostly occupied in the whole profile in contrast to the peripheral (p) and medullary hyphae (m,) where the localization of the tracer was more obviously in vacuoles. Figure 4.6. Same section as Figure 4.5 showing that fluorescence is not detectable in many out of focus peripheral hyphae. Bars = 50 Pm. Figure 4.7. A few inner cortical cells of a submerged rhizomorph show the tracer to be accumulated in the vacuoles. Bar = 20 Pm. Figure 4.8. Submerged rhizomorph showing labeling in the vacuoles of some medullary hyphae (m). Figure 4.9. DIC image of the section shown in Figure 4.8. Bars = 100 Pm.
111 112 Figures 4.104.14. Intact submerged rhizomorphs treated with CMAC and then hand sectioned to show which hyphae are permeable; epifluorescence micrographs (Figures 4.10, 4.12 and 4.14); DIC optics (Figures 4.11 and 4.13). Figure 4.10. Peripheral hypha showing labeling of the vacuoles (v) and no fluorescence in cytoplasm. Figure 4.11. DIC image of hypha shown in Figure 4.10 for comparison. Bar = 10 Pm. Figure 4.12. Transverse section of a rhizomorph showing fluorescence in peripheral (p) and inner cortical hyphae (i) but mostly not in the outer cortical cells (o). Occasionally a radial string of fluorescent cells bridges the outer cortex (arrows). Figure 4.13. DIC image of the section shown in Figure 4.12. Zones are labeled as in 4.12. Bars = 50 Pm. Note: majority of the fluorescence was observed across the entire cell profile and not all the inner cortical cells fluoresced. Figure 4.14. One of the inner cortical cells showing fluorescence in the vacuoles. Bar = 20 Pm.
113 114 Figures 4.15-4.18. Sections of rhizomorphs treated with CMAC for 0.5-1.5 h after cutting. Figure 4.15. Longitudinal section of a submerged rhizomorph showing fluorescence in most of the outer cortical cells (o) and few inner cortical cells (i). The labeling mostly occupied in the whole profile of outer cortical cells, in contrast to the inner cortical cells (i) where the localization of the tracer was obviously in vacuoles. Figure 4.16. Same section as Figure C for comparison. Bars = 20 Pm. Figure 4.17. Single peripheral hypha shows accumulated fluorescence in the vacuoles. Figure 4.18. The same field as in Figure A for comparison. Bars = 10 Pm.
115 116 Figures 4.194.22. Transverse sections of the rhizomorphs near the tip showing the fluorescence in the inner cortical zone. Figure 4.19. Hand-cut transverse section of submerged rhizomorph treated with carboxy-DFFDA showing labeling of cells across the entire inner cortical (i) zone and in the medulla (m), but not in the outer cortex (o). Figure 4.20. DIC image of the same for comparison. Note another half section overlapping the complete rhizomorph transverse section. Bars = 50 Pm. Figure 4.21. Intact aerial rhizomorph treated with CMAC and then sectioned showing bright fluorescence in most of the inner cortical (i) cells and lower fluorescence in outer cortex (o). Labeling was observed in entire cell profiles. Some medullary hyphae (m) show labeling in vacuoles (arrow). Figure 4.22. DIC image of the same section as in Figure 4.21 for comparison. Bars = 100 Pm.
117 118 Figures 4.234.30. Epifluorescence (Figures 4.23, 4.25, 4.27, 4.29) and corresponding DIC (Figures 4.24, 4.26, 4.28, 4.30) images of aerial rhizomorphs treated intact with apoplastic tracer HPTS, or RO water (controls) for 1.3 h, freeze substituted, and sectioned dry. Tissues are labeled as follows: peripheral zone (p), outer cortex (o), inner cortex (i) and medulla (m). Figure 4.23. Control. No autofluorescence can be seen in any part of the rhizomorph. Figure 4.24. DIC image of the same section for comparison. Bars = 100 Pm.
Figures 4.254.30. Aerial rhizomorphs treated with apoplastic tracer HPTS. Figure 4.25. HPTS fluorescence is most obvious in peripheral hyphal cell walls with no fluorescence in any other part of the rhizomorph. Figure 4.26. DIC image of the same section showing pigmentation in outer cortical cell walls. Bars = 20 Pm. Figure 4.27. Peripheral hypha (arrow) showing typical blue fluorescence of HPTS in the cell walls, but not the protoplasts. Note the fluorescence in the septa. Figure 4.28. DIC image of the same section as is Figure 4.27. Bars = 10 Pm. Figure 4.29. Cortical and medullary zone showing no fluorescence in any part of the section, including the cell walls. Figure 4.30. DIC image of section shown in Figure 4.29. Note the pigment in the cell walls and extracellular material of the outer cortex, and lack of free radiating peripheral hyphae in the peripheral zone (p). Bars = 20 Pm.
119 120 Figures 4.314.36. Epifluorescence (Figures 4.31, 4.33 and 4.36) and DIC (Figures 4.32, 4.34 and 4.36) images of transverse sections near the tip of a submerged rhizomorph treated with HPTS for 1.3 h. freeze substituted, and sectioned dry. Figure 4.31. Cell walls of free peripheral hyphae, walls and extracellular space in the peripheral zone, and outer tangential walls of the outer cortical all fluoresce strongly. There is a sharp drop in fluorescence in underlying outer cortical cell walls which mostly show a low level of fluorescence throughout, except in occasional patches where fluorescence is brighter (arrows). Inner cortical cell walls fluoresce, generally at a slightly higher level than the outer cortex. Figure 4.32. DIC image of the same section for comparison. Bars = 20 Pm. Figure 4.33. Detail of outer cortex (o) from a section in the same series as Figure 4.31 showing the distribution of HPTS fluorescence throughout the apoplast of the outer cortex, mostly at a low level. The brightest patches of fluorescence occur in the extracellular spaces at the cell corners. Figure 4.34. DIC image of same section for comparison. Bars = 20 Pm. Figure 4.35. Inner cortical cells showing fluorescence restricted to the cell wall. Note: No diffuse fluorescence was seen in the intercellular spaces. Figure 4.36. DIC image of the same section as in Figure 4.35 for comparison. Bars = 10 Pm.
121 122 Figures 4.374.42. Epifluorescence (Figures 4.37, 4.38 and 4.41) and DIC (Figures 4.38, 4.40 and 4.42) images of longitudinal (Figures 4.374.42) of submerged rhizomorphs treated intact with HPTS, freeze substituted and dry sectioned in the region 710 cm distant from the tip. Figure 4.37. Peripheral hyphae showing bright HPTS fluorescence in the cell walls; note septa are fluorescent. Figure 4.38. DIC image of the same section as in Figure 4.37. Bars = 10 Pm. Figure 4.39. Outer cortical cell walls showing no HPTS fluorescence in them. Figure 4.40. DIC image of Figure 4.39 showing pigmentation in the cell walls, and to a lesser extent in the extracellular matrix (arrow). Bars = 15 Pm Figure 4.41. Pale HPTS fluorescence is seen in a few inner cortical cell walls at high photographic exposures. Figure 4.42. DIC image of Figure 4.41 showing both ovoid and narrow hyphae typical of the inner cortical region. Bars = 20 Pm.
123 124 Chapter 5. Conduction and Transport of Oxygen to Rhizomorphs through ‘Air Pores’
5.1 Introduction
Several observations indicate that rhizomorphs of Armillaria species need a good supply of oxygen for their initiation and growth. They range from observations that submerged mycelium must reach the substrate-air interface for rhizomorphs to be initiated and grow (Snider, 1959; Smith & Griffin, 1971;
Pareek, Cole & Ashford 2001) and their abundance in the upper horizons of soil (e.g. Ono, 1965 and 1970; Redfern, 1973; Morrison, 1976) to data on the effects of restricting oxygen supply on rhizomorph growth in soil and culture
(Münch, 1909; Reitsma, 1932; Jacques-Félix, 1968; Smith & Griffin, 1971;
Rishbeth, 1978; Worrall, Chet, & Hüttermann 1986). Rhizomorphs grow faster than free hyphal tips (Rishbeth, 1968) and they contain a large consolidated mass of cells (Motta, 1969 and 1971; Motta & Peabody 1982), so it can be assumed that they have a high oxygen requirement.
Many authors have alluded to gas spaces within rhizomorphs of Armillaria
(e.g. Brefeld, 1877; De Bary, 1887), and some have postulated a role for rhizomorphs in aeration, and indicated that the central space of the rhizomorph is a gas-channel that might be involved in conducting oxygen (vide Garraway,
Hütterman & Wargo, 1991; Fox, 2000). For example, the experiments of
Reitsma (1932) indicated that the oxygen for growing apices in Armillaria mellea is supplied through rhizomorphs. Smith & Griffin (1971) in Armillaria luteobubalina (then known as Armillariella elegans) showed that high partial
125 pressures of oxygen at the origin of rhizomorphs promote their growth, but oxygen at the rhizomorph surface inhibits growth, as Jacques-Félix (1968) had thought for A. mellea. Smith & Griffin (1971) reported that when oxygen becomes limiting for apical growth, rhizomorphs flatten and become lobed, and growth becomes much slower; the distance at which flattening occurred depended on the partial pressure of oxygen applied. Investigations such as these are widely quoted in support of a role for rhizomorphs in aeration (e.g.
Watkinson, 1979) and for oxygen in their development (e.g. Rayner et al.,
1985). However the precise pathways by which oxygen from the atmosphere may reach the tips of the first cluster of rhizomorphs produced have never been elucidated.
In the seminal paper of Brefeld (1877) A. mellea rhizomorphs developed in culture were described as arising from clusters of sclerotia, presumably the same structures that Snider (1959) termed ‘microsclerotia’. There are many reports of the presence of structures described as ‘air pores’ or ‘breathing pores’ on various parts of the mycelium, but particularly on rhizomorphs.
Hartig (1874) described tufts of hyphae extending from rhizomorphs of A. mellea. Smith & Griffin (1971) showed tufts of hyphae on aborted side branches of rhizomorphs of A. luteobubalina growing into the air, and interpreted them to be ‘breathing pores’. The central canal of the aborted branch bearing the ‘breathing pores’ was shown to be directly connected with the central canal of the main rhizomorph, and Smith & Griffin postulated that oxygen diffused through the intertwining hyphae about 0.5 cm behind the tip of the aborted branch into the central canal of the rhizomorph. Granlund, Jennings
126 & Veltkamp (1984) and Intini (1987) used the term ‘air pores’ for structures on the surface of the rhizomorphs of A. mellea and A. obscura respectively.
Pareek et al., (2001) noted that fluffy structures developed when the mycelium reached the substrate-air interface in tubes of liquid medium inoculated with A. luteobubalina. These were necessary for the initiation or growth of rhizomorphs, but they did not look like sclerotia except in very early stages.
Closer inspection of these fluffy structures at the substrate-air interface suggested a similarity between them and the air pores developed on rhizomorphs. In this chapter we report the occurrence of ‘air pores’ consistently at inoculation sites, air pore development and internal structure, and their connection with rhizomorphs. Finally, their role in aeration is examined quantitatively by estimating oxygen conductances with an oxygen electrode. We conclude that rhizomorphs develop beneath clusters of air pores, not clusters of microsclerotia, and that the air pores and their subtending rhizomorphs conduct oxygen.
5.2 Materials and methods
5.2.1 Fungal material
Armillaria luteobubalina Watling & Kile (isolate 930199KGS) from the culture collection of the University of Western Sydney was used throughout.
This was originally isolated from fruit bodies collected from Gore Hill Park,
(33.5o S, 151.15o E) North Sydney in June 1993. Stock cultures were maintained in Petri dishes on malt marmite agar (2 % malt, 2 % agar and 0.1 %
127 marmite w/v) at 23 oC and sub-cultured every 4-8 weeks. To produce ‘air pores’ and rhizomorphs, 10 mm plugs of inoculum were cut from the edge of actively growing colonies and transferred to 2 % malt marmite agar in Petri dishes and incubated in the dark at 23 oC.
5.2.2 Air pore developmental studies
Cultures were monitored daily for at least 10 d using a Leica dissecting microscope to determine whether there was any relationship between air pore development and rhizomorph initiation. Photomicrographs of successive stages of air pore development were taken with a Zeiss Axiocam digital camera and images obtained were assembled using Adobe Photoshop 6.0 software.
5.2.3 Structural studies
For light microscopy, air pores judged to be mature and emergent from the mycelium growing at the agar-air interface were randomly selected. Air pores and associated mycelium were excised from the colony using a double-edged razor blade. The agar below the mycelium underlying the air pore was trimmed. Samples were first rinsed in reverse osmosis purified water and then fixed in situ in a glass vial with 5 % (v/v) glutaraldehyde in 0.08 M piperazine-
N-N’-bis (2-ethanol sulphonic acid) (PIPES) buffer at pH 7.0 for 3 h, rinsed 3-
4 times in 0.3M PIPES buffer at pH 7.2 (Salema & Brandão 1973), dehydrated through a graded ethanol-water series and gradually infiltrated with medium grade LR White resin (London Resin Co.) over 5 d, all at 4 oC. Samples were polymerised at 60 oC for 10-12 h in gelatin capsules. Semi-thin sections (1.5-2
128 Pm) were cut on a Reichert Ultracut microtome using glass knives, collected on to drops of water on slides and dried on a hot plate at 35 oC for 12 h.
Sections were stained with 0.05 % toluidine blue at pH 4.4 and observed with bright-field optics. Photomicrographs were captured using a Zeiss Axiophot microscope equipped with an image point cooled CCD video camera
(Photometrics, Tucson, AZ) and then processed using Adobe Photoshop 6.0 software.
5.2.4 Oxygen conductance measurements
Conductance of air pores and other tissues to oxygen from the air was measured with gas diffusion chambers fitted with oxygen electrodes, exactly as described by Curran (1985) except that the chamber volume was reduced to about 10 ml to increase sensitivity. The electrode output was displayed on an iMac computer (Apple Computer Inc., Cupertino, California, USA) through a
PowerLab/4SP analog:digital converter using Chart 3.6.6 software (AD
Instruments Pty Ltd, Castle Hill, Australia). Chambers were routinely leak- tested, and electrodes were calibrated twice daily and whenever significant drift was detected.
Fungal tissue was mounted in 6 mm plastic tubes, to allow its insertion into the electrode chamber which had originally been designed for mangrove pneumatophores. Tubes of 1.5 ml capacity cut from 3 ml plastic pipettes were used to punch pieces of agar with attached pigmented mycelium from the cultures. Each sample was then pushed slightly out of the tube and a thin slice of agar cut off with a double-edged razor blade, to expose cut ends of the
129 rhizomorphs. A thin layer of Vaseline petroleum jelly was applied around the circumference of the sample to prevent any leakage and it was pushed back into the tube. Extreme care was necessary to avoid any breaks in rind or agar, or contamination of the exposed surface with Vaseline. The outside of the plastic pipette tube with the sample in it was then lubricated and inserted through the rubber disc of the electrode chamber, as described by Curran
(1985). The layer of agar and mycelium with melanized crust formed a barrier between the air and the inside of the oxygen electrode chamber, and entry of oxygen depended on conduction through either the fungus or agar.
To identify the oxygen conducting capacity of different parts of the fungus, the experiments were divided into several sets. Oxygen conductances of groups of air pores with an inoculum plug, melanised crust, aerial rhizomorphs and plain agar (agar from a region of the same plate but not containing fungus) were measured. To clarify whether the inoculum plug was involved in conduction of oxygen, conductivities of groups of air pores grown on cultures without an inoculum plug, and melanised crust and aerial rhizomorphs, were measured. To ascertain whether the oxygen conducted from the air pores travelled through the rhizomorph, conductance with uncut and cut ends of rhizomorphs was compared. Aerial rhizomorphs with one or both ends cut were also investigated. All sets of experiments were replicated 6 times, each replicate from a different plate. Conductances were calculated as in Curran (1985) and results were analysed by t tests for paired comparisons.
130 5.3 Results
5.3.1 Development of air pores
Three to four days after inoculation a radially spreading sparse white mycelium was observed around the inoculation plug. From this several balls of clustered white hyphae emerged (Figure 5.1). These hyphal clusters were the first distinctive stage of air pore development and under our conditions they first became obvious on day 4 or 5 after inoculation. They were about 0.5 mm in diameter and white. They occurred on the agar surface in close proximity to the inoculum plug and careful observation revealed that the basal part was submerged (Figure 5.2). There was no sign of any rhizomorph tips and at this stage these clusters were designated as stage I of the air pore development. The developing air pores continued to grow and measured 0.5-0.7 mm in diameter on day 6. The majority of them were still white and appeared as undifferentiated aggregates of hyphae with the exception of a few that showed a dark coloured spot, which was less than 0.01 mm and was on one side of the air pore (Figure 5.3). Also, several rhizomorph tips were seen in the agar located specifically beneath the developing air pore clusters. This phase was designated as stage II of air pore development. The third stage of air pore development was defined by the onset of complete pigmentation. Under our conditions this had first occurred by day 7 after inoculation. At this stage the developing air pore was larger (diameter 1 mm) and was lightly pigmented
(Figure 5.4). This consolidated ball of pigmented hyphae was overlain by a fluffy mass of white hyphae from the colony (Figure 5.4). There was some
131 variability in time of initiation and rate of development of air pores, so that by the 7th day after inoculation there were numerous air pore initials but not all were yet pigmented (Figure 5.5). Beneath this cover of white mycelium the pigmented structure extended away from the agar surface to create a lightly pigmented cylinder that was best seen after removing the mycelial covering
(compare Figures 5.6 and 5.7, day 8 after inoculation). On the 9th day, with further growth, the entire columnar, pigmented air pore was seen to have emerged through the mycelium (Figure 5.8). By day 10 the air pores had attained a height of 3-7mm and were viewed as being fully developed and designated as stage IV of air pore development (Figure 5.9). At this stage, the apical part of the air pore was fluffy and white in contrast to the lightly pigmented basal part (Figure 5.9). On day 10 the radially spreading mycelium of the colony showed first clear signs of colour. Pigmentation was observed around the air pore spreading up to 5 mm in diameter from the basal part of it and by day 28 the entire mycelium was covered with a darkly pigmented crust
(Figure 5.10) as in pseudosclerotia or pseudosclerotial plates (PSP) as described by Campbell (1934) and Lopez-Real (1975). The air pores were also pigmented but much more lightly than the colony as a whole (Figure 5.10). The apical region of each air pore still retained its fluffy appearance (Figure 5.10).
