TAXONOMY OF MYCOHERBICIDE SPECIES: Fusarium & Pleospora

On this page, we address the difficult issues of the taxonomy of the mycoherbicides Fusarium and Pleospora. This classification is based upon the physical, chemical, biological, and microbiological characteristics of a the fungus under examination.  We will work almost exclusively here with the genus Fusarium, but hope to add information on Pleospora soon.

"Taxonomy of the genus Fusarium is complex and difficult to apply, partly because of the use of different taxonomic systems in different countries. The major taxonomic systems currently in use are discussed in a companion volume, "Fusarium Species: An Illustrated Manual for Identification" by Nelson et at. (1983). The taxonomy of the genus is further complicated by the extreme variability of Fusarium species in culture and the fact that they mutate and degenerate rapidly, particularly under conditions of repeated subculturing on common laboratory media. This situation has led to great confusion in the extensive literature on Fusarium mycotoxicology because the same fungus is known under a variety of different names, because different fungi are lumped together under names such as F. tricinctum Corda emend. Snyd. & Hans. and F. roseum Lk. emend. Snyd. & Hans., because several Fusarium toxins have been named for mis-identified producing species, because elaborate chemical and pathological studies have been reported in the literature and attributed to incorrectly named species, and because many toxigenic Fusarium strains have become degenerate and lost their toxigenic ability due to maltreatment in laboratories that do not specialize in the maintenance of Fusarium cultures. Consequently it has become impossible to relate toxicological studies done in different laboratories to each other because of the existing confusion with regard to the taxonomy and nomenclature of toxigenic Fusarium species (Smalley et al., 1977)." Toxigenic Fusarium Species by Marasas et alia, Penn State U, 1984

Below is a superb, but dated overview:

The Applied Mycology of Fusarium  Moss & Smith, 1984, Cambridge University Press.

Chapter 1: 

The Fusarium problem: historical, economic and taxonomic aspects
C. BOOTH
Commonwealth Mycological institute, Ferry Lane, Kent, Surrey TW9 3A F, U.K.

Historical aspects of the Fusarium problem.

When did our knowledge of Fusarium begin?  There is a tendency to regard it as all very new, in fact I recently read a thesis in which it was claimed that our knowledge of Fusarium, diseases of cereals began with a paper published by Bennett in 1928.

I can assure you that it began much earlier; precisely when depends upon what you mean by the Fusarium problem. One of the first written descriptions of ear rot of maize caused by F. moniliforme was described from native Aztec descriptions in the sixteenth century by a Franciscan friar in Mexico. When the plant pathologists first went overseas to look at peasant agriculture they found the farmers had local names for the diseases that occurred in their crops, and these names obviously went back for generations, usually being translated as red mould, white mould etc.

If we mean when were Fusarium diseases first known within the concept of modern terminology in a scientific system, then of course the date is 1809, when Link first described the genus. He described Fusarium roseum as the first species; unfortunately his collection was mixed and it has been used as a mixed-up name ever since, it is however surprising in such a supposedly unstable genus and in an era when we, as taxonomists, are constantly accused of changing names, how many of the original names have survived unchanged: F. lateritium Nees described 1817; F. Heterosporum Nees described 1818; F. oxysporum Schlechtendahl described 1824; F. avenaceum Fries (Fusisporium) described 1832: F. equiseti Corda (Selenosporium) described 1838;  F. graminearum Schwabe described 1838;  F. solani (Fusisporium) Martius described l842;  F. sambucinum Fuckel described 1869;  F. semitectum Berk. & Rav. described in 1875.

The Fusarium problem can he divided into four major aspects.  These are: storage rots, plant pathogens, toxins and hormones, and human and animal pathogens.

Fusarium as the cause of storage problems

A study of the problems caused by Fusarium began in a modern sense with an investigation into the rotting of potatoes carried out by Martius in 1840-41 and published in 1842. He found the causal organism to be a fungus which he called Fusisporium solani; this was later transferred to Fusarium as Fusarium solani (Mart.) Sacc.

