| Front page | | Contents | | Previous | | Next |

Entomophthorales on cereal aphids

2. Characterisation of Entomophthorales

The order Entomophthorales

The order Entomophthorales (subdivision: Zygomycotina) is characterised by a sporangium that has been reduced to a single conidium and is usually discharged forcibly at maturity. Primary conidia may produce secondary conidia. Fungi from Entomophthorales produce thick-walled resting spores (zygospores or azygospores). Most species are parasitic on insects and in a few cases on other animals and plants, or they live saprophytically in soil and dung (Webster, 1970; Humber, 1989; Tanada & Kaya, 1993). The mycelium of Entomophthorales is coenocytic, however it may become divided by septae into segments. Somatic protoplasts are common in the order. In some species the mycelium fragments into hyphal bodies (Alexopoulos et al., 1996). Generally the characterisation of Entomophthorales into families, genera and species is based on classical mycological features.

2.1 Characterisation by morphological and pathobiological methods

Families

Within the order Entomophthorales, family characteristics include nuclear cytology: relative nuclear size and appearance and the quantity and distribution of condensed chromatin during interphase, morphology of mitotic chromosomes, the relative placement of the spindle at metaphase and the fate of the nuclear envelope during mitosis (Humber, 1989). Knowledge concerning the modes of formation and germination of resting spores and the nature of vegetative growth and development is also used to separate families (Humber, 1989). The following five families are recognised within the Entomophthorales: Completoriaceae (Humber), Meristacraceae (Humber), Ancylistaceae (Fisher), Entomophthoraceae (Winter) and Neozygitaceae (Ben-Ze’ev & Kenneth). Only the latter three contain entomopathogens (Humber, 1989; Keller, 1999).

Genera

The classification of genera in the Entomophthorales has over the years undergone revision (McCoy et al., 1988). At present primarily three systems are normally used, referred to as Humber’s classification, Keller’s classification and Ba»azy’s classification. According to Keller (1991; 1994) the following characters should be used to define the genera: mode of discharge of primary conidia, number of nuclei per conidium, shape of primary and secondary conidia, mode of formation of secondary conidia and finally pathobiology (e.g. host symptoms). Humber (1989) further focused on the nature of the conidial wall and morphology of the primary conidiophores and / or conidiogenous cells but not the morphology of the secondary conidia and the pathobiology. However, Humber (1989) includes the presence and morphology of cystidia and rhizoids, types of secondary conidia formed and the pathobiology if these are correlated to characteristics of the primary conidia. Ba»azy (1993) agrees with Keller (1991; 1994) and Humber (1989) concerning most points, but includes Neozygites in Entomophthoraceae and includes all species with one nucleus per conidium in the genus Zoophthora. The current classifications for entomopathogens within Entomophthorales are shown in table 2.1. In this report the system proposed by Humber (1989) will be cited. Humber’s features for classification are illustrated in figure 2.1.

Table 2.1 Look here!
Classification of entomopathogenic fungi within the order Entomophthorales (Ba»azy, 1993; Remaudière & Hennebert, 1980; Remaudière & Keller, 1980; Keller, 1987, 1991, 1999; Humber, 1981, 1989; Steinkraus et al., 1998).

Figure 2.1 Look here!
Key for identification of entomopathogenic Entomophthorales to genus level following Humber’s system (Humber, 1989, 1997). (Figure modified after Keller, 1994).

Species

Species are identified on the basis of host insect species and morphological features, primarily the size and shape of the primary conidia and the number of nuclei per conidium. Keys for identification of species are given by Ba»azy (1993). Keys for the species found in Central Europe are given by Keller (1987, 1991). A key to all species of Neozygites is given by Keller (1997b) and an overview of all species of Eryniopsis is given by Keller & Eilenberg (1993). Finally an overview of the Entomophthora muscae (Cohn) Fresenius complex is given by Keller et al. (1999). An overview of morphological features for species infecting aphids is given in Appendix A.