From 28d onwards the air pores were regarded as being mature and designated as stage V of air pore development. They were hydrophobic and were oriented in various planes relative to the agar surface. At no stage of their development did air pores resemble rhizomorph tips and there were no indications of an organised meristem.
132 The average number of hyphal clusters per culture counted on day 5 after inoculation was 25 (Table 5.1) with a range from 13-39. It was difficult among these white hyphal clusters to pick out those that were destined to develop into air pores, and it was not until the air pores were all lightly pigmented on day 7 that this distinction could be made with confidence. The number of air pores
(about 21 per dish at day 7) did not change over the next three days, although some of them failed to develop fully, and remained as small rounded pigmented aggregations surrounded by a dense white mycelium (Figure 5.11,
Figures in parentheses Table 5.1).
Table 5.1. Initiation and growth of air pores and rhizomorphs in Armillaria luteobubalina cultures#
Day after Description of air pores No. of air pores Stage inoculation (mean + s.e.m.) designated 1-4 White mycelium 0 - 5 White hyphal aggregates 25.0 + 4.0 Stage I 6 White hyphal aggregates with patch 24.9 + 3.0 Stage II of dark colour 7 Lightly pigmented 21.0 + 3.0 Stage III 9 Lightly pigmented, elongated 21.6 + 2.4 Stage III 10 Fully grown 21.8 + 2.2 Stage IV (including dormant air pores) (10.8 + 2.0) # Five culture plates were examined daily for 10 days after inoculation, and the numbers of air pores and rhizomorph tips were counted. Greater numbers of air pores and variability on days 5 and 6 after inoculation indicate the difficulty in establishing whether an aggregation of hyphae would develop into an air pore.
In all cases, rhizomorphs were initiated specifically beneath clusters of growing air pores (Figures 5.12 and 5.13). Rhizomorph tips appeared after the initiation of the hyphal clusters at an early stage of their development; that is they first became apparent at stage 2. They rapidly grew down into the agar
(Figure 5.14) until they reached the bottom of the dish where they bent and
133 then grew horizontally outwards (Figure 5.15). From an early stage the central space of the rhizomorphs was filled with gas, as indicated by its shining appearance in Figure 5.14.
5.3.2 External Morphology of fully developed and mature air pores
Air pores grew out into the air up to a height of 7 mm. They were usually, but not always, more or less perpendicular to the agar surface from which they emerged. Externally, they appeared fibrous and contained fine more or less parallel hyphae intertwined together to form a long cylinder. They were designated as “fully developed” (Stage IV) when they reached their full length and “mature” (Stage V) when they had become fully pigmented. On the basis of colour and superficial arrangement of hyphae, fully developed air pores appeared as pigmented cylinders with white fluffy tips (Figure 5.9). When mature all the regions of the air pore were pigmented including the fluffy tip
(Figure 5.10) and careful observation revealed a basal zone of intense pigmentation.
5.3.3 The internal structure of air pores and maintenance of continuity of their
gas space with the central gas space of rhizomorphs throughout
development
The spatial relationship between air pores and rhizomorphs and their gas spaces is best shown in vertical sections through the site of emergence of individual air pores. However, since neither the air pores nor rhizomorphs are exactly perpendicular to the agar surface, they are sectioned in many planes.
134 The earliest stage sectioned (day 6 after inoculation) shows two rhizomorph tips located beneath the edge of an obliquely sectioned developing air pore at stage II (Figure 5.16). The gas space in the medulla of these rhizomorphs is continuous with the atmosphere via gas spaces between the loosely arranged hyphae of the colony surface and extends into the base of the air pore. Few pseudo-parenchymatous cells are observed on one side of the surface layer of the culture, next to the air pore cluster. Fully extended but not yet mature air pores are seen to consist of very loosely intertwined approximately parallel hyphae with intercellular spaces of various sizes (Figure 5.17). Serial sections indicate that the air pore at this stage is a roughly cylindrical loose mass of undifferentiated hyphae. Clusters of hyphal profiles appear bonded together in patches by an extracellular material that stains reddish purple with toluidine blue (see inset in Figure 5.17). Figure 5.17, a near median section, shows two distinct regions of more closely spaced hyphae on the flanks of a central space containing few hyphae located at the base of the air pore. The medulla in the base of a rhizomorph is situated immediately beneath the air pore and is in continuity with a narrow gas space located beneath the surface of the colony.
This region is shown better in Figure 5.18 where gas space continuity can be traced from the medulla of a submerged rhizomorph through the base of the air pore to the atmosphere. The pseudoparenchyma forming the cortex of the rhizomorph is easily distinguishable.
The hyphae at the colony surface adjacent to the air pore are loosely arranged at this stage but pigmentation begins around the air pores and the surface hyphae rapidly become consolidated into a pseudoparenchymatous crust with a
135 developing rind at its surface (Figure 5.19). The outermost cortical cells that will form the rind are small in diameter (5-18 Pm) and surrounded by extracellular lightly pigmented material while inner cortical cells are larger
(diameter 12-37 Pm) and without pigmentation. A wide gas space containing scattered fine hyphae has developed beneath this crust and separates it from the agar surface (Figures 5.19, 5.20).
The overall relationship between mature air pores, submerged rhizomorphs, and colony crust is shown in Figure 5.20. Beneath the two air pores shown is a cluster of rhizomorphs sectioned obliquely in various planes. The pseudosclerotial plate has differentiated into a consolidated pseudoparenchyma layer covered by a pigmented rind that appears brown to black. At higher resolution (Figure 5.21) the rind is seen to consist of tightly packed cells embedded in pigmented extracellular material, with no obvious gas spaces.
Several fine hyphae branch from wider hyphae in the underlying cortex and traverse the rind. Some emerge into the atmosphere while others appear cut off or broken. The rind appears to extend across the base of one of the air pores and is sectioned transversely as a ring surrounding a non-pigmented central region in the other. This region is shown at higher resolution in Figure 5.22.
Below the consolidated pseudoparenchyma layer forming the mycelial crust there is a large gas space containing loosely arranged hyphae (Figure 5.20). It is widest beneath each air pore and peters out with distance from air pore clusters. The medullary gas space in the base of rhizomorphs which also filled with loosely arranged hyphae is continuous with this gas space underlying the colony crust (Figures 5.20, 5.24).
136 The relationship between the tissues of individual air pores and underlying rhizomorphs at maturity is shown most clearly in sections that are near the median longitudinal plane of the air pore (Figure 5.24). The medullary gas space of an obliquely sectioned rhizomorph lies immediately below the air pore. It is surrounded by a pseudo-parenchymatous cortex except at its base. A mound of heterogeneous tissue has now differentiated at the base of the air pore (Figure 5.24). This consists of an outer pigmented rind and a narrow loosely arranged and discontinuous pseudoparenchyma surrounding a central region of fine sparsely arranged hyphae, with a large amount of extracellular space (Figures 5.22, 5.24). Groups of more or less parallel hyphae that appear to have their origin at the rhizomorph cortex (serial sections) have grown across the gas space and entered the basal region of the air pore in a plane oblique to the colony surface (Figures 5.24, 5.25). Many are cemented together in groups and anastomoses are frequent. This results in a complex arrangement of individual and aggregated hyphae bridging the gas space between the rhizomorph and the basal region of the air pore. Most of these appear to enter the air pore at its periphery, leaving a central region with a large amount of extracellular space containing loosely arranged fine hyphae, resembling the other gas spaces (Figures 5.22, 5.24, 5.25). Above the rind of the differentiated basal region the air pore comprises a cylinder of loosely arranged more of less parallel hyphae with large amount of extracellular space that forms a continuum with the atmosphere (Figures 5.23, 5.24), as shown at the earlier stage. Many of the hyphae are pigmented and/or have granular deposits on their walls and patches where hyphae are cemented together by extracellular
137 material are still apparent (Figure 5.23). The major differences between fully- developed and fully mature air pores are the presence of many more pigmented hyphae throughout the latter and the differentiated basal region.
It is clear from Figure 5.24 that continuity of gas space between the rhizomorph medulla and the atmosphere via the air pore depends crucially on the nature and amount of extracellular space in the rind, specifically in the basal region of the air pore. This region was therefore examined in more detail.
Figure 5.25 shows that the tissue of the rind and pseudoparenchymatous cortex in the basal mound of the air pore is not consolidated, in contrast to that over the colony crust. There are many areas where there is a clear continuity between the gas space within the basal mound and the atmosphere via gaps in the rind (Figure 5.25). In many of these areas the cortex is poorly developed or not at all developed (Figure 5.26). Many of the hyphae that extend into the air pore from the rhizomorph cortex are seen to have grown through this unconsolidated rind and contribute to the hyphal mass of the cylinder of loosely arranged hyphae comprising the main body of the air pore. Many of these emergent hyphae have pigmented walls.
A summary of the spatial relationship between the air pore and the rhizomorph is illustrated in a line diagram (Figure 5.27).
5.3.4 Structures similar to air pores observed on the rhizomorph surface
Apart from the air pores described above, structures very similar to them were seen but they did not develop from the mycelium. Instead they were found on
138 the surface of already existing aerial rhizomorphs, and at their tips. In vitro such structures were observed on aerial rhizomorphs either developed in soil or liquid medium. The relationship of these structures to rhizomorphs has not studied in detail here and warrants further investigation.
5.3.5 Oxygen conductance
Essentially no conductance to oxygen (average 4u10-12 m3s-1) was found unless the rhizomorphs originating from the region of the base of the air pore were sliced, exposing their cut ends inside the electrode chamber. With cut rhizomorphs below, the conductance of groups of air pores originating from inoculum blocks averaged 532u10-12 m3s-1 (Table 5.2 a). As a result of this finding, the base of the agar was sliced off in subsequent measurements, as described. The average calculated conductances to oxygen of plain agar blocks without mycelium, and regions of the colony with melanised crust, but no rhizomorphs or air pores, were 18u10-12 m3s-1 and 20u10-12 m3s-1, respectively
(Table 5.2 b, set 2). These very low values presumably represent the rate of oxygen leakage around the agar in the plastic tubes. Aerial rhizomorphs had very variable conductance which, although it averaged 187u10-12 m3s-1, was not statistically significantly different from the agar and colony with melanised crust. However, groups of air pores (which in this set all had part of an inoculum plug present) showed significantly greater conductance to oxygen, averaging 987u10-12 m3s-1. The presence of part of the inoculum plug with the air pores in this set introduced the concern that inoculum plugs might have been responsible for some of the conductance. To clarify whether or not the
139 inoculum plug was involved in oxygen conduction, cultures grown by inoculating from a loop rather than with a plug of agar were used for a further experiment. Groups of air pores grew near the site of inoculation, and showed average oxygen conductance of 619u10-12 m3s-1. This was significantly higher than the conductances of colony with melanised crust and aerial rhizomorphs
(mean conductances 7u10-12 m3s-1 and 34u10-12 m3s-1 respectively, Table 5.2b set 3). Although conductances of aerial rhizomorphs had not proved significantly different from those of colony with its melanised crust or agar alone, the variability of their conductances suggested that aerial rhizomorphs might have an oxygen conducting capacity that was not being adequately detected. Whether the conductance of aerial rhizomorphs would be increased by cutting off their tops was tested, but the increase in conductance was not statistically significant (Table 5.2c).
Across all our observations, the conductance of groups of air pores to oxygen averaged 679+68u10-12 m3s-1 (mean + standard error of the mean, n = 25).
These groups of air pores consisted of 4+0.3 air pores, resulting in a conductance per air pore of 198+27u10-12 m3s-1 (see also Table 5.3).
Table 5.2. Oxygen conductance in cultures of Armillaria luteobubalina# a. Effects of cutting rhizomorphs beneath the agar Set 1 N 6 Air pores (rhizomorphs not cut below, with melanised crust and agar) 6 a + 3 Number of air pores 3 + 0.3 Conductance per air pore 2 + 1 Air pores (rhizomorphs cut below, with melanised crust and agar) 532b + 132 Conductance per air pore 182 + 55
140 b. Conductances of parts of cultures Set 2 3 N 6 6 Agar 18a + 3 - Agar with melanised crust 21a + 7 7a + 12 Air pores (rhizomorphs cut below, with melanised 987b + 86 619b + 81 crust and agar) Number of air pores 4 + 0.7 5 + 0.7 Conductance per air pore 291 + 83 132 + 27 Aerial rhizomorph (rhizomorphs cut below, with 187a + 131 34a + 26 melanised crust and agar) c. Effects of cutting on aerial rhizomorphs Set 4 N 6 Aerial rhizomorph (rhizomorph cut below, with melanised crust and 29a + 30 agar) Aerial rhizomorph (cut top, with melanised crust and agar) 21a + 7 Aerial rhizomorph (cut top and rhizomorph cut below, with 212a + 94 melanised crust and agar) #Within each row of the Table, a difference in superscript letter indicates a statistically significant difference (p<0.05) by t-test for paired comparisons. All conductances are in 10-12 m3 s-1 (nanolitres per second); mean + standard error of the mean are shown.
Table 5.3. Summary of oxygen conductances in cultures of A. luteobubalina over all observations#
Mean s.e.m. n Agar 18 3 6 Agar with melanised crust 12 6 18 Aerial rhizomorph (rhizomorph cut below, with melanised 56 28 31 crust and agar) Air pores (rhizomorphs cut below, with melanised crust and 679 68 25 agar) Number of air pores 4 0.3 25 Conductance per air pore 198 27 25 # Conductances are in 10-12 m3s-1 (nanolitres per second); mean, standard error of the mean and number of samples are shown.
5.4 Discussion
This chapter reports the development of air pores at the substrate-air interface in pure cultures of Armillaria luteobubalina and confirms their role in
141 conducting oxygen to rhizomorphs. These air pores are not borne on aborted side branches of rhizomorphs as were the ‘breathing pores’ of Smith & Griffin
(1971), but are produced on recently inoculated mycelium and initiated before rhizomorphs. Nevertheless, there is a very clear spatial relationship between the two structures. Of the two terms previously used for aerating structures as found on rhizomorphs, ‘breathing pores’ (Smith & Griffin, 1971) and ‘air pores’ (Granlund et al., 1984; Intini, 1987), we chose ‘air pores’ because gas exchange is presumably by diffusion rather than mechanical ventilation.
Structures apparently identical to developing air pores were described in the first cultures of A. mellea (then known as Agaricus melleus) grown by Brefeld
(1877). A contemporary of Koch, he isolated the fungus from single spores and grew them in slide cultures on a thinned plum mixture. His superb illustrations show what must be early stages of air pore development. Brefeld did not have the advantage of modern sectioning techniques and his substrate was limited in amount, and he interpreted the structures from which rhizomorphs emerge to be sclerotia. Snider (1959) also thought that rhizomorphs developed from
'microsclerotia' but did not investigate these structures in detail. It is not surprising that air pores have been mistaken for microsclerotia. They initially have very similar morphology. They arise as more or less spherical loose hyphal aggregates and later become pseudoparenchymatous and brown- pigmented in the outer part - a description that applies to many sclerotia including microsclerotia (Willetts, 1972; 1978; Willetts & Bullock, 1992).
However air pores continue their development to become elongate structures with a totally different internal structure from microsclerotia.
142 Initiation of air pores early in mycelial development, before rhizomorphs, explains why rhizomorphs are not initiated in culture unless there is an interface between the mycelium and air: air pores must develop first. It explains Rishbeth’s observations (1968) that from their inception rhizomorphs grow much faster than hyphal tips. The strategic timing of air pore development and their position suggest that they play a strategic role in gas exchange and aeration of Armillaria rhizomorphs, and perhaps also associated mycelium as well.
The spatial relationships of the various regions of the air pore at maturity and its relationship with the rhizomorphs, melanised crust and surrounding mycelium are illustrated schematically in Figure 5.27. This figure also shows the distribution of gas spaces in these various regions and how they interconnect. It is a summary of the structural evidence, from sectioning data, of the capacity of the system to aerate. There is no apparent barrier to oxygen diffusion along the loose hyphae of the air pore region above the rind, and it would be expected that oxygen would diffuse throughout the intercellular spaces of this region from the atmosphere readily and without any impediment.
The spaces in the rind at the basal region of the air pore, given their size, would also be expected to be freely permeable to oxygen, and oxygen should diffuse readily into the mound of tissue beneath; this contains intertwined hyphae and large intercellular spaces. The gas space in the base of the air pore is in direct continuum with that of the central canal of all the rhizomorphs via the reservoir of gas space in the mycelium on the agar surface beneath the air pore, and this is continuous with that underlying the melanised crust of the colony. This gas
143 space is of greater depth under and adjacent to air pores than elsewhere. Its oxygen should be replenished via the air pore. This will supply the mycelium under the melanised crust with oxygen and provide an additional reservoir of oxygen for the rhizomorphs to draw upon. As oxygen diffuses down this gas space system as a whole, it will be utilised by hyphae: the gas space will be depleted of oxygen, and gradients will be set up. Regions with the most hyphae relative to the volume of gas space they draw on will be the most depleted, as also will be those furthest from the atmosphere. This all depends on the extracellular space being filled with gas and not liquid. This is difficult to assess from sections but it is obvious in fresh material: the silvery appearance of the rhizomorph canals in the agar indicates that they are filled with gas, and the melanised crust can be peeled away from the agar surface to reveal that the hyphae are free and not consolidated by extracellular mucilage. There is gas space in rhizomorphs of A. luteobubalina (Smith & Griffin, 1971; Pareek et al.,
2001).