Throughout the mid-nineteenth century a number of papers appeared describing Fusarium species associated with rots of potato tubers. The workers were mostly German and included Harting (1846). Schacht (1856) and Reinke & Berthold who wrote a marvellous paper called Zersetzung der Kartoffel (the decay of potatoes). published in 1879. All these authors regarded the Fusarium species they isolated as saprophytes. In fact, both de Bary (1861) and Reinke & Berthold (1879) regarded them as obligate saprophytes.

It was almost 20 years later that Pizzigoni (1896) and Wehmer (1897) showed by inoculation experiments that Fusarium species could cause tuber rots. These findings were not immediately accepted because they contradicted the authority of the great de Bary and of Reinke and Berthold and also because certain of their contemporaries such as Frank (1896 and 1898) tried and failed to produce comparable results. In fact it was not until 1904 that the parasitic nature of Fusarium species as storage rots was established in a paper published by Erwin F. Smith and Swingle (1904) in the U.S.A. Unfortunately Smith and Swingle regarded F. solani and F. oxysporum as identical species and took up F. oxysporum because it was the earlier name. Papers by Appel & Wollenweber (1910), Sherbakoff (1915) and others have since added a number of other Fusarium species to the list of storage rots.

Fusarium as the cause of plant disease

It is also to workers in the U .S. A. in the latter part of the last century that we owe our first specific knowledge of Fusarium, species as the causal agents of serious plant diseases and also as toxin producers which could cause serious problems when infected grain was fed to animals.

Fusarium oxysporum featured as the first and it is still the most important Fusarium species causing diseases of economic crops. Between 1892 and 1899 a number of papers were published in the U.S.A. demonstrating the pathogenicity of Fusarium oxysporum to living plants. Two papers in particular stand out. A disease of cotton called ‘Frenching’ was becoming serious in Alabama, and Atkinson was invited to investigate it; in 1892 he published his findings and described Fusarium vasinfectum as the causal organism. This is a form of F. oxysporum and the name is still used in relation to cotton wilt. Atkinson described the typical Fusarium wilt including the presence of gummy substances blocking the vascular tissue, and he illustrated what has proved to be the diagnostic character for the identification of F. oxysporum, namely the phialides producing the microconidia.

The second outstanding paper was by Erwin F. Smith. During the 1890s he extended this work by Atkinson to the wilt disease of cotton, watermelon and cowpea in South Carolina and applied the technique of true pathogenicity testing, almost to the use of Koch’s principles, he went on to tell us (1899) that a field infected with melon wilt fungus should not be planted with melons for a number of years but that canteloupes, cotton, peanuts, cowpeas or soybean could be grown. He also demonstrated how Fusarium was carried by seeds, water run-off, soil particles or farm implements. Smith called his fungi: F. vasinfectum (Neocosmospora vasinfecta) from cotton, F. niveum from water melon and F. thracheiphila from cowpea. he showed remarkable perception; he could not separate his three pathogenic fungi in pure culture but he used the three names because of their different pathogenicity.

It was not until 1940 that Snyder and Hansen placed these three names and about twenty other similar ones as synonyms of F. oxysporum and separated the pathogenic strains (because that is what they are) as formae speciales. With all due respect this was not a very remarkable advance after 40 years and now 40 years later we are still unable to differentiate one pathogenic strain from another without the help of a suitable host plant.

Fusarium toxins

Shortly after the work of Smith and his compatriots, evidence came to light regarding the serious toxic effects that may arise when grain infected with Fusarium species is fed to farm animals. In Nebraska in the 1890s, horses, cows and pigs were reported as losing hair and hooves after eating infected grain. The situation was summarised by Peters in a paper published in 1904. Much of this early work was confused because workers believed, as did Peters, that they were dealing with an outbreak of ergotism. However, by this time Sheldon (1904) had already confirmed that the toxicity was due to a Fusarium species which he isolated and named F. moniliforme - in fact the first Fusarium species to be named following an investigation into its toxicity.