Strain and isolate

Characterisation methods based on morphology and pathobiology are insufficient to distinguish between different strains and isolates of the same species. However, for studies on the development of epizootics it is essential to differentiate between strains. For monitoring the establishment and spread of a released fungus it is essential to identify the released isolate from naturally occurring strains. Methods based on either biochemical reactions or DNA are therefore used for that purpose. In this report the term ‘strain’ is used for a group of clonally related species and the term ‘isolate’ is used for the culture itself (Hawksworth et al., 1995).

2.2 Characterisation by biochemical methods

Biochemical data have been used to assess intraspecific and interspecific variation amongst isolates. The most used methods are electrophoretic mobility of enzymes and isozymes as well as fatty acid composition (May et al., 1979; Milner et al., 1983; Latgé & Boucias, 1984; Glare et al., 1987; Wilding et al., 1993).

Interspecific variation

In general, gel electrophoresis of enzymes and fatty acid compostion are suitable for distinguishing between species though not between isolates of the same species (May et al., 1979; Milner et al., 1983; Glare et al. 1987; Wilding et al. 1993). As an example Wilding et al. (1993) identified entomophthoralean fungi in aphid hosts to species level using gel electrophoresis of enzymes. They found that among the thirteen tested enzymes three of them were able to distinguish between the two aphid pathogens P. neoaphidis and Conidiobolus obscurus (Hall et Dunn) Remaudière & Keller but not between isolates of the same species.

Intraspecific variation

Electrophoresis of isozymes is in contrast useful for investigating intraspecific variation between species. Latgé & Boucias (1984) examined the intraspecific variation of C. obscurus using electrophoresis of isozymes. They demonstrated that isoenzyme banding patterns were independent of host species, geographical origin, mycelia age and culture medium among 30 isolates of C. obscurus. However, the isolates allowed a clustering based on their ability to produce resting spores, sporulate and infect aphids. Similarly, Silvie et al. (1990) demonstrated using isoenzyme analysis that a strain of P. neoaphidis released in a greenhouse experiment for controlling aphids was gradually replaced by naturally occurring strains of the same fungus species.

2.3 Characterisation by DNA based methods

The characterisation of fungi using DNA based techniques has during the last ten years become increasingly important. The methods are used for studies in evolutionary ecology, population genetics and systematics. DNA based methods has several significant advantages over alternatives such as morphological and biochemical characterisation because the genotype rather than the phenotype is assayed (Dowling et al., 1996). Furthermore, some methods are able to distinguish between isolates of the same species, and methods can be applied based on analysis of only a small amount of tissue.

PCR

Over the years different polymerase chain reaction (PCR) based methods have been developed. In table 2.2 some of the commonly used techniques and their applicability to different problems relevant to entomopathogenic fungi are listed.

Table 2.2 Look here!
Commonly used PCR based techniques and their applicability to different problems in characterisation. Techniques marked with * have been used for Entomophthorales (Hodge et al., 1995; Thomsen & Beauvais, 1995; Dowling et al., 1996; Hajek et al., 1996; Jensen et al., 1998; Rohel et al., 1997; Jensen et al., in press; Vestergaard & Eilenberg, in press).

2.3.1 Molecular and morphological variation in Pandora neoaphidis

Objective

The objective of our studies was to obtain information about the diversity of P. neoaphidis isolated from different hosts and geographical origins using both classical characterisation methods (conidial morphology) and molecular techniques (PCR).

RAPD

Random amplified polymophic DNA-PCR (RAPD) was the technique chosen for analysing the geographical variation among P. neoaphidis isolates. As can be seen in table 2.2, RAPD is one of the techniques that is useful for studies of geographical variation.

RAPD-PCR analysis involves amplification of random segments of genomic DNA. Usually 10 base oligonucleotide primers are used to amplify the DNA. A DNA amplification product is generated for each genomic region that happens to be flanked by a pair of 10 base priming sites. The DNA fragments generated in the RAPD reactions are separated electrophoretically on an agarose gel and visualised by ethidium bromide staining. Under standardised conditions, individuals of the same genotype can be expected to show identical RAPD fragment profiles that are likely to differ from those of other genotypes. The protocol used in our experiments is included in appendix B.