The ultimate test of whether the system works is whether it conducts oxygen.
This can be tested directly by measuring oxygen conductances with an oxygen electrode. Since there is no conductance through the agar alone the only way for oxygen to reach the oxygen electrode must be via the fungus and its gas spaces. The melanised colony with submerged mycelium, but no rhizomorphs or air pores, also has negligible conductance. The only two types of sample that showed appreciable conductance were some of those containing aerial rhizomorphs with both ends cut, and all of those containing air pores attached to rhizomorphs that were cut near their apex. These latter results indicate that
144 air pores and rhizomorphs both conduct oxygen and are connected. The significant difference between samples with and without cut rhizomorphs, however, does not allow us to conclude that the tip, or rind and cortex along the flanks of rhizomorphs, are impermeable to oxygen, since they were covered with agar which we know from the control is not permeable. For similar reasons, the oxygen permeability of the melanised rind cannot be assessed from these data. The closely packed hyphae of the rind differentiated into pseudoparenchyma where the cells are thick walled, cemented and melanised suggest low permeability, but the layer is pierced by emergent hyphae and, especially if broken, these may provide gas passages.
There is a body of evidence to suggest that air pores are common and widespread. Not only are they found in A. luteobubalina in culture, but a reassessment of the literature indicates that they occur in cultures of other species, on or as branches of rhizomorphs, and as extensions of pseudosclerotial plates. For example, illustrations in Garrett (1953) and
Mwenje & Ride, (1996) show what appear to be air pores. Structures identical in position and external morphology to ours are visible in illustrations of
Kenyan rhizomorph-producing isolates in agar culture (Otieno, Sierra &
Termorshuizen, 2003). Granlund et al. (1984) show a scanning electron micrograph in which they tentatively identified an air pore on a rhizomorph which previously grew in air. They thought these were side branches that had stopped growing. They compared them with Smith & Griffin's (1971)
‘breathing pores’, but provided no further information on their structure or development. The air pore they illustrated was elongated, with a tuft of loose
145 hyphae, very much like our air pores. Intini (1987) showed air pores on rhizomorphs, and commented that their principal function appears to be improvement of gas exchange between the inside of the rhizomorph and the outside air. Secondary air pores form on mycelium on the colony surface in our agar cultures and new rhizomorph clusters arise from them. We have also found air pores directly produced on rhizomorphs of A. luteobubalina in agar and soil grown cultures in Petri dishes (M. Pareek, unpublished). The A. mellea rhizomorphs many metres long found floating in water by Goffart (1903-4) and
Findlay (1950) also point to a need for some kind of aeration structure at intervals along their length. The little tufts of hyphae associated with rhizomorph branching described by Hartig (1874) may also have been air pores. However, we never saw any air pores develop into rhizomorph branches or fruiting bodies. Air pores in our situations always retained their internal structure even in cultures several months old.
The capacity of rhizomorphs to conduct oxygen was strongly suspected, and the basic structure of rhizomorph-derived breathing pores was previously reported to be simple tufts of hyphae bursting through the rind of rhizomorphs at the substrate air interface. The mature structure that we have found here is much more complex, but nevertheless contains hyphae that have grown through a rind layer. Those hyphae which cross the sub-air pore space into the basal tissue mound and thence through the spaces in the rind to form the apical cylinder of hyphae appear to have arisen from consolidated pseudoparenchyma around the site of origin of the developing rhizomorphs. The mature structure satisfies the requirement for longevity and survival and, since the crust and the
146 air pore are both hydrophobic, should prevent waterlogging of the air pores even when it is sufficiently wet to cause pooling on the crust surface.
It appears that we have here an elaborate, sophisticated aeration system that extends and incorporates all the mycelial thallus. It meets the following requirements for a good aeration system: abundant gas space relative to tissue volume; good connectivity with all parts that need aeration (involving if necessary longitudinal channels in conducting organs); and protected entry points that have longevity and survive adverse conditions, that resist waterlogging and yet also survive desiccation, and that have good support so that they do not collapse. The possession of a more elaborate gas space system than was formerly thought perhaps explains the success of some Armillaria spp. as pathogens in waterlogged situations. It may also be an important factor in pathogenesis in allowing a mycelium to grow on a wet root surface and to send hyphae or rhizomorphs into live roots or cut stumps, where conditions may be hypoxic. It would seem that a preoccupation with terminology and a shift in focus to transport of nutrients and water rather than gas have distracted attention from the importance of gas transport in Armillaria rhizomorphs, and from their fundamental differences from cords of wood rotting and ectomycorrhizal fungi such as Phanerochaete velutina and Suillus bovinus, not only in structure but also in function. It appears that their morphology and anatomy reflect the fact that Armillaria rhizomorphs are strongly suited for gas transport, a point well understood by early workers, and this is what differentiates them from other structures. Their role in this is supported by an
147 apparent developmental plasticity and ubiquity of air pores and an aeration system in associated mycelium that was formerly unappreciated.
148 5.5 Legends
Figures 5.15.11. Stages of air pore development. Figure 5.1. Several loose clusters of non-pigmented hyphae (arrows) are obvious on and around the inoculum plug, 4d after inoculation. Bar 1 mm. Figure 5.2. Stage I of air pore development. Clusters of white hyphae on day 5 after inoculation, measuring 0.5 mm in diam.. Arrows indicates region submerged in agar; there are no organised structures, which can be related to rhizomorph tips at this stage. Figure 5.3. Stage II of air pore development. Appearance of a hyphal cluster on day 6. The cluster now measures 0.7 mm in diam. with patch of dark colour (arrow). This is the first stage at which tips of rhizomorph are seen beneath the developing air pores. Bars 300 μm. Figure 5.4. By 7d the hyphal clusters are larger and more obvious and some are pigmented. Bar 10 mm. Figure 5.5. Stage III of air pore development. Detail of two pigmented hyphal clusters at day 7, the larger is approx 1 mm in diameter and lightly pigmented. Note white mycelium associated with the cluster. Figure 5.6. By 8d the white mycelium is denser and covers a pigmented cluster. Figure 5.7. Removal of the overlying white mycelium shows the "fluffy" tip of a more elongate structure, and the pigmented surface of the colony out of focus. Figure 5.8. Air pore at 9d with the white mycelium removed to show the appearance of the air pore as a lightly pigmented cylinder with a slightly fluffy tip. Bars 1 mm. Figure 5.9. Stage IV of air pore development. By 10d air pores have reached maximum length of about 7 mm and are termed "fully-extended" or “fully- developed”. Note the lightly pigmented cylindrical structures each with a white fluffy tip, emergent from mycelium at the agar surface. Pigmentation of the colony surface is apparent. Bar 5 mm. Figure 5.10. Stage V of air pore development where a cluster of mature, completely pigmented air pores has arisen from the interface between the inoculum plug and the agar surface, 30 d after inoculation. The surface of the colony is now darkly pigmented. Bar 10 mm. Figure 5.11. Air pore apparently arrested at an intermediate stage of development at 21d after inoculation, similar to the stage of development typically found at 7/8 d in air pores, which will mature fully. It is pigmented and covered by the white mycelium. Bar 250 μm.
149 150 Figures 5.125.13. Relationship between emergent air pores and submerged rhizomorphs in Petri plate cultures. Figure 5.12. Surface view of colony 30d after inoculation, showing clusters of air pores in 3 distinct locations (circles A, B and C). Air pores at position A around the inoculum plug are fully pigmented and are horizontal while those at positions B and C are vertical. Figure 5.13. Underside of the same Petri-plate on day 30 showing that there are 3 sites of rhizomorph initiation that correspond with the air pore clusters (A, B, and C). Rhizomorphs emerging from position ‘A’ are short in contrast to those from positions B and C, which have grown out and reached to the edge of the plate. Bars 10 mm. Figure 5.14. Vertical section through the agar in the centre of a group of developing air pores showing a cluster of rhizomorphs. Bar 1 mm. Figure 5.15. View from the underside of the plate showing rhizomorphs growing down into the agar and then turning as they reach the bottom of the agar. Bar 5 mm.
151 152 Figures 5.165.18. Vertical sections through air pores and adjacent areas at various stages of development. Figure 5.16. Edge of a developing air pore at stage II fixed 6d after inoculation, sectioned obliquely, with two very young rhizomorphs (r) in longitudinal section beneath. A gas space is apparent in the rhizomorph medulla (m) and this can be traced to the junction with the air pore (A) and the atmosphere (arrowheads). Note the darkly stained apical meristem of each rhizomorph. Stained with toluidine blue (pH 4.4). Also note few parenchyma-like cells on the left hand side of the air pore (P). Bar 100 Pm. Figure 5.17. Longitudinal section through an air pore at stage IV showing continuity between the medullary gas space (m) of a developing rhizomorph and gas space in the base of the air pore (a) which has loosely arranged hyphae and an extracellular space continuous with the atmosphere. The rhizomorph medullary gas space contains fine hyphae and is surrounded by the pseudoparenchymatous rhizomorph cortex (c). Arrows indicate agar surface. Note that the hyphae at the colony surface are loosely arranged with intercellular spaces; they are not yet consolidated into a rind and cortex, nor separated from the agar surface by a wide gas space. Bar 100 Pm. Figure 5.18. Median longitudinal section through base of a “fully–developed’ air pore at stage IV showing that it consists of more or less parallel hyphae in a loose arrangement with a large amount of extracellular space. The basal region is not yet fully differentiated; it consists of a central air space with few hyphae interspersed in all directions (s), surrounded by a zone with more abundant hyphae around the periphery (arrows). There is as yet no differentiated rind within the air pore base. Bar 50 Pm. Figure 5.19. Developing crust at the colony surface in the vicinity of a fully-developed air pore (stage IV). The outer (o) cortex is pigmented and becoming rind like. It grades into a non- pigmented pseudoparenchymatous inner cortex (i) containing hyphal profiles of various sizes. Beneath this is a gas space containing loosely arranged hyphae. Continuity between these hyphae and those of inner cortex is apparent. Bar 30 Pm.
153 154 Figures 5.205.23. Sections showing mature air pores (Stage V) and surrounding regions of the colony. Figure 5.20. Vertical section through the
colony showing the base of two obliquely sectioned mature air pores (A1 &
A2) with a cluster of rhizomorphs (sectioned in various planes) beneath, surrounded by loosely arranged mycelium (my) growing submerged in the agar. The rind (R) and pseudoparenchyma (c) at the surface of the colony are fully differentiated and separated from the agar surface (arrows) by a gas space (g) containing fine loosely arranged hyphae. The medullary gas space (m) of several of the submerged rhizomorphs is seen to open into this. Rind in the base of air pores is continuous with the rind over the rest of the colony. The basal region of A2 is sectioned approximately transversely so that the rind with pseudo-parenchyma cells beneath it forms a ring of darkly pigmented tissue surrounding more loosely arranged central tissue (asterisk). Bar 250 Pm. Figure 5.21. Vertical section through the colony surface near a mature air pore showing fully pigmented rind (R) and underlying pseudoparenchymatous cortex (c). Rind cells (R) have a thick wall and a small lumen; walls and surrounding matrix are a deep brown with no obvious gas spaces. The rind grades into the underlying cortex; cells vary in size and wall thickness and are surrounded by a matrix that grades from brown to non-pigmented in the innermost region. Fine hyphae arising from the cortex traverse the rind (arrows) and are cut at their point of emergence. One arises from an inner cortical cell (i) of large diameter. Bar 20 Pm. Figure 5.22. TS of air pore basal region showing a central gas space containing loosely arranged hyphae (*) surrounded by a more consolidated cortex and rind. Bar 50 Pm. Figure 5.23. TS across the air pore column above the basal region showing profiles of many parallel loosely arranged hyphae with a large amount of extracellular space. Bar 20 μm.
155 156 Figures 5.245.26. Longitudinal sections through individual mature air pores (stage V) fixed at 30d after inoculation. Figure 5.24. Median longitudinal section showing differentiation of the basal region into a consolidated mound of tissue comprising a central region (*) with a large amount of extracellular space surrounded by a discontinuous pseudoparenchyma layer with an outer pigmented layer that is continuous with the rind of the colony (arrows R). The region above contains more or less parallel loosely arranged hyphae with occasional cemented patches (arrowheads) and variously sized intercellular spaces. The rhizomorph medullary gas space (m) is continuous with gas space beneath the air pore extending up into the basal mound of differentiated tissue and continuous with gas space underlying the adjacent colony crust. Individual hyphae and hyphal aggregates arising from the cortex (c) at the base of the rhizomorph cross this gas space and enter the base of the air pore at its periphery. Bar 100 Pm Figure 5.25. Detail from an adjacent section showing hyphal connections traversing the gas space from the cortex at the rhizomorph base into the base of the air pore. Bar 100 Pm Figure 5.26. Median longitudinal section showing detail of the basal region from an adjacent section of the same air pore to that shown in Figure 5.24. Note the large gaps in the rind and discontinuity of the cortex where there is very obvious continuity between the internal gas space and the atmosphere (large arrows). Bar 50 Pm.
157 158 Figure 5.27. The rhizomorph gas canal (gc) connects with the atmosphere vis gaps in the rind within the air pore and the extracellular space between the parallel hyphae cemented together by extracellular material (E). The gas canal also connects with the large amount of gas space in the mycelium growing over the agar surface and under the colony crust (AM) consisting of a rind (R) and cortex (C) and the rhizomorph (r) is surrounded by submerged mycelium (SM) at least in its basal region. Bar = 500 Pm.
159 160 Chapter 6. Numerical modelling of oxygen transport through an air pore
6.1 Introduction
The rate at which oxygen diffuses through biological entities is of critical importance in the analysis and understanding of many biological activities. In many cases, the role of oxygen diffusion is quite clear and understandable. For example, the oxygen carried by blood transport in our bodies (for that matter, in all mammals) eventually has to diffuse to cells for cellular respiration.
However, for smaller biological entities, oxygen uptake and transport mechanisms are not so obvious. For instance, fungal rhizomorphs, which can be responsible for spreading infection (Garrett, 1970) and transporting nutrients
(Granlund, Jennings & Veltkamp, 1984; Eamus et al., 1985; Cairney, 1992) and oxygen (De Bary, 1887; Smith & Griffin, 1971) are found to be closely associated with additional structures called air pores (Snider, 1959; Smith &
Griffin, 1971; Pareek, Cole & Ashford, 2001) which conduct and transport atmospheric oxygen to rhizomorphs in the gas phase (Chapter 5).
In order to understand the physiology of these fungi it is important to quantify clearly the effects of oxygen diffusivity through air pores. In this chapter, three different models for oxygen transport are investigated. The experimental data used here are collected from the experiments done on air pores as described in
Chapter 5. This information was then used to model rigorously the mass transport equations describing the diffusion of oxygen within the air pore.
161 6.2 Experimental Set-Up
The experimental set has been detailed in Chapter 5. For the purpose of numerical modelling, a schematic of the experimental set-up is shown in Figure
6.1.
6.3 Unsteady State Diffusion Model
In the diffusion electrode experiment, the oxygen conductivity is normally calculated on the basis of the cumulative oxygen transferred to the chamber in a given time period:
'C uV / 't C Ac (1) C Aa C Ac where, C is the conductivity of oxygen (m3 s1), the term in the parenthesis is the net amount of oxygen transferred to the chamber (mol), V is the volume of oxygen chamber, 'CAc is the change in oxygen concentration in the time interval 't, CAc is the oxygen concentration in the chamber, and CAa is the ambient oxygen concentration during the experiment (mol m3).
Clearly, equation (1) is a simple expression and gives a quick first estimate of the oxygen conductivity (Hovenden & Allaway, 1994; Curran, 1985). This formula is based on an implicit assumption that the oxygen driving force (CAa
CAc) is constant during the experiment; however, in practice, the driving force does change with time (due to build up of oxygen in the chamber).
Furthermore, the respiration of the air pore (however, small) is also neglected.
162 Therefore, in this chapter, a modified form of equation (1) is proposed to account for this behaviour.
Let 't in equation (1) be very small, i.e., limit 't o0, so that 'CAco0; with these limits the equation reduces to:
dC Ac C V u dt (2) C Aa C Ac
After a rearrangement:
dC V Ac C C C (3) dt Aa Ac
A physical interpretation of equation (3) reveals that the term on the left hand side is the net rate of oxygen accumulation in the chamber (mol s1) and the term on the right hand side (RHS) is the net rate of oxygen transferred to the chamber due to the concentration gradient. However, a part of the oxygen transferred through the air pore is utilised in respiration, and a correction term can be subtracted from the RHS:
dC V Ac C C C k l (4) dt Aa Ac d
where, kd is the coefficient of respiration (mols consumed per unit length per unit time, mol m1s1) and l is the length of the air pore.
Equation (3) can now be integrated within the limits, at t 0 CAc 0:
163 CAc dC t V ³ Ac ³ dt (5A) 0 C C Aa C Ac kd l 0
§ k l · C d V ¨ Aa ¸ ln¨ C ¸ t (5B) C ¨ kd l ¸ ¨ C Aa C Ac ¸ © C ¹
§ kd l · § C ·½ C Ac ¨C Aa ¸®1 exp¨ t ¸¾ (5C) © C ¹¯ © V ¹¿
Values of the parameters kd and C should be estimated using the experimental data through a linear regression. However, the current set-up of oxygen electrode does not permit a decoupling of the two parameters. Therefore, the respiration coefficient was neglected (kd 0) from the analysis:
V § C Aa · ln¨ ¸ t (6A) C © C Aa C Ac ¹ on rearranging:
§ C Ac · C ln¨1 ¸ t (6B) © C Aa ¹ V
The value of C should be estimated using a linear regression on the experimental observations. A plot between ln(1CAc/CAa) and t should give a straight line with a slope C/V.
