MOLECULAR CHARACTERIZATION OF Paecilomyces fumosoroseus ( Deuteromycotina : Hyphomycetes ) ISOLATES

ITS and RAPD analyses were used to investigate molecular variations within samples of Paecilomyces isolates and to resolve five morphologically atypical isolates resembling P. fumosorosus, obtained from whitefly in Northern Paraná State. The ITS4-ITS5 amplicon was 700 base pairs (bp) long in all isolates. The five isolates of Paecilomyces not assigned to species produced restriction profiles identical to all the reference strains of P. fumosoroseus. The extent of fingerprint variability observed by RAPD was sufficient to discriminate all the isolates. The genetic similarity among unidentified isolates and strains of P. fumosoroseus was even higher than that observed among reference strains of this species, allowing us to conclude that isolates CNPso-P77, CNPso-P78, CNPso-P80, CNPso-P85 and CNPso-P91 are P. fumosoroseus.


INTRODUCTION
The whitefly Bemisia argentifolli is one of the most severe agricultural pests of the 20 th century.It is widespread throughout the world, and causes considerable damage to a variety of high value crops, including in Brazil where B. argentifolli was recently introduced.As chemical controls were shown to be ineffective, efforts have been made to develop natural remediation in applying biological controls of this pest.Over the world, one of the most common sources of whitefly population mortality is a mycosis caused by the entomopathogenic fungus Paecilomyces fumosoroseus (Wise) Brawn & Smith (Lacey et al., 1996;Humber, 1992).This fungus is associated with whitefly in different regions of the globe (Lacey et al., 1996), including observations performed in Brazil on soybean and other vegetables.The pathogenicity of P. fumosoroseus against whitefly has recently been studied (Vidal et al., 1997;Wraight, et al., 1998).
Despite common outbreaks of B. argentifolli, little is known about the genetic variability of its potential nemesis, P. fumosoroseus.The detection of variability and knowledge of various strain identities are basic prerequisities for either the development of an effective biocontrol strategy or to evaluate the impact of artificial epizootics.The morphological characteristics used for classification of Paecilomyces species frequently do not resolve new isolates into clearly defined species and provide no isolate-level characterisation.Currently, more sensitive genetic methods such as random amplification of polymorphic DNA (RAPD) (Williams et al., 1990) and nuclear rDNA analysis have been extensively employed for genetic variability analyses (Fungaro & Vieira, 1998) and for the identification of species and isolates including entomopathogenic fungi (Fegan et al., 1993;Leal et al., 1994, Bidochka et al, 1994;Tigano-Milani et al., 1995a;Tigano-Milani et al., 1995b;Fungaro et al., 1996;Cravanzola et al., 1997;Berretta et al., 1998, Cantone & Vandenberg, 1998;Piatti et al. 1998).
In this study polymorphisms of internal transcribed spacers (ITS) of ribosomal DNA amplified by polymerase chain reaction (PCR) and RAPD were used to investigate Scientia Agricola, v.57, n.4, p.729-732, out./dez.2000 molecular variations of Paecilomyces isolates and to resolve five morphologically atypical isolates resembling P. fumosoroseus, obtained from whitefly in Northern Paraná State (Sosa-Gómez, 1998).

DNA isolation
Conidial suspensions of 17 isolates in complete medium (Pontecorvo et al., 1953), were incubated for 2 days under agitation at 28°C.The resulting mycelium was harvested by filtration.To extract the DNA, 1g of mycelium was ground to a fine powder in liquid nitrogen and incubated in 800 µL DNA extraction buffer (100mM Tris-HCl (pH8.0),25mM EDTA, 1% SDS, 25mM NaCl) at 65°C for 20 min.The suspension was deproteinized by extracting once with an equal volume of phenol and once with an equal volume of chloroform: isoamyalcohol (24:1).DNA was precipitated by adding two volumes of ice-cold ethanol and 10% 3M sodium acetate.The precipitate was collected by centrifugation, washed with 70% ethanol and dried.The pellet was resuspended in TE (10mM Tris, 1mM EDTA pH 8.0).

rDNA ITS Regions Analyses
Primers ITS4 (5'TCCTCCGCTTATTGATATGC-3') and ITS5 (5'GGAAGTAAAAGTCGTAACAAGG-3') were used to amplify the ITS region of rDNA.The 50 µL of reaction mixture contained 10 ng template DNA, 1 µM of each primer, 100 µM of each dNTP, 5 µL 10x PCR buffer, 1.5 mM MgCl 2 , and 2.5 Units of Taq Polymerase.The mixture was subjected to amplification in a MJ Research thermal cycler.An initial denaturation step of 2 min at 95°C was followed by 35 cycles of annealing for 1 min at 55°C, extension for 1 min at 72°C and denaturation for 2 min at 95°C before a final extension step of 5 min at 72°C.The resultant reaction mixture (10 µL) was separated on a 1.4% horizontal agarose gel, stained with ethidium bromide and visualised under UV illumination.For restriction analysis, 10 µL of each PCR product was digested in a 20 µL volume, using AluI, Xho I, Dde I, Bam H1, Pst I, Pvu I, Hind III, Ras II, Sau 3A1, Eco RI, Hae III under conditions supplied by the manufacturer (Gibco-BRL).The restriction reactions were separated by electrophoresis in 1.4% agarose gels.After staining, the gels were photographed over a UV transilluminator.