Morphology

Length and width of conidia were measured.

Isolates

Sixteen isolates of P. neoaphidis were selected from the ARSEF collection of entomopathogenic fungi. The isolates were selected to cover as many geographical regions in the world as possible. Isolates from either Acyrthosiphon pisum (Harr.), Brevicoryne brassicae (L.) or S. avenae were chosen primarily.

For comparison, other species of Pandora isolated from aphids were selected for the study (Pandora nouryi (Remaudière & Hennebert) Humber and Pandora kondoiensis (Milner in Milner, Mahon & Brown) Humber). In addition two species of Pandora isolated from other insect families were included (Pandora bullata (Taxter & MacLeod in Humber) Humber and Pandora delphacis (Hori) Humber). Finally two isolates from the genus Conidiobolus isolated from aphids were included as outgroups. A list of the isolates included in our study is given in table 2.3.

Table 2.3 Look here!
Insect host and geographic origin of isolates used for RAPD analyses. Isolates written in bold and italics were additionally used for morphological studies.

Results

RAPD-PCR was implemented and further developed for this system. Ten out of fourteen primers amplified multiple DNA fragments for all isolates included in this study. A total of 568 discrete bands were scored from photographs using the ten primers. Phenetic similarity was calculated (UPGMA using Jaccard’s coefficient) by using the statistical software NTSYSpc (V.2.01e). The tree generated by the analysis is shown in figure 2.2. All P. neoaphidis isolates shared more than 40% of the scored bands while sharing virtually none with other species, even other Pandora species isolated from aphids. Among the P. neoaphidis isolates, RAPD grouping could be related to geographical origin. No relationship between host insect and RAPD groups was seen.

Figure 2.2 Look here!
Dendrogram based on RAPD fragment pattern of Pandora neoaphidis and related species. Pn = Pandora neoaphidis; Pno = Pandora nouryi; Pk = Pandora kondoiensis; Pd = Pandora delphacis; Pb = Pandora bullata; Ct = Conidiobolus thromboides and Co = Conidiobolus obscurus. Bold lines are isolates belonging to P. neoaphidis.

Isolates representing different groups in the RAPD-PCR analysis were selected for measurement of length and width of the primary conidia from in vitro material. Results are shown in table 2.4 and figure 2.3. Conidia length, width and length / width ratios were subjected to analysis of variance by Tukey’s multiple comparison procedure. Analysis of variance showed a significant effect of isolate for both length and width measurements and for ratio of length to width (length: F_{12, 259} = 268.0, P<0.0001; width: F_{12, 259} = 145.5, P<0.0001; L/W F_{12, 259} = 77.3, P<0.0001). Differences between P. neoaphidis isolates were small compared to other Pandora species. However conidia from the isolate originating from Chile were significantly longer and conidia from the Australian isolate of P. neoaphidis were significantly broader. These two isolates were also those which differed the most from the rest of the P. neoaphidis isolates in the analysis of RAPD-PCR data. However measurements fell in all cases within the species descriptions. Analysis of the RAPD-PCR data and the conidia measurements showed that the two species P. kondoiensis and P. nouryi possess huge variation.

Table 2.4
Pandora spp. primary conidia (in vitro) dimensions (? m) and length-width ratio. Means within the same column followed by different letter are significantly different by analysis of variance (Tukey’s procedure, P<0.05)

Figure 2.3
Length-width ratio as a function of length +/- S.E. (:m) for Pandora spp. primary conidia (in vitro).

2.4 Conclusions

Both morphological and DNA based methods have been used for characterisation of P. neoaphidis isolates, and our results proved that P. neoaphidis consists of several phenotypes and genotypes. Analysis of data showed that among the P. neoaphidis isolates RAPD grouping could be related to geographical origin whereas no clear conclusions could be drawn based solely on classical morphological measurements. No relationship between host insect and RAPD groups was observed.
In summary, we may conclude:

| Front page | | Contents | | Previous | | Next | | Top |