To illustrate this approach, the sample data in Figure 6.2 were used. The first plot shows calibration of the equipment and the second plot is for the leak test.
164 The next three plots indicate oxygen diffusion behaviour within the plain agar, the pigmented mycelium and the aerial rhizomorph respectively. Finally, the time-dependent concentration behaviour for the air pore is shown in the last plot (shown within the red rectangle in Figure 6.2). The calculation using
9 3 1 3 equation (1) gave C 1.38u10 m s (with 't = 480 s, 'CAc = 0.54 mol m ,
6 3 3 V = 10.5u10 m , CAa = 0, CAc = 8.55 mol m ). When ln(1CAc/CAa) was plotted against t, a straight line was obtained with a slope C/V (1.0u104 s1)
(Figure 6.3a); thus, C 1.05u109 m3 s1. The correlation coefficient (R2) between the fitted straight line and the actual data was 0.9999, which was an indication of an excellent fit. Clearly, equation (1) over-predicted the oxygen conductivity by about 24%. This can be attributed to one of the underlying assumptions of equation (1) that the oxygen driving force was constant throughout the observation period. However, the actual driving force decreases progressively with the time; the higher average driving force (concentration gradient) used in equation (1) would over-estimate the overall oxygen conductivity. Equation (6), on the other hand, is based on an unsteady state mass balance of oxygen, and therefore, makes more accurate predictions of the oxygen conductivity.
A comparison between experimental observations and model predictions is shown in Figure 6.3(b). Clearly, the model allowed predictions of oxygen concentrations over a much bigger range than what was possible using the experimental set-up. However, a magnified view of the data (shown in the inset) showed a significant deviation between the model predictions and experimental observations and this time, equation (6) slightly under-predicted
165 the oxygen conductivity. Therefore, a need to apply more rigorous modelling techniques is not overemphasized.
6.4 Oxygen Diffusivity
The oxygen conductivity calculated in the previous section is a dimension dependent quality. For example, with all other quantities fixed, a bigger air pore (length-wise) due to more diffusional resistance will conduct less oxygen than a small one. Similarly, a broader (bigger diameter) air pore will conduct more than a narrow one. To offset for this scale-dependence the oxygen conductivity should be multiplied by a factor:
L D C (7) A where, D is diffusivity of the air pore, L and A are the length and area of the air pore. For a typical case, in Figure 6.1, L 3.5u103 m and the area of the air pore was A ¼(Sd2) ¼S(300u106)2 7.06u108 m2. Using C 1.05u109 m3 s1, we get D 5.2u105 m2 s1. On Comparing this diffusivity with that of oxygen in water (2.4u109 m2 s1), it becomes very clear that a mechanism other than just the diffusion of oxygen through the liquid phase is operational in the air pore. To further quantify this transport mechanism a CFD-based model was developed.
166 6.5 Computational Fluid Dynamic (CFD) Modeling
The transport of oxygen through an air pore to the electrode chamber can be treated as an unsteady state diffusion. Based on this premise, the transport of oxygen through an air pore was modelled using a computational fluid dynamics (CFD) software – FLUENT. Important numerical dimensions for the sample air pore are listed in Table 6.1.
Table 6.1. Numerical dimensions for CFD simulations.
Item Value Diffusion electrode Volume 10.5 ml Diameter 3.8 cm Height 1.0 cm Air pore Diameter 350 μm Length 0.5 cm
The following assumptions have been made in the development of this model:
1. The transport of oxygen in an air pore can be written as the two different terms in the species transport equations – namely, the convective and diffusional terms. However, the transport due to the convective term can be neglected. 2. The diffusional resistance due to the mycelium and residual rhizomorph is negligible. 3. There are no chemical reactions in the air pore. 4. The resistance to oxygen transport due to gas film outside the air pore is negligible. 5. The oxygen diffusivity is constant throughout the air pore.
The diffusion of oxygen through the air pore may be written as (Bird, Stewart
& Lightfoot, 2001):
167 wC A C V ( J) (8) wt A with
w w w i j k (9) wx wy wz
where, CA is the concentration of oxygen, V is the velocity of fluid, J is diffusive flux of oxygen, and i, j, k indicate to unit vectors in the directions x, y, z. The term on the LHS indicates the rate of increase of oxygen concentration.
The first term on the RHS represents the rate of oxygen transport due to convection, and the rate of oxygen transport due to the diffusion is accounted for in the second term.
The diffusive flux in equation (8) may be approximated by Fick’s law:
J DC A (10)
On substituting equation (10) in equation (8), and assuming constant diffusivity: wC A C V D( 2C ) (11) wt A A
Equation (11) can also be written as (using equation 9):
2 2 2 wC A ªwC AVx wC AVy wC AVz º ª w C A w C A w C A º « » «D 2 D 2 D 2 » (12) wt ¬ wx wy wz ¼ ¬ wx wy wz ¼
168 where, Vx, Vy and Vz are the velocity components in x, y and z directions respectively.
Finally, the transport due to convection may be assumed to be negligible, and the first term on the RHS of equation (12) can be neglected:
2 2 2 wC A ªw C A w C A w C A º D« 2 2 2 » (13) wt ¬ wx wy wz ¼ where D is the net diffusivity of oxygen in the air pore. To solve the above partial differential equation, the air pore and the oxygen chamber were divided into small finite volumes. A 2D grid (or mesh) of a simplified computational domain is shown in Figure 6.4. The grid construction was done using a
FLUENT pre-processor called GAMBIT. Most of the oxygen chamber had a grid size of 0.005u0.005 cm2. However, a graded mesh was used within the air pore and the region immediately beneath it. The complete grid consisted of
151550 computational nodes, and one unsteady state simulation using adaptive time step (from 0.01 s to 5 s) required 8 hours of computer time to simulate 10 minutes of transient diffusion (on a Pentium 2.8 GHz processor). Initially, the diffusivity a calculated using equation (7) was used (D 5.2u105 m2 s1).
Following initial and boundary conditions were employed:
1. At t 0, CA 0 (in the air pore and the chamber).
2. For t t 0, CA CAa (only on the tip of the air pore).
A commercial CFD code called FLUENT was used to solve equation (13) subjected to above initial and boundary conditions. Figure 6.5 shows contours
169 of oxygen mass fraction within the oxygen chamber (a colour indicates a particular concentration level). As expected, the concentration was highest near the tip of the air pore, and beyond that the concentration decreased quite progressively. Also shown in Figure 6.5 are representative contours of oxygen concentration (but with a different colour map) within the air pore. Since only the tip of the air pore was assumed to be conducting, the concentration was highest on the top tip of the air pore, and decreased rapidly from the top to the bottom.
During the simulations, the volume averaged concentration within the diffusion chamber was monitored. As shown in Figure 6.6, the monitored concentrations were compared with the experimental observations (red coloured plot). With the initially estimated diffusivity (5.2u105 m2 s1) the calculated concentration profile (blue coloured curve) was significantly different from the observed profile. Therefore, simulations were performed for discrete values of oxygen diffusivities until a match was found. For this set of experimental data, the best fit value of diffusivity was found to be 5.9u105 m2 s1.
6.6 Conclusions
Different results were produced by the linear formula of Curran (1985) (see
Chapter 5), the logarithmic model, and the CFD simulation. The first of these
(eq. 1) resulted in D of 6.8u10-5 m2s-1, the second gave D of 5.2u10-5 m2s-1, and the CFD simulation gave an intermediate value of D of 5.8u10-5 m2s-1. In consequence, using equation 1 as in Chapter 5 probably overstated the best
170 estimate of conductance by about [(6.8-5.8)/5.8], i.e. 17% (equation 7). The overestimation is linear and applies in general to every calculation of conductance in Chapter 5. The results can not, however, now be recalculated because of the dependence (section 6.4) on air pore dimensions, which were not individually recorded. Table 6.2 indicates approximate values for the mean conductances from the Summary Table (5.3), allowing for 17% overestimation.
Table 6.2. Revised estimates: summary of oxygen conductances in cultures of A. luteobubalina over all observations #
Agar 15 Melanised crust and agar 10 Aerial rhizomorph (rhizomorph cut below, with melanised crust and agar) 48 Air pores (rhizomorphs cut below, with melanised crust and agar) 579 Conductance per air pore 169 #Conductances are in 10-12 m3 s-1.
The CFD simulation corrects the scale, but does not change the conclusions to be reached from Chapter 5. Although it is outside the scope of this thesis, it would be worthwhile for mangrove researchers, who worked with larger material and apparatus, to re-examine their findings using modelling such as that described here.
171 Air-pore Mycelium
Oxygen Electrode Agar
Figure 6.1. Schematic of a diffusivity chamber; oxygen electrode is situated at the base. On top of the chamber is the fungal sample inserted in the plastic tube of known length and volume.
Figure 6.2. Data set for the sample calculation where conductivities to oxygen are shown on X-axis (mol O m3) and samples tested are revealed on Y-axis. (From experiments in Chapter 5). Electrode readings are as follows: 'Calibration': after label 4, zero oxygen; after label 5, air. 'Leak test': between labels 7 and 10, zero oxygen (no leak is evident). 'Plain agar': 15 to 17, very slight increase in oxygen. 'Pigmented mycelium': 18 to 24, very slight increase in oxygen. 'Aerial rhizomorph': from after 29, no increase in oxygen. 'Air-pore': from 37, rapid increase in oxygen in the chamber. Horizontal scale bar: 10 minutes.
172 0 y = -0.0001x -0.01 R2 = 0.9999
-0.02
-0.03
-0.04 (1-Cac/Caa) ln -0.05
-0.06
-0.07 0 200 400 600 Time (s-1)
D
9
8
7
6 0.6 ) -3 0.5 Model 5 Experimental 0.4
(mol m (mol 4
Ac 0.3 C 3 0.2 2 0.1
1 0 0 100 200 300 400 500 0 0 10000 20000 30000 40000 50000 Time (s)
E
Figure 6.3. (a) Linear regression of oxygen diffusivity data for calculating the slope of equation (6). (b) Comparison between experimental and model predictions.
173 (a)
(b)
Figure 6.4. Computational grid (mesh) for mathematical calculations. Geometry created in Gambit showing air pore as a cylindrical structure sitting on top of the diffusivity chamber.
174 0.1 0.5 1 s
Figure 6.5. Typical contour of oxygen concentration in the diffusion electrode chamber and air pore taken at 0.1, 0.5 and 1 s.
0.5 D = 5.1x10-5 D = 5.5x10-5 D = 5.9x10-5
) 0.4 -3 Experimental
0.3
0.2 Concentration (mol m
0.1
0 0 50 100 150 200 250 300 350 400 450 500 Time (s)
Figure 6.6. Repetitive simulations to evaluate best fit value for air pore diffusivity.
175 Chapter 7. Conclusions & General Discussion
7.1 Introduction
The research presented in this thesis aimed to clarify the proposed role of rhizomorphs in absorption, aeration and long distance translocation where some things have been taken as accepted fact without any hard evidence. It must, however, be made clear that the work described in this thesis is exclusively with axenically-grown material, and therefore extension of the findings to the field can only be speculative in absence of further experimentation. The primary aspect of this research was to study the structure and role of rhizomorphs of Armillaria luteobubalina to understand the association of these vegetatively spreading infecting structures to moist conditions where these propagate in relatively anaerobic conditions. This was approached in a number of ways. Firstly, structural studies were carried out on two different types of rhizomorphs produced under in vitro conditions (Chapter
3). Secondly, the apoplastic and symplastic permeabilities of these rhizomorphs were assessed (Chapter 4) and thirdly, the structures previously described as micro sclerotia were studied, which were in fact, air-pores
(Chapter 5). Lastly, modelling of oxygen conduction through an air-pore was carried out (Chapter 6).
7.2 Rhizomorph Structure and Organization
It has been established that A. luteobubalina produces two different types of rhizomorphs, i.e. aerial and submerged rhizomorphs which are similar in
176 structure to those that have been described in A. mellea in axenic culture
(Chapter 3). The two types of rhizomorph are different in external morphology and to some extent in internal structure. Differences appear to be due to the different environments where they are growing, and they can be compared to subcorticalis and subterranea rhizomorphs produced by A. mellea in natural forest situations, as described by Hartig (1874). The aerial rhizomorphs are pigmented in contrast to non-pigmented submerged rhizomorphs. Also, submerged rhizomorphs have aerenchyma-like cells in the inner cortical zone with intercellular space that connect with the central gas canal (Chapter 3) that provides a continuous, intercellular pathway for the rapid flux of oxygen through tissues growing in anaerobic conditions.
Once initiated, rhizomorphs grew mostly submerged into the agar medium and at a very fast rate (5-8 mm/day). In some instances, a few of the submerged rhizomorphs emerged out of the medium to become aerial and pigmented
(Chapter 3). After becoming aerial, rhizomorphs slowed down in growth rate to
0-2mm/day. Factors that triggered the submerged rhizomorphs to change the growth direction and become aerial are not clear but once emerged they appeared superficially similar to subterranea rhizomorphs found in the natural environment. Also, the transition of the two forms into one another is demonstrated several times in illustrations by Hartig (1874). There is a consensus with the scientific community that the rhizomorphs growing aerially
(in vitro), first become pigmented and that the pigmentation then reduces their growth rate (see Smith & Griffin, 1971). Smith & Griffin (1971) showed that exposure to certain partial pressures of oxygen causes rhizomorphs to become
177 pigmented and growth rates concurrently slow down. Similar observations were made on hydroponically grown roots of rice, when submerged roots also became pigmented after exposure to the atmosphere (Enstone & Peterson,
1998). However, the existing notion that pigmentation reduces the growth of aerial rhizomorphs is a correlation rather than conclusive proof. Furthermore, subterranea rhizomorphs, which are variously pigmented, have been found to grow considerable distances suggesting that pigmentation alone may not be responsible for the inhibited growth of the aerial rhizomorph. However, the information on growth rates of subterranea rhizomorphs is unavailable and it can be argued that these may have been growing at slow rates for a long time.
One of the reasons for the inhibited growth of aerial rhizomorphs in vitro may be reduced nutrient or water availability. It is important to note that aerial rhizomorphs in in vitro cultures grow in air away from the water and/or nutrient medium and cannot absorb nutrients directly as subterranea rhizomorphs appear to do. This may also apply to short rhizomorphs (3 cm) of
Armillaria luteobubalina found extending from root surfaces in some field situations in Western Australia where the soil is reported to be poor in nutrients/water (Pearce et al., 1986).
Another factor that potentially affects rhizomorph extension is moisture level because cell expansion or extension results from water uptake. Rhizomorphs, which have been dried for a long time and were thought dead, have been found to come back to life again with exposures to moisture (Hartig (1874).
Association of Armillaria rhizomorphs with construction wood in bridges,
178 water pipes, water tunnels or utility poles submerged in water (Hartig, 1874), and their colonization of sprinkled irrigated logs which are completely water saturated (Metzler & Hecht, 2004) further suggests their association with high moisture levels and ability to sustain in relatively anaerobic conditions. More peripheral hyphae on submerged rhizomorphs in contrast to aerial rhizomorphs indicate a potential role of these structures to exploit the nutrients available in moist conditions. This explains the association of rhizomorphs with moist conditions in field situations.
The differences seen in the inner cortex of aerial and submerged rhizomorphs can be interpreted in the context to the role of rhizomorphs in aeration. There is much more gas space in submerged rhizomorphs than in aerial rhizomorphs.
Aerial rhizomorphs lack ovoid swollen cells and instead cells appear to be cylindrical structures cemented together. This structural variation can be interpreted in terms of differences in oxygen availability. The inner cortex tissue in submerged rhizomorphs can be compared to plant aerenchyma. Plant aerenchyma contains a high ratio of intercellular gas space to the volume occupied by cells. This has at least two potential functions; it acts as a large reservoir for the gas (oxygen) storage and by reducing the amount of cells it reduces the oxygen requirement per unit volume of tissue. The ovoid cells also have been reported to act as storage cells (Intini, 1987). The inner cortex of submerged rhizomorphs could thus be seen as serving a similar purpose to aerenchyma. It is the first ‘fungal aerenchyma’ to be described. It is next to the medullary region which has large gas spaces. The differences observed here in aerial and submerged rhizomorphs in relation to the inner cortical zone also
179 contradict the existing notion that the only difference between two different types of rhizomorph is pigmentation.
7.3 Permeability
7.3.1 Radial permeability
The peripheral hyphae, which form the outermost zone of the rhizomorph, are composed of fine hyphae, many of which run parallel to the rhizomorph axis and branch out perpendicular to it (Chapter 3), to provide a large surface area.
These peripheral hyphae are surrounded by mucilage, and they superficially resemble the root hairs of flowering plants. The acidic nature of the mucilage enables it to act as an ion-exchanger and change the relative mobilities of anions and cations near the root surface (Clarkson. 1991). As with root hairs of higher plants, the peripheral hyphae and their associated mucilage appeared to be involved in nutrient uptake. The peripheral hyphae are permeable to both apoplastic and symplastic tracers and can take up relatively large molecules
(i.e. fluorescent probes) from the surrounding solution (Chapter 4). This means by inference that they should also readily be able to take up nutrients (ions) which are much smaller. This is in accordance with previous literature showing absorption of nutrients along the length of rhizomorphs (Morrison, 1975). The role of the associated copious mucilage of the peripheral hypha is not studied in this research.