Fusarium infections of humans

In recent years we have had many examples of Fusarium solani and F. oxysporum causing a direct fungal infection of the eye. Ironically these troubles only go back to around the time when antibiotics came into general use in the 1940s and to when, more specifically, the use of steroids began in the l950s. There are also many records of Fusarium species associated with various types of ulcers. An interesting paper by Greco was published in 1916 on ‘The origin of tumours’. He describes a fungal infection of the nose which he said was caused by Fusarium vinosum, a species Wollenweber placed as a synonym of F. flocciferum, one of the rarer members of the Discolor group.

What is Fusarium and what is its biology?

Originally, as proposed by Link in 1809, Fusarium was the name of hyaline, fusiform, non-septate asexual spores borne on a stroma. Note the ‘non-septate’. A definition such as this is not specific and could apply to at least fifty different genera of Hyphomycetes. The ‘borne on a stroma’ was given considerable stress before the days of cultures and hence the genus was included by Fries (1821) in the Tuberculariaceae. With the development of pure culture methods for the identification of Fusarium species the presence of a stroma or sporodochium was no longer regarded as an essential character of the genus. For more details, see Wollenweber (1913).

Various people have tossed the generic description around in the past 70 years and all we can say is that Fusarium is basically a genus of hyaline, septate, phialidic (enteroblastic) asexual spores whose foot cell bears a heel. The only significant or specific character in this description is the heel. This whole genus is, if you like, bound together by a heel. The origin of the heel of the foot cell was considered by Wilcox et al. (1913) to be a function of the medium and growth rate, chiefly because it is absent from many conidia of the Martiella group, notably Fusarium solani and F. coeruleum. Pethybridge & Lafferty (1917) considered it to be a question of ageing as they found it to be more prominent on spores produced in older cultures. What is its biological or ecological function? I would say nil. Yet obviously it binds together a group of species which have many specialised characters in common. In fact it transcends the divisions of the perfect state genera which have a Fusarium conidial state and it has made us look again at our classification of these perfect or teleomorphic states.

To me this heel on the basal cell of the macroconidium or asexual spore represents a vestigial character which illustrates a line of genealogical development of the highly organised Fusarium phialide from the more primitive sporogenous cell as found in Trichothecium roseum. More primitive forms are found in other conidial Hypomyces species. As the major toxins produced by Fusarium species are trichothecenes it is appropriate that there exists such a morphological link of this specialised nature between Fusarium and Trichothecium, from which the name for the toxin was derived.

The fact that Fusarium species have a highly effective enteroblastic phialidic method of spore formation implies that the conidia are slime spores and that they are dispersed by water, soil particles or on seed etc. The exception to this would be if the spore had undergone a secondary modification to dry spores as found in Penicillium and Aspergillus but there is no evidence of this in Fusarium. This statement is supported by the fact that although Fusarium species are very common, especially in soil, they are seldom caught in spore traps. Certainly in our Fusarium laboratory no Fusarium species were isolated from the air, nor do they normally occur as laboratory contaminants. However, there are exceptions. In the first place about half of the Fusarium species are capable of producing a perithecial state; records from aerial spore-trapping experiments seldom show whether a Fusarium colony developed from an ascospore or a conidium. Nelson, White & Toussoun (1971) demonstrated how effective the ascospores of Gibberella zeae, the ascospore stage of F. graminearum, could be in the dispersal of Fusarium graminearum in their carnation houses. The interesting point to me is the situation where Fusarium species without a known perithecial state have been authentically recorded from a spore trapping experiment. Such a report was produced by Lukezic & Kaiser (1966). They found that conidia of Fusarium semitectum (which they called F. roseum - Gibbosum), that formed on aerial mycelium, were easily dislodged by wind and dispersed in the atmosphere by wind speeds as low as 2.4 km.p.h. They also found that conidia formed on sporodochia on banana fruit could not be dislodged by air blasts; however, they offered no explanation of this apparent discrepancy. In fact the explanation is very simple: Fusarium, semitectum, and several other species of Fusarium, have both slime-spored and often almost vestigial dry-spored conidia.