RAPD Analyses
The 10-mers used as the random primer in the RAPD procedure was purchased from Operon Technologies, Alameda, CA.Taq DNA polymerase and DNA polymerisation mix (dNTP) were supplied by Pharmacia.The 10x concentrated buffer, supplied with the enzyme, contained 100 mM Tris-HCl pH 9.0, 500 mM KCl, 15 mM MgCl 2 .The amplification reactions were conducted in a total volume of 25 µL.The mixture contained 8.8 µl of fresh ultra-pure sterile water, 2.5 µl buffer, 4.0 µL of dNTP (200 µM each), 2.5 µL of primer (2,5 mM), 0.3 µL of enzyme (1.5 units), 3.0 µL of MgCl 2 (12.5 mM) and 4.0 µL of DNA (20-25 ng).The final concentration of magnesium in the reaction was 3.0 mM.Control reactions were run containing all components except genomic DNA and none of the primers employed yielded detectable amplified products.The mixture was subjected to the following temperature conditions: 92°C for 5 min, followed by 40 cycles of 92°C for 45 sec, 39°C for 1 min 30 sec, 72°C for 2 min and ending with 3 min at 72°C.Samples of 22 µL of the RAPD products were assayed by electrophoresis in 1.4% agarose gels run with TBE buffer.Electrophoresis was conducted at 3V/cm.Gels were stained with ethidium bromide and photographed under UV using a Pan F Ilford film.For data analysis, each isolate was scored for the presence or absence of amplification products.The data were entered into a binary matrix and a pairwise similarity matrix was constructed using the Sorensen coefficient (Sneath & Sokal, 1973).An UPGMA cluster based on Sorensen values was generated using the NTSYS (Numerical Taxonomy System, Applied Biostatistics, Setauket, New York) computer programme (Rohlf, 1994).

RESULTS AND DISCUSSION
The development of PCR and the design of primers for the amplification of the various rDNA regions have considerably facilitated taxonomic studies of fungi.ITS sequences are generally constant, or show little variation within species but vary between species in a particular genus.As such, these sequences have been widely used to develop rapid procedures for the identification of fungal species by PCR-RFLP analysis (Edel, 1998).In this study ITS analyses were conducted by restriction enzyme digestion of the amplified region using the primers ITS 4 and ITS 5.The ITS4-ITS5 amplicon was 700 base pairs (bp) long in all isolates of Paecilomyces sp., P. fumosoroseus and P. tenuipes.Of the 12 restriction enzymes used, differences between isolates were detected with Hae III only.No restriction sites were found when DNA was treated with AluI, Xho I, Dde I, Bam H1, Pst I, Pvu I and Hind III.The enzymes Ras II, Sau 3A1, Eco RI digested the amplicon, but showed no polymorphisms among isolates.Interspecific polymorphisms detected with Hae III, discriminated P. fumosoroseus from P. tenuipes (Figure 1).
The five isolates of Paecilomyces not assigned to species produced restriction profiles identical to all the reference strains of P. fumosoroseus.In addition, ITS intraspecific polymorphisms were observed for neither P. tenuipes nor P. fumosoroseus.
RAPD analyses using 31 primers produced 536 scored bands within all of the 17 isolates.An example of a typical RAPD pattern is given in Figure 2. The extent of fingerprint variability was sufficient to discriminate all isolates.This feature is very important because the identification of isolates of Paecilomyces, has often turned out to be difficult or even impossible when based solely on morphological traits.
A dendrogram based on the amplification fragments is shown in Figure 3.The isolates were separated into 2 large phenetic groups.The first group is formed by all isolates of P. fumosoroseus and the five unidentified isolates.The second group includes all isolates of P. tenuipes.Despite the high genetic divergence between these two groups (0.75), a common amplification product obtained using the primer OPAK-16 (Figure 2) was observed for all isolates.The five unidentified isolates CNPso-P77, CNPso-P78, CNPso-P80, CNPso-P85 and CNPso-P91 were very similar to each other, with an average pairwise similarity of 92.5%.This high genetic similarity was expected since they were isolated from the same epizootic phase in northern regions of Paraná State.The genetic similarity among the unidentified isolates and strains of P. fumosoroseus was even higher than that observed among reference strains of this species.These data allow us to conclude that isolates CNPso-P77, CNPso-P78, CNPso-P80, CNPso-P85 and CNPso-P91 are P. fumosoroseus.A similar strategy was employed by Tigano-Milano et al. (1995b) to study 27 P. fumosoroseus isolates, of which 15 were obtained from B. tabaci, in comparison with one strain of P. lilacinus and nine strains of Paecilomyces not assigned to species.