Peripheral hyphae also have potential to supply the absorbed material to inner tissues of the rhizomorph via symplastic transport (Chapter 4). The symplastic
180 tracers are taken up across the plasmalemma and into the symplast of these peripheral hyphae. Vacuoles took up the symplastic fluorescent tracers, and showed motility similar to motile tubular vacuole system demonstrated for
Pisolithus tinctorius. This suggests that radial transport in rhizomorphs could occur via the vacuoles. The vacuolar system in peripheral hyphal cells penetrates the dolipore septa that separate cells in a similar fashion as shown in the cords of Pisolithus tinctorius (Allaway & Ashford, 2001), providing the possibility for intercellular transport via the organelle (Shepherd et al., 1993).
However, it must be pointed out that, while this is a potential conduit, it has not been proven that any long distance transport actually occurs in this system
(Allaway & Ashford, 2001; Cairney, 2005).
Adjacent to the peripheral hyphae is the outer cortical zone, which is compactly arranged with little or no intercellular space and is variously pigmented. It has always been believed that, in common with roots, changes in uptake along rhizomorphs may be attributable to the development of various permeability barriers in the cortex such as pigmentation (Morrison 1975;
Clipson et al., 1987). In in-vitro cultures, submerged rhizomorphs grown in moist situations lacked pigmentation in contrast to aerial rhizomorphs.
Nevertheless, when submerged rhizomorphs were left in the culture for longer durations (for a month or more), they showed initiation of pigmentation in outer cortical zone. This began in the basal part of the rhizomorph but the pigmentation was not yet pronounced as seen in aerial rhizomorphs. This dark pigment has been identified as melanin in fungi (see Bloomfield & Alexander
1967) and lignin in plants (Henson, Butler & Day 1999)].
181 Pigment formation in fungi is generally viewed as a response to environmental stress (Bell & Wheeler 1986; Henson et al., 1999). Melanin is deposited in the cell walls or in extracellular space and may be a major component of the cell, for example, up to 30% for the spores of Agaricus bisporus (Rast &
Hollenstein 1977); this represents a considerable allocation of energy and resources and prompts an assumption that melanin production in fungus serves a useful purpose (Butler et al., 2001). The importance of melanin deposition in the survival and longevity of fungal propagules has been widely recognised for many years; melanin is reported to be essential for protecting fungi against UV, irradiation, radio-waves, desiccation and temperature extremes (Bell &
Wheeler 1986; Henson et al., 1999; Butler et al., 2001). Melanin in fungi is also reported to be important for resistance to microbial attack. For example, hyaline spores or hyphae in soil are seen to be quickly killed and lysed in comparison to melanised cells which may survive for several years (Bell &
Wheeler 1986; Henson et al., 1999; Butler et al., 2001).
The onset of pigmentation in fungi has been correlated with reduced permeability. Melanin deposition in sclerotia and other fungal structures is widely thought to occur to withstand desiccation in unfavourable conditions
(Willets, 1971; Willets & Bullock, 1992). Melanin in the cell walls is thought to prevent or at least reduce loss of water from the structure and to act as a barrier. Melanin in outer cortical cell walls of sclerotia (Young & Ashford
1992; 1995; 1996) and similarly suberin (which contains phenols similar to lignin) in the exodermis or endodermis of roots (Peterson, Emanuel &
Humphreys, 1981; Enstone & Peterson, 1997; Zimmermann & Steudle 1998;
182 Steudle 2000; Vesk et al., 2000) limits the permeability of the wall and/or extracellular material in the region of its deposition and acts as an apoplastic barrier. Pigmentation in outer cortical cell walls of the rhizomorphs similarly reduces apoplastic permeability and acts as a barrier to the probes. However, the probes are bigger than nutrients which are taken up as ions. Nevertheless, the nutrient requirement of the internal tissues of these pigmented rhizomorphs is likely to be fulfilled by symplastic transport, where peripheral hyphae radially uptake the tracer and transport it to the inner cortical cells to overcome the barrier of pigmented cortical layer (Chapter 4). The non-pigmented outer cortex of the submerged rhizomorph was permeable to the apoplastic tracer but did not show the probe neatly into the cell walls. Instead, it appeared diffused with few bright patches in corners of adjoining cell walls and or in inter/extra- cellular spaces. This may imply that cell walls of the outer cortex may have some barriers apart from pigmentation. It is known that rhizomorphs exposed to air are more pigmented and it has been shown that oxygen stimulates the activity of p-diphenol oxidase, an enzyme which catalyses the formation of brown pigment in the outer cortical cell walls (Smith & Griffin 1971). This outer cortical layer is comparable to exodermis of submerged roots where suberisation in the exodermis of hydroponically grown plants is found to be increased after the roots are exposed to air (Clarkson et al., 1987; Enstone &
Peterson, 1998). This patchy development is a characteristic of the uniform exodermis (Enstone & Peterson, 1998). In submerged roots, 11% of exodermal cells had either developing or mature suberin in contrast to 92% in the air- treated region. Exodermis eventually matures even when roots are completely
183 submerged but this occurs at considerable distance from the root tip (120 mm)
(Enstone & Peterson, 1998). The possibility of development of barriers in the outer cortical cell walls of rhizomorphs may be patchy and may be responsible for the diffused fluorescence of the probe. Nevertheless, there are apparent permeable pathways through this zone that provide the apoplastic pathway to inner regions of the rhizomorph where the probe is neatly restricted to the cell walls of inner cortical and medullary hyphal walls.
Interestingly, the pigment present in cortical regions does not make any difference in the radial permeability to the symplastic probe, as a similar permeability pattern is observed for non pigmented outer cortical regions.
However, the permeability is selective to only few outer cortical cells that connect the inner permeable cortex and medullary zones (Chapter 4). This pattern implies that either the symplastic probe is not moving into the cells through the apoplast of the outer cortex or there are fewer cortical cells with permeable apoplast. In Phragmites plantlets, Armstrong et al. (2000) have shown ‘passage areas’ that they called ‘windows’ in the hypodermis which are thin walled in comparison to other hypodermal cells. These windows are opposite to the initiating lateral and these have been shown to permit radial oxygen loss and therefore nutrient or water uptake. Another explanation for a patchy permeability in the outer cortex may be due to cell to cell transfer of the probe. Some outer cortical cells may be either connected to lateral peripheral hyphae which may be then inter connected to other cortical cells via pores as suggested by Cairney et al. (1988b) (Chapter 4).
184 7.3.2 Long Distance Transport
Vessel-hyphae have been reported to occur in the medullary region of field grown Armillaria mellea and have been given a role in the long distance transport of nutrients or water (Cairney, Jennings & Agerer, 1991). However, no such obvious specialised hyphae were found in A. luteobubalina rhizomorphs produced in vitro conditions. Instead, the medullary region was seen to consist of a large intercellular space containing interspersed fine hyphae. Whether rapid long distance transport occurs in the absence of vessel hyphae is unclear. Slower, long distance transport may occur via cell to cell transport in the symplast. Tiny droplets on the peripheral hyphae of aerial rhizomorphs of A. luteobubalina suggest bulk flow of water similar to that seen in the cords of Serpula lacrymans (Brownlee & Jennings 1982).
Transport via the symplast may occur not only through cytoplasmic continuities between hyphae via their septa but also via continuities in the vacuolar system. In Pisolithus tintorius, where vacuoles are motile and cross the septa (Sheperd, Orlovich & Ashford, 1993), transport could occur within the vacuolar compartment either by diffusion or bulk flow. Motile, tubular vacuoles have now been described in a wide range of fungi from all divisions
(Rees, Shepherd & Ashford, 1994) and tubular vacuoles have been found in cords of P. tinctorius as well as mycelium (Allaway & Ashford, 2001).
Rhizomorphs of A. luteobubalina contain motile vacuoles but in this work they are not obviously tubular and the question of any role in long distance transport is yet unresolved, as it is for other species.
185 7.4 Air pores: Initiation, Development and Significance
The two different types of rhizomorphs (aerial and submerged) are readily produced in vitro by Armillaria luteobubalina. However, they are only initiated after air pores (the structures previously described as microsclerotia or tufts) are first produced at the substrate air-interface on mycelium (Chapter 5). While studying the morphology and physiology of Agaricus melleus (=Armillaria mellea) in nature, Hartig (1874) first described these structures as ‘tufts’ that would seem to be air pores. He found these structures to occur widely from the wood colonised by the rhizomorphs and described air pores to be the precursors of branches, since he observed primordia of new rhizomorph branches to develop beneath them. He concluded, “up to this point the description of the origin of branches of the rhizomorph agrees essentially with the origin of young fruit bodies which I was successful in observing”. He also stated that these tufts disappeared as the rhizomorph was initiated, which is not the case in the current research. Air pores observed in the present study always persisted, remained intact and closely associated with the submerged rhizomorphs. Furthermore, when extending rhizomorphs came to the agar surface, groups of secondary air pores developed at this point.
Although structures that are apparently air pores were first observed and described by Hartig (1874), it was Brefeld (1877) who systematically studied their initiation and development in-vitro. He was first to culture spores of
Armillaria mellea (then called Agaricus melleus) on artificial nutrient solution and observed dense hyphal masses which became spherical in the centre of the
186 circular white mycelium. Brefeld (1877) reported that these originated from either single branches or several adjacent hyphae and thought they were sclerotia. He also cultured isolates from single spores and found out that they produced mycelium which produced air-pores followed by rhizomorphs.
Brefeld’s work is historically important in two respects: it was the first instance of use of a pure culture technique for any fungus (contemporary to Koch’s famous work, 1877), and it was also the first careful description of initiation and apical growth of rhizomorphs, based upon microscopic observation.
Brefeld (1877) saw early stages of air pore development but did not see later stages (presumably because his medium – plum decoction - ran out before they matured properly), and he misinterpreted them as microsclerotia. Since then, several studies have been done making many important contributions (see
Chapter 2), but the focus of these was mostly on factors affecting the initiation and growth of rhizomorphs such as pH, temperature, moisture, light, nutrient level of growth medium, and oxygen status of environment. Snider (1959) followed up Brefeld’s work. He again showed that fluffy ball-like structures appear on the mycelium prior to rhizomorphs. He also found these to be embedded in the agar but virtually always in contact with its surface
(analogous to the stage I of air pores in our conditions). However, he did not study their structure or development in detail and again misinterpreted them as
‘micro sclerotia’.
In the current study, rhizomorph tips were always found to be initiated after the air pores had emerged and they invariably developed beneath the developing air pores from non-pigmented areas. This agrees with the observations made by
187 Brefeld (1877) who described the development of rhizomorphs from “growing points” (a small distinctive area of growth) on the “sclerotia”. This is also consistent with the observations of Snider (1959) who correlated initiating rhizomorphs with clusters of hyphae borne at the substrate-air interface. At this stage, these authors observed slight pigmentation on these hyphal “balls” or
“microsclerotia”, but they both emphasized that it was from areas which remained white that the rhizomorph tips were initiated.
In the present study, light pigmentation was observed at one side of some air pores, probably the pseudoparenchyma cells as seen in micrographs of thin sections at stage II of air pore development. These pseudoparenchyma cells are part of the developing colony crust with its pigmented rind; this was clearly distinguished at stage IV of air pore development as a dark pigmentation extending from the base of the air pore (Chapter 5). It could be argued that this pigmented, pseudoparenchymatous crust which develops over the colony surface on agar is the equivalent of a pseudosclerotial envelope, the protective layer of thick-walled fungus cells which covers the bulk of infected wood. It both encloses and protects a mass of hyphae together with the woody substrate that they digest, thus named “pseudosclerotium”, because of its apparent similarity to sclerotia (Fox, 2000). Initiation of pseudosclerotial envelopes and pseudosclerotial plates has been reported to be influenced by various mechanical and physical factors (Fox, 2000). Gas phase composition is one of these factors (Hartig, 1874; Lopez-Real & Swift, 1977; Fox, 2000). However,
Rayner et al. (1985) commented that experimental studies done by Lopez-Real and Swift, (1975 and 1977) failed to confirm a role for oxygenation as
188 “stimuli” for pseudosclerotial plate formation. However, the role of oxygen in promoting melanisation is established and the cells at the crust surface are in direct contact with air.
Brefeld (1877) and Snider (1959), both found that rhizomorphs are not produced from the submerged mycelium until it has reached the surface and the structures they described as “microsclerotia” have developed aerially. Snider
(1959) related this requirement to the need for ready access to oxygen in the atmosphere and postulated that initiating rhizomorph tips have higher oxygen requirement per unit weight of tissue than hyphae of the mycelium due to high metabolic activities, a point also noted by Motta (1969 and 1971) as rhizomorph apices grow much faster than the hyphal tips. Snider (1959) presumed that the oxygen supply for initiating rhizomorph tips comes from the aerially borne “microsclerotia”. However, he lacked any hard evidence for this.
The best evidence for a special connection between the internal gas space of rhizomorphs and the atmosphere comes from the work of Smith & Griffin
(1971) who found that in A. luteobubalina, optimal growth of the rhizomorph required a high partial pressure of oxygen within the apex, but a low one outside. In the present study, it has been shown that air pores and initiating rhizomorphs not only have a spatial relationship but are linked by a connected gas space.
The presence of a gas space in rhizomorphs, and the potential for aeration of the rhizomorph tip via this gas space to sustain the rapid growth observed in
Armillaria rhizomorphs, has long been suggested (Watkinson, 1979) and there
189 is considerable circumstantial evidence for an oxygen-transporting capacity in rhizomorphs. However, evidence that air pores conduct oxygen from the atmosphere to the medullary space of rhizomorphs, and is essential for their development, is shown in the present research (Chapter 5). At stage II of air pore development, the gas spaces within the air pore have already formed a continuum with the intercellular gas-spaces of the medullary region of rhizomorph. The concept of rhizomorph aeration via air pores agrees with the work done on Armillaria luteobubalina by Smith & Griffin (1971), where they showed that the central canal of the main rhizomorph and gas spaces of aborted side branches bearing breathing pores are directly connected. The difference between findings of the current research and those of Smith & Griffin (1971) is that in this work, where Petri-plate cultures were used in contrast to the test- tube cultures used by Smith & Griffin (1971), air pores originated from the mycelium rather than an aborted side branch of a rhizomorph.
The air pores described here are neither aborted branches as described by
Smith & Griffin (1971) nor the precursors of rhizomorph branches or fruiting bodies (Hartig, 1874) but they show many parallels with them, as discussed before. Under the present experimental conditions, air pores are part of a much more extensive sophisticated aeration system that includes the rhizomorphs and the extracellular space of the mycelium between the crust and agar. Spaces within the air pore body connected up with air-space system via gaps in the rind that conducted oxygen with the conductivity ranging from 500900u1012 m3s1. The measured oxygen conductivity of air pores was much less tan that of pneumatophores, the aerial roots found in vascular plants but air pores are
190 much smaller than pneumatophores. Gas spaces within pneumatophores provide the pathway for the diffusion of oxygen from the air into the tissues submerged in anaerobic mud (Curran, Cole & Allaway 1986; Hovenden &
Allaway 1994). Air pores, in a similar fashion, have gas spaces within them that permit oxygen diffusion from the air to submerged rhizomorphs, growing in relatively anaerobic conditions. This possibly explains the extraordinary ability of Armillaria species to grow to meet waterlogged situations provided they have air. It may be an important factor in pathogenesis, where mycelium may grow on a root surface and send hyphae or a rhizomorph into the root, where oxygen concentrations are low.
7.5 Numerical Modeling of Oxygen Diffusion in Air pores
In this study, three different methods to estimate oxygen conductivity or diffusivity in an air pore of a rhizomorph have been presented. Firstly, a model based on the constant gradient of oxygen was used, which for the sample data, overestimated the value of the diffusivity as 6.8u105 m2 s1. Secondly, a model was developed on the basis of an unsteady state macroscopic balance of oxygen. Using this approach, the value of oxygen diffusivity was estimated as
5.2u105 m2 s1, which on comparison with experimental data underestimated the oxygen concentration. This prompted the use of a more rigorous CFD- based method to model the oxygen transport through the air pore. After a trial and error, the best fit value of oxygen diffusivity was found to be as 5.9u105 m2 s1, which was in the middle of the first two approaches. This analysis has permitted to precisely evaluate the oxygen diffusivity within this small
191 biological entity. The calculated oxygen diffusivity was at least about 10,000 times higher than the diffusivity of oxygen in pure water, which has indicated the existence of a strong oxygen conducting tendency of the air pore consistent with conduction in the gas phase.
7.6 Comparison of Rhizomorphs with Higher Plant Roots
Rhizomorphs have been reported superficially to resemble roots of higher plants (Cairney et al., 1991); however it is of interest to determine just how similar they are. Rhizomorphs, especially those of A. luteobubalina consist of a large gas space, which is involved in the transport of oxygen (Chapter 5).
Given the fact that rhizomorphs in axenic culture prefer to grow submerged and have large central space with extensive aerenchyma in cortical region, rhizomorph structure can be more conveniently compared with those of roots of wetland plants that grow submerged in anaerobic conditions and have specialized aerenchyma cells that provides a low-resistance internal pathway for gas transport between shoot and root extremities.
Oxygen diffusion in submerged roots is influenced by anatomical, morphological and physiological characteristics, as well as environmental conditions. Roots of many wetland species contain a large volume of aerenchyma that develops a barrier (suberised exodermis) that is impermeable to radial oxygen loss. These traits (aerenchyma and barriers) act synergistically to enhance the amount of oxygen diffusing to the root apex and enable the development of an aerobic rhizosphere around the root tip, which enhances root penetration into anaerobic substrates (Colmer, 2003). However, the
192 characteristics of wetland roots allowing internal aeration may conflict with those for nutrient acquisition because oxygen is essential for most of nutrient and water uptake (Gibbs et al., 1998). The thickenings and suberisation in hypodermis lessen the loss of oxygen to the surrounding medium, permitting gas transport from the aerial part of the plant to the apical region of roots
(Armstrong et al., 2000). This is suggested to imply low permeability to nutrients and reduced oxygen loss to the surrounding medium. Lateral roots that remain permeable to oxygen are the main surface for the exchange of substances between the roots and rhizosphere. In rice or Phragmites, radial oxygen loss (ROL) is typically highest in the apical (elongating) regions and declines basipetally (Armstrong et al., 2000). It is not unusual for ROL to be virtually undetectable further than a few millimetres behind the root tip, and the greatest oxygen release to the substrate is often from basally-borne laterals.
In rice, laterals of primary root overcome the barrier of thickenings in hypodermis. The laterals lack thickenings and also lose oxygen radially which is sufficient enough to detoxify the surroundings. These laterals are also responsible for nutrient or water uptake.
Similar concepts may apply to rhizomorphs based on their morphology, anatomy, permeability patterns and their role in aeration as shown in this research. In rhizomorphs oxygen is diffusing from the base to tip. Permeability tests done in the current study on both aerial and submerged rhizomorphs confirmed that the outer cortex is relatively impermeable and the possibility that it may be involved in conserving oxygen in a similar manner to the exodermis of legume root nodules (Jacobsen et al., 1997; 1998). Suberised
193 lamellae, as found in the exodermis of roots are also shown to be impermeable to apoplastic tracers (Enstone & Peterson, 1997) as well as ions such as La3+
(Vesk et al., 2000). It is possible that the pigmented layer may similarly prevent the radial outward diffusion of oxygen travelling in the medulla from the source to the tip of the rhizomorph but with the potential to allow transfer of nutrients (taken up by peripheral hyphae) across this zone via the symplast.
Also, the apical region of the rhizomorph is highly permeable and as in roots may be losing some oxygen to the surroundings and/or absorbing nutrients or water in this elongating region. Away from the tip peripheral hyphae increase in number and length and are permeable. The peripheral hyphae may be involved in ROL and nutrient or water uptake. A basic comparison between roots and rhizomorphs zone is given in table 7.1.
Table 7.1. Comparison between the rhizomorph zone to the tissues of the roots of wetland plants in particular relation to their roles.
SIMILARITIES BETWEEN RHIZOMORPH AND ROOT ZONES Rhizomorph zone Its properties Comparable root zone Its properties Rhizomorph apex ROL/Absorption/Infection Root apex ROL/Absorption Peripheral ROL/Absorption/Infection Roots hairs/laterals ROL/Absorption hyphae Outer cortex Protection/Barrier Exodermis Protection/Barrier Storage Inner cortex Storage (gas/nutrients) Aerenchyma (gas/nutrients) Enlarged intercellular Medulla Diffusion of O2 spaces in various Diffusion of O2 tissues of wetland sp. DISSIMILARITIES BETWEEN RHIZOMORPH AND ROOT ZONE) Rhizomorph zone Its properties Comparable root zone Its properties - - Stele Water/nutrients
7.7 Oxygen Diffusion in Rhizomorphs
Apart from several structural similarities (as described above), roots of wetland plants and rhizomorphs show some mechanistic attributes, particularly in terms
194 of oxygen diffusion mechanism. Diffusion is the mechanism by which gases move within roots of all plant species (Armstrong, 1979; Beckett et al., 1988).
For efficient longitudinal oxygen transport, wetland plant roots possess certain features (Colmer, 2003) that are expressed constitutively and/or can be enhanced when plants are exposed to low concentrations of oxygen in the root zone. These are:
(1) Anatomical features: large aerenchyma lacunae, a proportionally large
cortex (meaning narrow stele), and a barrier to radial oxygen loss
(ROL) exterior to the cortical aerenchyma that diminishes O2 loss from
the aerenchyma to rhizosphere.
(2) Morphological features: Thick roots with small number of laterals, or if
present, short laterals that emerge close to the well aerated root base
rather than further down the root.
(3) Physiological traits: Roots with large amounts of aerenchyma have
lower respiration rates on a volume basis which lowers demand for O2
consumption in respiration.
(4) Environmental conditions: Cooler conditions decrease O2 consumption
along the diffusion path by slowing respiration in root tissues.
Mathematical modelling indicated that several of the anatomical and morphological features listed above act synergistically to enhance O2 diffusion to the apex of roots in anaerobic substrates (Armstrong, 1979; Armstrong and
Beckett, 1987; Armstrong, Armstrong & Beckett, 1990).
195 A similar model can be applied to rhizomorphs, where the medullary region has large air space in the centre (approx. 50% of the total cross-sectional area of a rhizomorph) surrounded by a loose aerenchyma layer of inner cortex (15% of the total cross-sectional area of a rhizomorph), which is then surrounded by a compact, variously pigmented, relatively impermeable outer cortex. This makes total porosity of the rhizomorph to be close to 65%, which is the same order as submerged roots, where root porosity can reach up to 55%).
Pigmentation present in the cortical region further prevents any loss of oxygen into the near-by vicinity making it available for the transportation to the tip region. Aerenchyma present in rhizomorphs lowers respiration rates per unit rhizomorph volume, and rhizomorph growth in moist and cool places may result in lower oxygen consumption along the diffusion path. Based on these points, a model can be proposed for oxygen flow from air pore to rhizomorph where pigmentation in cortical cells prevents oxygen loss allowing oxygen to travel down towards the tip. Hyphae borne on the rhizomorph surface overcome the barrier of compact cortical layer and may lose oxygen and accomplish the nutrient requirement of the internal tissues. A schematic of such a model is presented in Figure 7.1.
7.8 Concluding Remarks
Rhizomorphs of A. luteobubalina in axenic culture have large central gas space that transports oxygen rather than water and nutrients. As a result we may speculate that a role very different from other cord producing fungi. This specialized feature of A. luteobubalina rhizomorph makes it suitable for fungal
196 spread, pathogenicity and substrate exploitation in moist potentially anaerobic conditions, rather than for water and nutrient transportation. Rhizomorph’s role in allowing gaseous diffusion, which is very different from translocation of water or nutrients, might also explain the rapid growth of rhizomorphs, which is 56x faster than the fringing hyphae.
197 Figure 7.1. A schematic diagram of the model for the diffusion of oxygen from an air pore to the mycelium and its rhizomorphs. Arrows indicate the direction of oxygen diffusion. The grey area at the base of the rhizomorph indicates the pigmentation in the outer cortical region (o). The two double lines on outer side of the rhizomorph show the zone of peripheral hyphae (p). Oxygen diffuses through the air pore into the medullary region (m) of the rhizomorph. Some oxygen is likely to diffuse to the radially spreading mycelium above the agar surface (AM) which contains a lot of gas space and is enclosed by pseudo-parenchyma (C) and rind cells (R) at its surface. The rind and pigmented outer cortex of the rhizomorph is likely to prevent the oxygen from diffusing out and thus oxygen travels down towards the tip region. It also travels radially to the inner cortical region (i) from whence it may diffuse out to the submerged mycelium through the peripheral hyphal zone (p).
198 Chapter 8. References
Adaskaveg, J.E. & Ogawa, J.M. (1990). Wood decay pathology of fruit and nut trees in California. Plant Disease 74, 341-352.
Allaway, W.G. & Ashford, A.E. (2001). Motile tubular vacuoles in extramatrical mycelium and sheath hyphae of ectomycorrhizal systems. Protoplasma 215, 218-225.
Aloni, R., Enstone, D.E. & Petertson, C.A. (1998). Indirect evidence for bulk water flow in root cortical cell walls of three dicotyledonous species. Planta 207, 1-7.
Altus, D.P. & Canny, M.J. (1985). Loading of assimilates in wheat leaves. II. The path from chloroplast to vein. Plant, Cell and Environment 8, 275-285.
Anderson, J.B. & Ullrich, R.C. (1982). Translocation in rhizomorphs of Armillaria mellea. Experimental Mycology 6, 31-40.
Armstrong, W. (1979). Aeration in higher plants. Advances in Botanical Research 7, 225-332.
Armstrong, W., Armstrong, J. & Beckett, P.M. (1990). Measurement and modelling of oxygen release from roots of Phragmites australis. In The Use of Constructed Wetlands in Water Pollution Control [P. Cooper and B.C. Finklater (Eds)] pp. 41-54. Pergamon Press, UK.
Armstrong, W., Cousins, D., Armstrong, J., Turner, D.W. & Beckett, P.M. (2000). Oxygen distribution in wetland plant roots and permeability barriers to
199 gas-exchange with the rhizosphere: a microelectrode and modelling study with Phragmites australis. Annals of Botany 86, 687-703.
Armstrong, W. & Beckett, P.M. (1987). Internal aeration and the devdelopment of stelar anoxia in submerged roots. A multishelled mathematical model combining axial diffusion of oxygen in the cortex with radial losses to the stele, the wall layers and the rhizosphere. New Phytologist 105, 221-245.
Ashford, A.E., Allaway, W.G., Peterson, C.A. & Cairney, J.W.G. (1989). Nutrient transfer and the fungus-root interface. Australian Journal of Plant Physiology 16, 85-97.
Ashford, A.E., Peterson, C.A., Carpenter, J.L., Cairney, J.W.G. & Allaway, W.G. (1988). Structure and permeability of the fungal sheath in the Pisonia mycorrhiza. Protoplasma 147, 149-161.
Barnabas, A.D. (1994). Apoplastic and symplastic pathways in leaves and roots of the seagrass Halodule uninervis (Forssk.) Aschers. Aquatic Botany 47, 155- 174.
Baron-Epel, O., Gharyal, P.K. & Schindler, M. (1988). Pectins as mediators of wall porosity in soybean cells. Planta 175, 389-395.
Beckett, P.M., Armstrong, W., Justin, S.H.F.W. & Armstrong, J. (1988). On the relative importance of convective and diffusive gas-flows in plant aeration. New Phytologist 110, 463-468.
Bell, A.A. & Wheeler, M.H. (1986). Biosynthesis and functions of fungal melanins. Annual Review of Phytopathology 24, 411-451.
200 Bending, G.D. & Read, D.J. (1995). The structure and function of the vegetative mycelium of ectomycorrhizal plants. V. Foraging behaviour and translocation of nutrients from exploited litter. New Phytologist 130, 401-409.
Bird, R.B., Stewart, W.E. & Lightfoot, E.N. (2001). Transport Phenomena, 2nd Edition. Wiley, New York.
Bloomfield, B.J. & Alexander, M. (1967). Melanins and resistance of fungi to lysis. Journal of Bacteriology 93(4), 1276-1280.
Boddy, L. (1999). Saprotrophic cord-forming fungi: meeting the challenge of heterogeneous environments. Mycologia 91(1), 13-32.
Bohrmann, B. & Kellenberger, E. (2001). Cryosubstitution of frozen biological specimens in electron microscopy: use and application as an alternative to chemical fixation. Micron 32, 11-19.
Brefeld, O. (1877). Botanische untersuchungen über Schimmelpilze. III. Heft: Basidiomyceten I. Verlag von Arthur Felix, Leipzig.
Briggs, G.E. & Robertson, R.N. (1957). Apparent free space. Annual Review of Plant Physiology 8, 11-30.
Briggs, G.E., Hope, A.B. & Robertson, R.N. (1961) Electrolytes and plant cells. Blackwell Scientific Publications, Oxford.
Brownlee, C. & Jennings, D.H. (1982a). Long Distance translocation in Serpula lacrimans: velocity estimates and the continuous monitoring of induced perturbations. Transactions of the British Mycological Society 79(1), 143-148.
201 Brownlee, C. & Jennings, D.H. (1982b). Pathway of Translocation in Serpula lacrimans. Transactions of the British Mycological Society 79(3), 401-407.
Butler, G.M. (1966). Vegetative structures. In The Fungi: An advanced treatise [G.C. Ainsworth and A.S. Sussman (Eds)] Vol. II, pp. 83-109. Academic Press, New York.
Butler, M.J., Day, A.W., Henson, J.W. & Money, N.P. (2001). Pathogenic properties of fungal melanins. Mycologia 93(1), 1-8.
Cairney, J.W.G. (1991). Rhizomorphs: Organs of exploration or exploitation? The Mycologist 5, 5-10.
Cairney, J.W.G. (1992). Translocation of solutes in ectomycorrhizal and saprotrophic rhizomorphs. Mycological Research 96(2), 135-141.
Cairney, J.W.G. (2005). Basidiomycete mycelia in forest soil: dimensions, dynamics and roles in nutrient distribution. Mycological Research 109(1), 7- 20.
Cairney, J.W.G., Jennings, D.H. & Agerer, R. (1991). The nomenclature of fungal multi-hyphal linear aggregates. Cryptogamic Botany 2/3, 246-251.
Cairney, J.W.G., Jennings, D.H., Ratcliffe, R.G. & Southon, T.E. (1988a). The physiology of basidiomycete linear organs. II. Phosphate uptake by rhizomorphs of Armillaria mellea. New Phytologist 109, 327-333.
Cairney, J.W.G., Jennings, D.H. & Veltkamp, C.J. (1988b). Structural differentiation in maturing rhizomorphs of Armillaria mellea (Tricholomatales). Nova Hedwigia 46, 1-25.
202 Cairney, J.W.G., Jennings, D.H. & Veltkamp, C.J. (1989). A scanning electron microscope study of the internal structure of mature linear mycelial organs of four basidiomycete species. Canadian Journal of Botany 67, 2266-2271.
Campbell, A.H. (1934). Zone lines in plant tissue. II. The black lines formation in Armillaria mellea (Vahl) Quél. Annals of Applied Biology 21(1), 1-22. (+3 Plates)
Canny, M.J. (1986). Sixth international congress of biorheology plenary lecture: Transport in living plants. Biorheology 23, 605-612.
Canny, M.J. (1988). Water pathways in wheat leaves. IV. The interpretation of images of a fluorescent apoplastic tracer. Australian Journal of Plant Physiology 15, 541-555.
Canny, M.J. (1990). What becomes of transpiration stream? New Phytologist 114, 341-368.
Canny, M.J. (1995). Apoplastic water and solute movement: new rules for an old space. Annual Review of Plant Physiology and Plant Molecular Biology 46, 215-236.
Canny, M.J. & McCully, M.E. (1986). Locating water-soluble vital stains in plant tissues by freeze-substitution and resin-embedding. Journal of Microscopy 142(1), 63-70.
Carpita, N., Sabularse, D., Montezinos, D. & Delmer, D.P. (1979). Determination of the pore size of cell walls of living plant cells. Science 205, 1144-1147.
203 Clarkson, D.T. (1991). Root structure and sites of ion uptake. In Plant roots. The hidden half [Y. Waisel, A. Eshel and U. Kafkafi, (Eds)] 2nd edition, pp. 483-508. Marcel Dekker, Inc., New York.
Clarkson, D.T. (1993). Roots and the delivery of solutes to the xylem. Philosphical Transactions of the Royal Society of London 341, 5-17.
Clarkson, D.T., Robards, A.W., Stephens, J.E. & Stark, M. (1987). Suberin lamellae in the hypodermis of maize (Zea mays) roots; development and factors affecting the permeability of hypodermal layers. Plant Cell and Environment 10, 83-93
Clipson, N.J.W., Cairney, J.W.G. & Jennings, D.H. (1987). The physiology of basidiomycete linear organs. I. Phosphate uptake by cords and mycelium in the laboratory and in the field. New phytologist 105, 449-457.
Cole, L., Hyde, G.J. & Ashford, A.E. (1997). Uptake and compartmentalisation of fluorescent probes by Pisolithus tinctorius hyphae: evidence for an anion transport mechanism at the tonoplast but not for fluid-phase endocytosis. Protoplasma 199, 18-29.
Cole, L., Orlovich, D.A. & Ashford, A.E. (1998). Structure, function, and motility of vacuoles in filamentous fungi. Fungal Genetics and Biology 24, 86- 100.
Colmer, T.D. (2003). Long-distance transport of gases in plants: a perspective on internal aeration and radial oxygen loss from roots. Plant, Cell and Environment 26, 17-36.
Conway, E.J. & Downey, M. (1950). An outer metabolic region of the yeast cell. Biochemical Journal 47, 347-355.
204 Cooke, R.C. & Rayner, A.D.M. (1984). Ecology of saprotrophic fungi. Longman, London.
Cruickshank, M.G., Morrison, D.J. & Punja, Z.K. (1997). Incidence of Armillaria species in precommercial thinning stumps and spread of Armillaria ostoyae to adjacent Douglas-fir trees. Canadian Journal of Forest Research 27, 481-490.
Curran, M. (1985). Gas movements in the roots of Avicennia marina (Forsk.) Vierh. Australian Journal of Plant Physiology 12, 97-108.
Curran, M., Cole, M. & Allaway, W.G. (1986). Root aeration and respiration in young mangrove plants (Avicennia marina (Forsk.) Vierh.). Journal of Experimental Botany 37(181), 1225-1233.
De Bary, A. (1887). Differentiation of the thallus. Mycelial strands. In Comparative morphology and biology of the fungi, mycetozoa and bacteria (English translation by H.E.F. Garnsey; revised by I.B. Balfour) pp. 21-29. Clarendon Press, Oxford.
Dighton, J. & Boddy, L. (1989). Role of fungi in nitrogen, phosphorus and suphur cycling in temperate forest ecosystems. In Nitrogen, phosphorus and sulphur utilization by fungi [L. Boddy, R. Marchant and D.J. Read (Ed)] pp. 269-298. Cambridge University Press.