I think the precise method of spore formation in conidial fungi is a theme somewhat overplayed by taxonomists looking for genealogical relationships. There is a strict limit to the number of ways a conidium can be formed from a strand of mycelium and similar methods have obviously arisen several times in unrelated species. These various methods have been studied in great detail but what is biologically significant is whether a conidium ends up as a dry spore or a slime spore. In fact there is considerable evidence that many fungi produce both. Fusarium in the course of evolution has become adapted largely as a specialised soil fungus with the main spore slimy, but there is plenty of evidence available for the existence of a somewhat vestigial dry spore or blastosporic form. In F. semitectum and F. avenaceum these spores are roughly equal in size to the phialidic macroconidia whereas in F. fusarioides (F. chlamydosporum) and F. sporotrichioides the blastosporic form represents the microconidial state. It is (lie presence of these dry blastospore forms which explains frequent reports from aerial trapping experiments of Fusarium: spores from species without a known perithecial state.

To return for the moment to my statement that the heel on the conidial cell transcends the perithecial state as a taxonomic character and makes one look again at our generic classification of these perithecial states. The Saccardoan emphasis on spore septation as a generic character within the Hypocreales has had to be discarded. As it existed, the three basic genera with Fusarium: conidial states were Nectria, with yellow to red perithecia and l-septate ascospores, Caloneciria, with a similar pigmentation and 2-or 3-septate ascospores and Gibberella, also with 3-septate ascospores and a purple pigmentation in the perithicial wall. These are really very minor differences. Spore septation alone cannot be accepted as a generic character within the Hypocreales and for further discussion of this see recent papers by Booth (1978), Samuels (1978) and Amy Rossman (1983). In fact, ascospore septation is not a generic character and it is certainly a very minor diagnostic character compared to the heel on the foot cell of the phialidic macroconidium. One can say this, because this heel binds together a group of fungi which have numerous growth, pathogenic and toxigenic characters in common.

Fungi which have been placed in other genera, or even in another class, because of some stromatic character, but which produce septate conidia with a foot cell have been shown, in culture, to be typical Fusarium species. For instance, the Coelomycete genus Botryocrea, has a pycnidium-like structure producing Fusarium-like spores with a foot cell. This is merely an environmental reaction to dry conditions. In nature many stromatic fungi develop, when iii a dry atmosphere, the first conidia below the upper surface of the stroma. Botryocrea sclerotioides develops its stromata on species of Astragalus. However, as Amy Rossman (1983) has just demonstrated, when this species is grown in culture it assumes normal Fusarium-like growth. PycnoFusarium rusci on Ruscus aculeatus, another pycnidial genus with a Fusarium-like spore and which is based on a single collection, will, I am prepared to wager, prove to be a similar case.

In this area it is my opinion that we largely create our own problems. Fusarium species are a labile group of organisms which when grown on artificial media may produce a multitude of variants both in their morphology and cultural characteristics depending upon the cultural conditions. There is no mystery about Fusarium taxonomy. In fact I can state categorically that if you make fresh collections from nature and place single conidium or single ascospore isolations on to a neutral culture medium then with the possible exception of Fusarium avenaceum you will find no greater variation in your resultant cultures than you would find in any other group of microfungi in culture. Of course if you are looking for trouble Fusarium cultures are ideal candidates for creating problems. William Brown, in a series of papers published between 1924 and 1928 (two of which were written with A. S. Horne), demonstrated how cultural conditions could affect both the morphology of the spores produced and the pigmentation of the medium. The C/N ratio and the amount and type of phosphate were all shown to he important factors.

Apart from the medium, if you are careless with your culture techniques you will produce your own problems, e.g. mass transfer from old cultures on to different media will often make you think you have contaminants present. In Fusarium avenaceum, probably the most unstable Fusarium species, mass transfer will soon reduce the long aciculate conidia to a short, variable almost microconidia-like mass; however, this is a cultural effect and if single conidia are taken then the normal form is quickly restored. Even preparing slides needs care. The tip of a needle introduced into a sporodochium will produce a slide with a spore morphology different from one made from the edge of a growing colony.

What we are aiming for is the concept that members of a species taxon are similar because they share a common heritage; they are not placed together because under certain conditions they can he made to look alike.

Alan Burgess (1955) wrote an interesting essay entitled ‘problems associated with the species concept in mycology’. He compared species concepts for fungi with those for flowering plants and came to the conclusion that in flowering plants species concepts are extremely easy to formulate compared to those for fungi. It is interesting to note that whereas there are far more fungi then flowering plants, theme are probably five times more flowering plant taxonomists than fungal taxonomists.

However, to return to fungal taxonomy. In spite of my categorical statement, variation does occur in Fusarium species amid these variations often blur the characters one accepts for species limitation. What these limitations are depends upon a number of things amid it is a question of determining the species characters within the limits of this natural variation. If the species has a known perithecial state, these limitations cover the range of variation that comes from a series of ascospore isolations on a range of what has come to be accepted as standard Fusarium media. One would not include the effects of staling on the spore morphology.

Some names are used to cover what at first appears to be a complex of species. In Fusarium oxysporum for instance, there are six or seven distinct morphological and cultural forms, all of which have in the past been given different names. A taxonomist would be happy to accept these as distinct species except that in doing so he would ignore the biology of the species. In F. oxysporum it is the strain of the organism that matters, i.e. its capacity to attack a specific host. We refer to these strains as formae speciales and the specific gene or genes they carry can be present in any one of the six or seven cultural forms mentioned.

There have been a number of myths perpetrated about the ability of Fusarium species to produce genetically induced variation. To sonic extent one must blame those who write reviews. Although there are people who are masters of this art, my complaint is that the reviewer likes to tell a story and tends to accept published statements which support it (in this case that of genetically controlled variation), instead of attributing so much of this apparent variation to slap-dash cultural methods.

One such story concerns the variation in Fusarium species due to heterokaryosis as a result of either branching of, or conidial production from, a heterothallic thallus. This has been maintained in a number of publications amid is based on (lie most flimsy evidence. As Puhalla (1981) stated (and our own observations confirm), hyphal tips are uninucleate and both micro- and macroconidia begin with only one nucleus; thus all nuclei in a macroconidium are mitotic descendants of this original haploid nucleus. Certainly hyphal fusion occurs in cultures but I would suggest that this relates more to nutritional needs than to genetical exchanges. In the first place there is a tendency for one strain of a species to rapidly occupy a whole area, with fusion thus occurring between hyphae with the same mitotic history. Also the occurrence of new strains of formae speciales with their specific pathogenicity is an extremely rare phenomenon which would surely not be the case if genetic interchange was as frequent as the presence of hyphal fusion might suggest. Available evidence suggests that the spread of pathogenic strains outside their original or immediate area is due more to the ineptitude of man than to a new mutation. The occurrence of widely dispersed opposite strains of heterothallic species will be discussed later.

Another chestnut is parasexuality which is often quoted as a means by which Fusarium species achieve variability. However there is little supporting evidence and a clear demonstration of parasexuality in Fusarium species is still not available. Evidence suggests that true variation in Fusarium, when it does occur, is in fact due to mutational changes in the nuclei and there are various techniques by which these may be induced in nature. These are, in relationship to the growth rate of these fungi, extremely rare.

Fusarium sexual stages

Finally we need to discuss the perithecial or sexual stages of Fusarium species. I have said elsewhere that taxonomists are apparently more interested in sex than are species of Fusarium. Whilst we are struggling to find the sexual or perithecial state, and to study the sexual processes which produce them, Fusarium species are as rapidly as possible dispensing with these processes altogether. In the genus as a whole these species which apparently live a life without sexual conjugation appear to compete equally with those that still have a sexual stage in their life-cycle.