Dowson, C.G., Rayner, A.D.M. & Boddy, L. (1986). Outgrowth patterns of mycelial cord forming basidiomycetes from and between woody resource units in soil. Journal of General Microbiology 132, 203-211.
205 Drew, M.C. (1987). Function of root tissues in nutrient and water transport. In Root development and function [P.J. Gregory, J.V. Lake and D.A. Rose (Eds)] pp. 71-101. Cambridge University Press.
Duddridge, J.A., Malibari, A. & Read, D.J. (1980). Structure and function of mycorrhizal rhizomorphs with special reference to their role in water transport. Nature 287, 834-836.
Eamus, D., Thompson, W., Cairney, J.W.G. & Jennings, D.H. (1985). Internal structure and hydraulic conductivity of basidiomycete translocating organs. Journal of Experimental Botany 36(168), 1110-1116.
Edelmann, L. (1991). Freeze-substitution and the preservation of diffusible ions. Journal of Microscopy 161(2), 217-228.
Enstone, D.E. & Peterson, C.A. (1998). Effects of exposures to humid air on epidermal visbility and suberin deposition in maize (Zea mays L.) roots. Plant Cell and Environment 21, 837-844.
Enstone, D.E. & Peterson, C.A. (1997). Maturation of the endodermis and exodermis in corn root is accelerated by exposure to air (Abstract number 66). In Session: Root Physiology. American Society of Plant Biologists.
Epstein, E. (1972). "Outer" or "free" spaces. In Mineral nutrition of plants: principles and perspectives, pp. 85-102. John Wiley and sons, Inc. New York.
Falk, S.P. & Parberry, D.G. (1995). Armillaria luteobubalina population structure in horticultural plantings in Victoria, Australia. Mycological Research 99(2), 216-220.
206 Feder, N. & Sidman, R.L. (1958). Methods and principles of fixation by freeze- substitution. Journal of Biophysics and Biochemical Cytology 4(5), 593-600.
Findlay, W.P.K. (1951). The development of Armillaria mellea rhizomorphs in a water tunnel. Transactions of the British Mycological Society 34, 146-147.
Fisher, D.B. (1972). Artifacts in the embedment of water-soluble compounds for light microscopy. Plant Physiology 49, 161-165.
Fisher, D.B. & Housley, T.L. (1972). The retention of water-soluble compounds during freeze-substitution and microautoradiography. Plant Physiology 49, 166-171.
Fitzgerald, M.A. & Allaway, W.G. (1991). Apoplastic and symplatic pathways in the leaf of the grey mangrove Avicennia marina (Forsk.) Vierh. New Phytologist 119, 217-226.
Fox, R.T.V. (2000). Armillaria root rot: Biology and control of honey fungus. Intercept Limited, Andover.
Garraway, M.O. & Evans, R.C. (1985). Fungal nutrition and physiology. John Wiley and sons, New York.
Garraway, M.O., Hüttermann, A. & Wargo, P.M. (1991) Ontogeny and Physiology. In Armillaria root disease [C.G.I. Shaw and G.A. Kile (Eds)]. Agriculture Handbook No 691, 21-47. Forest Service, USDA. Washington, DC.
Garrett, S.D. (1944). Root disease fungi. In Chronica Botanica Co. Waltham, Mass.
207 Garrett, S.D. (1953). Rhizomorph behaviour in Armillaria mellea (Vahl) Quél. I. Factors controlling rhizomorph initiation by Armillaria mellea in pure culture. Annals of botany 17(65), 63-79.
Garrett, S.D. (1956). Rhizomorph behaviour in Armillaria mellea (Vahl) Quél. II. Logistics of infection. Annals of botany 20(78), 193-209.
Garrett, S.D. (1960a). Inoculum Potential. In Plant Pathology [J.G. Horsfall and A.E. Dimond (Eds)] Vol. III, pp. 23-56. Academic Press, New York and London.
Garrett, S.D. (1960b). Rhizomorph behaviour in Armillaria mellea (Fr.) Quél. III. Saprophytic colonization of woody substrates in soil. Annals of botany 24(94), 275-285.
Garrett, S.D. (1963). Form and function in fungi. In Soil fungi and soil fertility. An Introduction to soil mycology. Vol. 2, pp. 43-63. Pergamon press, Oxford.
Garrett, S.D. (1970). Specialized parasites: ectotrophic root-infecting fungi. In Pathogenic root infecting fungi, pp. 80-142. Cambridge University Press.
Gibbs, J., Turner, D.W., Armstrong, W., Darwent, M.J. & Greenway, H. (1998). Response to oxygen deficiency in primary maize roots. I. Development of oxygen deficiency in the stele reduces radial solute transport to the xylem. Australian Journal of Plant Physiology 25, 745-758.
Goffart, J. (1903-4). Contribution à l’étude du rhizomorphe de l’Armillaria mellea Vahl. Mémoires couronnés et Mémoires des Savants étrangers publiés par l’Académie Royale des Sciences, des Lettres et des Beaux-Arts de Belgique 62, 3-29.
208 Granlund, H.I., Jennings, D.H. & Thompson, W. (1985). Translocation of solutes along rhizomorphs of Armillaria mellea. Transactions of the British Mycological Society 84(1), 111-119.
Granlund, H.I., Jennings, D.H. & Veltkamp, K. (1984). Scanning electron microscope studies of rhizomorphs of Armillaria mellea. Nova Hedwigia 39, 85-100.
Gray, S.N., Dighton, J. & Jennings, D.H. (1996). The physiology of basidiomycete linear organs. III. Uptake and translocation of radiocesium within differentiated mycelia of Armillaria spp. growing in microcosms and in the field. New Phytologist 132, 471-482.
Gregory, S.C. (1989). Armillaria species in northern Britain. Plant Pathology 38, 93-97.
Guillaumin, J-J. & Berthelay, S. (1981). Détermination spécifique des armillaires par la méthode des groupes de compatabilité sexuelle. Spécialisation écologique des espèces françaises. Agronomie 1(10), 897-908.
Guinel, F.C. & McCully, M.E. (1986). Some water-related physical properties of maize root-cap mucilage. Plant Cell and Environment 9, 657-666.
Gunning, B.E.S. & Steer, M.W. (1975). The outer surface of the cell - cell wall and plasma membrane. In Ultrastructure and the Biology of Plant Cells, pp. 15-32. Edward Arnold, London.
Harley, J.L. & Smith, S.E. (1983). Mycorrhizal Symbiosis. Academic Press, London.
209 Hartig, R. (1870). Das auftreten der rhizomorpha in nadelholzkulturen. Zeitschrift für Forst und Jagdwesen 2, 359-361.
Hartig, R. (1874). Agaricus (Armillaria) melleus L. In Wichtige krankheiten der waldbäume. Beiträge zur mycologie und phytopathologie für botaniker und forstmänner. Verlag von Julius Springer, Monbijouplatz 3, Berlin. [In German] (English translation by M. William, D.H. Lambert and W. Liese, 1974. Important diseases of forest trees. Contributions to mycology and phytopathlogy for botanists and foresters. Phytopathological Classics No. 12, 12-41. The American Phytopathological Society: St Paul, Minnesota).
Harvey, D.M.R. (1982). Freeze-substitution. Journal of Microscopy 127(2), 209-221.
Harvey, D.M.R., Hall, J.L. & Flowers, T.J. (1976). The use of freeze- substitution in the preparation of plant tissue for ion localization studies. Journal of Microscopy 107(2), 189-198.
Haugland, R.P. (2002). Handbook of Fluorescent Probes and Research Chemicals, 9th ed. Molecular Probes Inc., Eugene, Oregon.
Henson, J.M., Butler, M.J. & Day, A.W. (1999). The dark side of the mycelium: melanins of phytopathogenic fungi. Annual Review of Phytopathology 37, 447-471
Hoch, H.C. (1986) Freeze substitution of fungi. In Ultrastructure Techniques for Microorganisms, pp. 183-212. (H.C. Aldrich and W.J. Todd (Eds)]. Plenum Publishing Corp.
Hoch, H.C. & Howard, R.J. (1980). Ultrastructure of freeze-substituted hyphae of the basidiomycete Laetisaria arvalis. Protoplasma 103, 281-297.
210 Hovenden, M.J. & Allaway, W.G. (1994). Horizontal structures on pneumatophores of Avicennia marina (Forsk.) Vierh. – A new site of oxygen conductance. Annals of Botany, 73, 377-383.
Howard, R.J. & Aist, J.R. (1979). Hyphal tip cell ultrastructure of the fungus Fusarium: Improved preservation by freeze-substitution. Journal of Ultrastructure Research 66, 224-234.
Howard, R.J. & Ferrari, M.A. (1989). Role of melanin in appressorium function. Experimental Mycology 13, 403-418.
Howard, R.J. & O'Donnell, K.L. (1987). Freeze substitution of fungi for cytological analysis. Experimental Mycology 11, 250-269.
Intini, M.G. (1987). Le rizomorfe di Armillaria obscura al microscopio elettronico a scansione. (Scanning electron microscope study of Armillaria obscura). Micologia Italiana 2, 40-48.
Jacobsen, K.R., Shimada, S., Rousseau, R.A. & Denison, R.F. (1998). Tracing the path of O2 into birdsfoot trefoil and alfalfa nodules using iodine vapor. Bot Acta 111, 193-203.
Jacobsen, K.R., Shimada, S., Rousseau, R.A. & Denison, R.F. (1997). Suberized cell layers restrict oxygen diffusion into legume root nodules. Abstract number 113. In session: Plant Ineractions with other Organisms American Society of Plant Biologists.
Jacques-Félix, M. (1968). Recherches sur la morphogenèse des rhizomorphes et des télépodes en culture pure. Bulletin Trimestrid de la Société Mycologique de France 84, 166-307.
211 Jennings, D.H. (1976). Transport and translocation in filamentous fungi. In The filamentous fungi: Biosynthesis and metabolism [J.E. Smith and D.R. Berry (Eds)] Vol. 2, pp. 32-63. Edward Arnold.
Jennings, D.H. (1987). Translocation of solutes in fungi. Biological Review 62, 215-243.
Jennings, D.H. (1991). The role of droplets in helping to maintain a constant growth rate of aerial hyphae. Mycological Research 95(7), 883-884.
Jennings, D.H. (1995). Nutrient movement within the colony. In The physiology of fungal nutrition, pp. 447-467. Cambridge University Press.
Kellenberger, E. (1987). The response of biological macromolecules and supramolecular structures to the physics of specimen cryopreparation. In Cryotechniques in Biological Electron Microscopy, pp 35-63. [R.A. Steinbrecht and K. Zeirold (Eds)]. Spriger-Verlag, Berlin.
Kellenberger, E. (1991) The potential of cryofixation and freeze substitution: observations and theoretical considerations. Journal of Micros. 161, 183-203.
Kile, G.A. (1981). Armillaria luteobubalina: a primary cause of decline and death of trees in mixed species eucalypt forests in central Victoria. Australian Forestry Research 11, 63-77.
Kile, G.A. (1983a). Armillaria root rot in eucalypt forests: Aggravated endemic disease. Pacific Science 37(4), 459-464.
Kile, G.A. (1983b). Identification of genotypes and the clonal development of Armillaria luteobubalina Watling & Kile in eucalypt forests. Australian Journal of Botany 31, 657-671.
212 Kile, G.A., McDonald, G.I. & Byler, J.W. (1991). Ecology and disease in natural forests. In Armillaria root disease [C.G.I. Shaw and G.A. Kile (Eds)]. Agriculture Handbook No 691, 102-121. Forest Service, USDA. Washington, DC.
Kile, G.A. & Watling, R. (1981). An expanded concept of Armillaria luteobubalina. Transactions of the British Mycological Society 77(1), 75-83.
Kile, G.A. & Watling, R. (1983). Armillaria species from south-eastern Australia. Transactions of the British Mycological Society 81(1), 129-40.
Kile, G.A. & Watling, R. (1988). Identification and occurrence of Australian Armillaria species, including A. pallidula sp. nov. and comparative studies between them and non-Australian tropical and Indian Armillaria. Transactions of the British Mycological Society 91(2), 305-315.
Kile, G.A., Watling, R., Malajczuk, N. & Shearer, B.L. (1983). Occurence of Armillaria luteobubalina Watling and Kile in Western Australia. Australasian Plant Pathology Society 12, 18-20.
Kuhn, P.J. & Trinci, A.P.J. (1990). Cell walls and membranes in fungi - an introduction. In Biochemistry of cell walls and membranes in fungi [P.J. Kuhn, A.P.J. Trinci, M.J. Jung, M.W. Goosey and L.G. Copping (Eds)] pp. 1-4. Spriger-Verlag. Heidlberg.
Kuo, M-J. & Alexander, M. (1967). Inhibition of the lysis of fungi by melanins. Journal of Bacteriology 94(3), 624-629.
Läuchli, A. (1976). Apoplasmic transport in tissues. In Transport in plants II. Part B: Tissues and Organs [U. Lüttge and M. G. Pitman (Eds)] pp. 3-34. Springer-Verlag, Berlin.
213 Lopez-Real, J.M. (1975). Formation of pseudosclerotia ('zone lines') in wood decayed by Armillaria mellea and Stereum hirsutum: I. Morphological aspects. Transactions of the British Mycological Society 64(3), 465-471.
Lopez-Real, J.M. & Swift, M.J. (1975). The formation of pseudosclerotia ('zone lines') in wood decayed by Armillaria mellea and Stereum hirsutum: II. Formation in relation to the moisture content of the wood. Transactions of the British Mycological Society 64(3), 473-481.
Lopez-Real, J.M. & Swift, M.J. (1977). Formation of pseudosclerotia ('zone lines') in wood decayed by Armillaria mellea and Stereum hirsutum: III. Formation in relation to composition of gaseous atmosphere in wood. Transactions of the British Mycological Society 68(3), 321-325.
Lüttge, U. & Higinbotham, N. (1979). Transport in plants. Springer-Verlag, New York.
Maeda, H. & Ishida, N. (1967). Specificity of binding of hexopyranosyl polysaccharides with fluorescent brightener. Journal of Biochemistry 62(2), 276-278.
Mallett, K.I. & Colotelo, N. (1984). Rhizomorph exudate of Armillaria mellea. Canadian Journal of Microbiology 30, 1247-1252.
Marshall, A.T. (1980) Freeze-substitution as a preparation technique for biological X-ray microanalysis. Scanning Electron Microscopy Part II, 395- 408.
McCully, M.E. & Canny, M.J. (1989). Pathways and processes of water and nutrient movement in roots. In Structural and functional aspects of transport in roots (Third International Symposium on "Structure and Function of Roots".
214 Nitra Czechoslovakia, 3-7 August, 1987) [B.C. Loughman, O. Gašparíková and J. Kolek (Eds)] Vol. 36, pp. 3-14. Kluwer Academic Publishers. Dordrecht.
McCully, M.E. & Boyer, J.S. (1997). The expansion of maize root-cap mucilage during hydration. 3. Changes in water potential and water content. Physiologia Plantarum 99, 169-177.
McIntosh, J.R. (2001). Electron microscopy of cells: a new beginning for a new century. The Journal of Cell Biology 153(6), F25-F32.
Metzler, B. & Hecht, U. (2004) Three-dimensional structure of tubular air channels formed by Armillaria spp. in water-saturated logs of silver fir and Norway spruce. Canadian Journal of Botany 82, 1338-1345.
Monaghan, P., Perusinghe, N. & Müller, M. (1998). High-pressure freezing for immunocytochemistry. Journal of microscopy 192(3), 248-258.
Moody, A.R. & Weinhold, A.R. (1972). Stimulation of rhizomorph production by Armillaria mellea with lipid from tree roots. Phytopathology 62, 1347-1350.
Moon, G.J., Clough, B.F., Peterson, C.A. & Allaway, W.G. (1986). Apoplastic and symplastic pathways in Avicennia marina (Forsk.) Vierh. Roots revealed by fluorescent tracer dyes. Australian Journal of Plant Physiology 13, 637-648.
Moreshet, S., Huang, B. & Huck, M.G. (1991). Water permeability of roots. In Plant roots. The hidden half (2nd Edition) [Y. Waisel, A. Eshel and U. Kafkafi, (Eds)] pp. 659-678. Marcel Dekker, Inc., New York.
Morgan, A.J., Davies, T.W. & Erasmus, D.A. (1978). Specimen preparation. In Electron probe microanalysis in biology [D.A. Erasmus (Ed.)] pp. 94-147. John Wiley and sons, New York.
215 Morrison, D.J. (1975). Ion uptake by rhizomorphs of Armillaria mellea. Canadian Journal of Botany 53, 48-51.
Morrison, D.J. (1976). Vertical distribution of Armillaria mellea rhizomorphs in soil. Transactions of the British Mycological Society 66(3), 393-399.
Morrison, D.J. (1982). Effects of soil organic matter on rhizomorph growth by Armillaria mellea. Transactions of British Mycological Society 78(2), 201-207.
Morrison, D.J. (1988). Pathogenicity of Armillaria species is related to rhizomorph growth habit. In Root and Butt Rots, 7th International Conference Papers, 584-589.
Morrison, D.J. & Mallett, K. (1996). Silvicultural management of armillaria root disease in western Canadian forests. Canadian Journal of Plant Pathology 18, 194-199.
Motta, J.J. (1969). Cytology and morphogenesis in the rhizomorph of Armillaria mellea. American Journal of Botany 56(6), 610-619.
Motta, J.J. (1971). Histochemistry of the rhizomorph meristem of Armillaria mellea. American Journal of Botany 58(1), 80-87.
Motta, J.J. & Peabody, D.C. (1982). Rhizomorph cytology and morphogenesis in Armillaria tabescens. Mycologia 74, 671-674.
Münch, E. (1909). Untersuchungen über immunität und krankheitsempfänglichkeit der holzpflanzen. Naturwissenschaftliche Zeitschrift für Forst- und Landwirtschaft 7, 129-160.
216 Mwenje, E. & Ride, J.P. (1996). Morphological and biochemical characterization of Armillaria isolates from Zimbabwe. Plant Pathology 45, 1036-1051.
O'Brien, T.P. & McCully, M.E. (1981). The study of plant structure. Principles and selected methods. Termarcarphi Pty Ltd, Melbourne.
O'Brien, T.P., Feder, N. & McCully, M.E. (1964). Polychromatic staining of plant cell walls by Toluidine blue O. Protoplasma 59, 368-373.
Ono, K. (1965). Armillaria root rot in plantations of Hokkaido. Effects of topography and soil conditions on its occurece. (In Japanese). Bulletin of the Goverment Forest Experimental Station, Meguro 179, 1-62 (+5 Plates).
Ono, K. (1970). Effect of soil conditions on the occurrence of Armillaria root rot of Japanese Larch. (In Japanese). Bulletin of the Government Forest Experiment Station 229, 123-219 (+ 9 Plates).
Orlovich, D.A. & Ashford, A.E. (1995). X-ray microanalysis of ion distribution in frozen salt/dextran droplets after freeze-substitution and embedding in anhydrous conditions. Journal of Microscopy 180(2), 117-126.
Otieno, W., Sierra, A.P. & Termorshuizen, A. (2003). Characterization of Armillaria isolates from tea (Camellia sinensis) in Kenya. Mycologia 95, 160- 175.
Pålsgård, E., Lindh, U. & Roomans, G.M. (1994) Comparative study of freeze- substitution techniques for X-ray microanalysis of biological tissue. Microscopy Research and Technique 28, 254-258.
217 Papendick, R.I. & Mulla, D.J. (1986). Basic principles of cell and tissue water relations. In Water, fungi and plants [P.G. Ayers and L. Boddy (Eds)] pp. 1-25. Cambridge University Press.
Pareek, M., Cole, L. & Ashford, A.E. (2001). Variation in structure of aerial and submerged rhizomorphs of Armillaria luteobubalina indicate that they may be organs of absorption. Mycological Research 105(11), 1377-1387
Pearce, M.H. & Malajczuk, N. (1990a). Factors affecting growth of Armillaria luteobubalina rhizomorphs in soil. Mycological Research 94(1), 38-48.
Pearce, M.H. & Malajczuk, N. (1990b). Stump colonisation by Armillaria luteobubalina and other wood decay fungi in an age series of cut-over stumps in karri (Eucalyptus diversicolor) regrowth forests in south western Australia. New Phytologist 115, 129-138.
Pearce, M.H., Malajczuk, N. & Kile, G.A. (1986). The occurrence and effects of Armillaria luteobubalina in the karri (Eucalyptus diversicolor F. Muell.) forests of Western Australia. Australian Forests Research 16, 243-259.
Peberdy, J.F. (1990). Fungal cell walls - a review. In Biochemistry of cell walls and membranes in fungi [P.J. Kuhn, A.P.J. Trinci, M.J. Jung, M.W. Goosey and L.G. Copping (Eds)] pp. 1-4. Springer-Verlag. Heidlberg.
Peterson, C.A. & Edington, L.V. (1976). Entry of pesticides into the plant symplast as measured by their loss from an ambient solution. Pesticide Science 7, 483-491.
Peterson, C.A., Emanuel, M.E. & Humphreys, G.B. (1981). Pathway of movement of apoplastic fluorescent dye tracers through the endodermis at the
218 site of secondary root formation in corn (Zea mays) and broad bean (Vicia faba). Canadian Journal of Botany 59, 618-625.
Podger, F.D., Kile, G.A., Watling, R. & Fryer, J. (1978). Spread and effects of Armillaria luteobubalina sp. nov. in an Australian Eucalyptus regnans plantation. Transactions of the British Mycological Society 71(1), 77-87.
Radin, J.W. & Mathews, M.A. (1989). Water transport properties of cortical cells in roots of nitrogen- and phosphorous-deficient cotton seedlings. Plant Physiology 89, 264-268.
Ramsbottom, J. (1953). Mushrooms and Toadstools. London: Collins.
Rast, D.M. & Hollenstein, G.O. (1977). Architecture of the Agaricus bisporus spore wall. Canadian Journal of Botany 55, 2251-2262.
Rayner, A.D.M., Powell, K.A., Thompson, W. & Jennings, D.H. (1985). Morphogenesis of vegetative organs. In Developmental biology of higher fungi [D. Moore, L.A. Casselton, D.A. Wood and J.C. Frankland (Eds)] pp. 249-279. Cambridge University Press.
Rayner, A.D.M. & Todd, N.K. (1979). Population and community structure and dynamics of fungi in decaying wood. Advances in Botanical Research 7, 333-420.
Read, D.B., Gregory, P.J. & Bell, A.E. (1999). Physical properties of axenic maize root mucilage. Plant and Soil 211, 87-91.
Reaves, J.L., Shaw III, C.G. & Roth, L.F. (1993). Infection of ponderosa pine trees by Armillaria ostoyae: residual inoculum versus contagion. Northwest Science 67(3), 156-162.
219 Rebhun, L.I. (1965). Freeze-substitution: Fine structure as a function of water concentration in cells. Federation Proceedings 24 (suppl. 15), 217-232.
Rebhun, L.I. (1972). Freeze-substitution and freeze drying. In Principles and techniques of electron microscopy. Vol. 2, 3-49. [M.A. Hayat (Ed.)]. Van Nostrand-Reinhold, New York.
Redfern, D.B. (1968). The ecology of Armillaria mellea in Britain: Biological control. Annals of Botany 32, 293-300.
Redfern, D.B. (1970). The ecology of Armillaria mellea: rhizomorph growth through soil. In Root disease and soil borne pathogen: Proccedings of the symposium, July, 1968, Imperial College, London, pp. 147-149. [T.A. Toussoun, R.V. Bega and P.E. Nelson (Eds)]. University of California Press, Berkeley.
Redfern, D.B. (1973). Growth and behaviour of Armillaria mellea rhizomorphs in soil. Transactions of the British Mycological Society 61(3), 569-581.
Redfern, D.B. & Filip, G.M. (1991). Inoculum and infection. In Armillaria root disease [C.G.I. Shaw and G.A. Kile (Eds)]. Agriculture Handbook No 691, 48- 61. Forest Service, USDA. Washington, DC.
Rees, B.J., Shepherd, V.A. & Ashford, A.E. (1994). Presence of a motile tubular vacuole system in different phyla of fungi. Mycological Research 98, 985-992.
Reitsma, J. (1932). Studien über Armillaria mellea (Vahl) Quél. Phytopathologisches Zeitschrift 4, 461-522.
220 Rhoads, A.S. (1956). The occurence and destructiveness of Clitocybe root rot of woody plants in Florida. Lloydia 19(4), 193-240.
Rishbeth, J. (1968). The growth rate of Armillaria mellea. Transactions of the British Mycological Society 51(3&4), 575-586.
Rishbeth, J. (1978). Effects of soil temperature and atmosphere on growth of Armillaria rhizomorphs. Transactions of the British Mycological Society 70(2), 213-220.
Rishbeth, J. (1982). Species of Armillaria in southern England. Plant Pathology 31, 9-17.
Rizzo, D.M. & Harrington, T.C. (1993). Delineation and biology of clones of Armillaria ostoyae, A. gemina and A. calvescens. Mycologia 85(2), 164-174.
Roberson, R.W. & Fuller, M.S. (1988). Ultrastructural aspects of the hyphal tip of Sclerotium rolfsii preserved by freeze substitution. Protoplasma 146, 143- 149.
Roll-Hansen, F. (1985). The Armillaria species in Europe: A literature review. European Journal of Forest Pathology 15, 22-31.
Rotman, B. & Papermaster, B.W. (1966). Membrane properties of living mammalian cells as studied by enzymatic hydrolysis of florigenic esters. Biochemistry 55, 134-141.
Rykowski, K. (1984). Niektore troficzne uwarunkowania patogenicznosci Armillaria mellea (Vahl) Quél). w uprawach sosnowych. [Some trophic factors in the pathogenicity of Armillaria mellea in Scots pine plantations.] (In Polish.). Prace Instytutu Badawczego Lesnictwa 640, 1-140.
221 Salema, R. & Brandão, I. (1973). The use of PIPES buffer in the fixation of plant cells for electron microscopy. Journal of Submicroscopic Cytology 5, 79- 96.
Sattelmacher, B. (2001). The apoplast and its significance for plant mineral nutrition. New Phytologist 149, 167-192.
Schlafke, E. (1959). Kritische untersuchungen zur wanderung von fluorochromen in Blättern. Planta 50, 388-422.
Schmid, R. & Liese, W. (1970). Feinstruktur der rhizomorphen von Armillaria mellea. Phytopathologische Zeitschrift 68, 221-231.
Schütte, K.H. (1956). Translocation in the fungi. New Phytologist 55, 164-182.
Shaw III, C.G. (1985). In vitro responses of different Armillaria taxa to gallic acid, tannic acid and ethanol. Plant Pathology 34, 594-602.
Shaw III, C.G. & Kile, G.A. (1991). Armillaria root disease. Agriculture Handbook No 691. Forest Service, USDA. Washington, DC.
Shearer, B.L., Byrne, A., Dillon, M. & Buehrig, R. (1997a). Distribution of Armillaria luteobubalina and its impact on community diversity and structure in Eucalyptus wandoo woodland of southern western Australia. Australian Journal of Botany 44, 151-165.
Shearer, B.L., Crane, C.E., Fairman, R.G. & Grant, M.J. (1997b). Occurrence of Armillaria luteobubalina and pathogen-mediated changes in coastal dune vegetation of south-western Australia. Australian Journal of Botany 45, 905- 907.
222 Shearer, B.L., Crane, C.E., Fairman, R.G. & Grant, M.J. (1998). Susceptibility of plant species in coastal dune vegetation of south-western Australia to killing by Armillaria luteobubalina. Australian Journal of Botany 46, 321-334.
Shearer, B.L. & Tippett, J.T. (1988). Distribution and impact of Armillaria luteobubalina in the Eucalyptus marginata forest of south-western Australia. Australian Journal of Botany 36, 433-445.
Shepherd, V.A., Orlovich, D.A. & Ashford, A.E. (1993). A dynamic continuum of pleiomorphic tubules and vacuoles in growing hyphae of a fungus. Journal of Cell Science 104, 495-507.
Singh, P. (1975). Armillaria root rot: Distribution and severity in softwood plantations in Newfoundland. Acta Phytopathologica Academiae Scientiarum Hungaricae 10, 389-406.
Singh, P. (1981). Armillaria mellea: Growth and distribution of rhizomorphs in the forest soils of Newfoundland. European Journal of Forest Pathology 11, 208-220.
Smith, A.M. & Griffin, D.M. (1971). Oxygen and the ecology of Armillariella elegans Heim. Australian Journal of Biological Science 24, 231-262.
Smith, M.L., Bruhn, J.N. & Anderson, J.B. (1992). The fungus Armillaria bulbosa is among the largest and oldest living organisms. Nature 356, 428-431.
Snider, P.J. (1959). Stages of development in rhizomorphic thalli of Armillaria mellea. Mycologia 51, 693-707.
223 Stanosz, G.R. & Patton, R.F. (1991). Quantification of Armillaria rhizomorphs in Wisconsin aspen sucker stands. European journal of Forest Pathology 21, 5- 16.
Stanosz, G.R., Patton, R.F. & Spear, R.N. (1987). Structure of rhizomorphs from Wisconsin aspen stands. Canadian Journal of Botany 65, 2124-2127.
Steinbrecht, R.A. & Müller, M. (1987). Freeze-substitution and freeze drying. In Cryotechniques in biological microscopy [R.A. Steinbrecht and K. Zeirold (Eds)] pp. 149-172 Springer-Verlag, Berlin.
Stelzer, R. & Lehmann, H. (1993). Recent developments in electron microscopical techniques for studying ion localization in plants cells. Kluwer, Amsterdam.
Steudle, E. (1997). Water transport across plant tissue: Role of water channels. Biology of the Cell 89, 259-273.
Steudle, E. (2000). Water uptake by roots: effects of water deficit. Journal of Experimental Botany 51(350), 1531-1542.
Steudle, E. & Peterson, C.A. (1998). How does water get through roots? Journal of Experimental Botany 49(322), 775-788.
Stewart, A. & Deacon, J.W. (1995). Vital fluorochromes as tracers for fungal growth studies. Biotechnic and Histochemistry 70(2), 57-67.
Strugger, S. (1939). Studien über den transpirationsstrom im blatt von Secale cereale und Triticum vulgare. Zeitschrift furBotanik 35, 97-113.
224 Swift, M.J. (1968). Inhibition of rhizomorph development by Armillaria mellea in Rhodesian forest soils. Transactions of the British Mycological Society 51(2), 241-247.
Thomas, H.E. (1934). Studies on Armillaria mellea (Vahl) Quel., infection, parasitism, and host resistance. Journal of Agricultural Research 48(3), 187- 218.
Thompson, W. (1984). Distribution, development and functioning of mycelial cord systems of decomposer basidiomycetes of the deciduous woodland floor. In The ecology and physiology of the fungal mycelium [D.H. Jennings and A. D.M. Rayner (Eds)] pp. 185-214. Cambridge University Press.
Tlalka, M., Watkinson, S.C., Darrah, P.R. & Fricker, M.D. (2002). Continuous imaging of amino-acid translocation in intact mycelia of Phanerochaete velutina reveals rapid, pulsatile fluxes. New Phytologist 153, 173-184.
Townsend, B.B. (1954). Morphology and development of fungal rhizomorphs. Transactions of the British Mycological Society 37, 222-233.
Twery, M.J., Mason, G.N., Wargo, P.M. & Gottschalk, K.W. (1990). Abundance and distribution of rhizomorphs of Armillaria spp. in defoliated mixed oak stands in western Maryland. Canadian Journal of Forest Research 20, 674-678.
Unestam, T. & Sun Y.-P. (1995). Extramatrical structures of hydrophobic and hydrophilic ectomycorrhizal fungi. Mycorrhiza 5, 301-311.
Van Harreveld, A., Crowell, J. & Malhotra, S.K. (1965). A study of extracellular space in central nervous tissue by freeze-substitution. The Journal of Cell Biology 25, 117-137.
225 Varney, G.T., McCully, M.E. & Canny, M.J. (1993). Sites of entry of water into the symplast of maize roots. New Phytologist 125, 733-741.
Vesk, P.A., Ashford, A.E., Markovina, A-L. & Allaway, W.G. (2000). Apoplasmic barriers and their significance in the exodermis and sheath of Eucalyptus pilularis-Pisolithus tintorius ectomycorrhizas. New Phytologist 145, 333-346.
Watkinson, S.C. (1979). Growth of rhizomorphs, mycelial strands, coremia and sclerotia. In Symposium of the British Mycological Society. Fungal walls and hyphal growth, pp. 93-113. British Mycological society, Cambridge, Queen Elizabeth College, London.
Watkinson, S.C., Boddy, L., Burton, K., Darrah, P.R., Eastwood, D., Fricker, M.D. & Tlalka, M. (2005). New approaches to investigating the function of mycelial networks. Mycologist 19, 11-17.
Watling, R., Kile, G.A. & Burdsall, Jr. H.H. (1991). Nomenclature, taxonomy, and identification. In Armillaria root disease [C.G.I. Shaw and G.A. Kile (Eds)]. Agriculture Handbook No 691, 1-9. Forest Service, USDA. Washington, DC.
Whiting, E.C. & Rizzo, D.M. (1999). Effect of water potential on radial colony growth of Armillaria mellea and A. gallica isolates in culture. Mycologia 91(4), 627-635.
Willetts, H.J. (1971). The survival of fungal sclerotia under adverse environmental conditions. Biological Reviews 46, 387-407.
Willetts, H.J. (1972). The morphogenesis and possible evolutionary origins of fungal sclerotia. Biological Reviews 47, 515-536.
226 Willetts, H.J. (1978). Sclerotium formation. In The Filamentous Fungi, vol. 3, Developmental Mycology (J.E. Smith & D.R. Berry, eds): 197-213. Edward Arnold: London.
Willetts, H.J. & Bullock, S. (1992). Developmental biology of sclerotia. Mycological Research 96, 801-816.
Wolkinger, F., Plank, S. & Brunegger, A. (1975). Rasterelektronenmikroskopische untersuchungen an rhizomorphen von Armillaria mellea. Phytopathologische Zeitschrift 84, 352-359.
Worrall, J.J., Chet, I. & Hüttermann, A. (1986). Association of rhizomorph formation with laccase activity in Armillaria spp. Journal of General Microbiology 132, 2527-2533.
Young, N. & Ashford, A.E. (1992). Changes during development in the permeability of sclerotia of Sclerotinia minor to an apoplastic tracer. Protoplasma 167, 205-214.
Young, N. & Ashford, A.E. (1995). Apoplastic permeability of sclerotia of Sclerotium rolfsii, Sclerotium ceviporum and Rhizoctonia solani. New Phytologist 131, 33-40.
Young, N. & Ashford, A.E. (1996). The effects of rind damage and regeneration on permeability of the apoplast in sclerotia of Sclerotinia minor. New Phytologist 134, 13-24.
Zimmermann, M.H. (1971). Transport in the phloem. In Trees: structure and function [M.H. Zimmermann and C.L. Brown (Eds)] pp. 221-279. Springer- Verlag, New York.
227 Zimmermann, H.M. & Steudle, E. (1998). Apoplastic transport across young maize roots: effect of exodermis. Planta 206, 7-19.
228