In fact Fusarium oxysporum, the most common species, the most economically important in the most studied species, has no perithecial state. Even in call the countries, where there may be up to six months of enforced dormancy of each year, there is no greater predominance of species producing perithecia than those without this facility. In fact, in contrast to most Pyrenomycetes, perithecial states of Fusarium species which include Nectria, Calonectria and Gibberella have a comparatively short period of survival, especially if dried. In species with a sexual phase there is a tendency to change from a homothallic to a heterothallic state and for one strain to dominate a given area. Fusarium solani, a very common world_wide species, occurs both as homothallic and heterothallic strains which produce perithecia of Nectria haematococca. Each year the Commonwealth Mycological Institute (CMI) receives from all over tile world what may be regarded as a random sample of isolates and the number of heterothallic isolates amongst these is probably 100 for each homothallic isolate received. A similar situation exists in Fusarium graminearum (Gibberella zeae) but in this species homothallic isolates are more common. In F. decemcellulare (Caloneciria rigidiuscula) a more spccialised situation exists. This tropical species has a more restricted distribution and, whereas in homothallic strains of, for example, F. solani it is difficult to stop perithecia being formed in culture, in F. decemcellulare perithecial production has to be encouraged by specific cultural techniques. Also it has been stated that strains of C. rigidiuscula with four-spored asci are saprophytic whereas those with eight spores are parasitic. We have observed only strains with four-spored asci to 1)e homothallic.

Thus homothallic strains are definitely in the minority and our evidence for stating that in heterothallic species asexual conjugation is restricted comes from the observation that opposite mating types of heterothallic strains are often widely dispersed geographically. Gordon (1954) was probably the first worker to observe this phenomenon. He found that the mating types of Fusarium sulphureum (Gibberella cyanogena) occurred separately in nature and in no instance in his experience were they found together. Thus mating type ‘a’ from British Columbia mated with ‘A' from Manitoba and Tasmania. Both were found in Britain and in Prince Edward Island but not together in the same area. Gordon went on to develop this work with 13 other heterothallic species. He found that most species were bisexual but self-sterile and opposite mating types were geographically separate, which explains the rarity of Gibberella species in nature. One is left with the impression that like galaxies these opposite mating strains are rushing away from each other.

Burnett (1975), in his review of the situation existing for heterothallic strains of Nectria haematococca f. sp. cucurbitae, stated that in nature no two morphological and physiological compatible types of species exist together. Now this may be true to a large extent for other species of Fusarium or at least true for N. haematococca f. sp. cucurbitae; however it is a fact that one does occasionally find perithecia of these heterothallic strains in nature, which suggests that these opposite strains do occasionally meet and overcome their apparent sexual aversion. indeed, even with F. sulphureum Gordon did finally find one collection of Gibberella cyanogena in nature and we have collections of heterothallic F. sambucinum (G. pulicaris), F. graminearum (G. zeae) and several of F. lateritium (G. baccata), all of which provides evidence for heterothallic strains making contact in nature. Nevertheless the phenomenon is much rarer than one would expect. For the taxonomist this rarity does provide problems in establishing species limitations.

Having discussed the history and importance of Fusarium species I would like to end with a word of warning. If you set your sights on Fusarium species alone, and this is the subject of the Symposium. then you may be blinkered as to what is happening in related genera. Available evidence suggests that the production of trichothecenes is a phenomenon of the Hypocreales, the name itself being derived from Trichothecium roseum; therefore if you are concerned with these toxins you cannot be concerned with Fusarium alone. In Plant disease, apart from the rather specialised ‘oxysporum wilt’ and the relationships which exist between Fusarium species and cereals, almost all symptoms of Fusarium diseases can be simulated by species of Cylindrocarpon, especially on acid soils. Also with regard to storage problems, other hypocreaceous fungi. such as Trichoderma and Cylindrocarpon, can cause the same problems and produce the same or similar toxins. So BEWARE.

Brown, W. & Horne, A. S. (1924). Studies of the genus Fusarium. I. General account.

• To some mycologists this piece may appear dated, but we have yet to find a better summary, and therefore include it in our presented literature.
• Thanks to the DEA FOIA office for pointing this out to mycoherbicide.info

Keys for the identification of Fusarium.

Fusarium:

Fusarium Key

Pleospora: