SciELO - Scientific Electronic Library Online

vol.22 issue1Co-evolution model of Colletotrichum lindemuthianum (melanconiaceae, melanconiales) races that occur in some Brazilian regionsGenetic maps of Saccharum officinarum L. and Saccharum robustum Brandes & Jew. ex grassl author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Genetics and Molecular Biology

Print version ISSN 1415-4757On-line version ISSN 1678-4685

Genet. Mol. Biol. vol.22 n.1 São Paulo Mar. 1999 

Transformation of the entomopathogenic fungi, Paecilomyces fumosoroseus and Paecilomyces lilacinus (deuteromycotina: hyphomycetes) to benomyl resistence


Peter W. Inglis, Myrian S. Tigano and M. Cléria Valadares-Inglis
Laboratório de Genética Molecular Microbiana (GMM), Cenargen/Embrapa, Caixa Postal 02372, 70849-970 Brasília, DF, Brasil. Send correspondence to M.C.V.-I. Fax: +55-61-340-3573, E-mail:




The entomopathogenic fungi Paecilomyces fumosoroseus and P. lilacinus have been transformed to resistance to the fungicide benomyl by a polyethylene glycol (PEG)-mediated procedure using a mutant b-tubulin gene from Neurospora crassa carried on plasmid pBT6. Benomyl-resistant transformants of P. lilacinus were obtained that could tolerate greater than 30 µg/ml benomyl and P. fumosoroseus transformants were obtained that could tolerate 20 µg/ml benomyl. Following 5 serial passages of transformants on benomyl-containing media and 5 serial passages on non-selective media, 100% of P. lilacinus transformants were found to be mitotically stable by a conidial germination test. In contrast, only 4 out of 9 transformants of P. fumosoroseus were mitotically stable. Southern blot analysis of genomic DNA from both species suggested that the mechanism of transformation in all transformants was by gene replacement of the b-tubulin allele. Non-homologous vector sequences were not detectable in the genomes of transformants.




Paecilomyces fumosoroseus and Paecilomyces lilacinus (Deuteromycotina: Hyphomycetes) are both entomopathogenic fungi currently receiving attention for their potential in biological control programs. P. fumosoroseus is geographically widespread, found colonising many different orders of insects in all stages of the life cycle, and can also be isolated from soil samples (Samson, 1974). It is being commercially developed for use against a range of agricultural pests, including the whitefly, Bemisia tabaci (Osborne and Landa, 1992; Lacey et al., 1993). P. lilacinus is typically isolated from soil, particularly in samples originating from warmer regions (Domsch and Gams, 1980). Additionally, this fungus has been found infecting a range of different hosts (Samson, 1974) and is a mycoparasite, colonising sclerotia of several species of fungi (Gupta et al., 1993). Of particular interest, however, is the potential of P. lilacinus to infect insects (Rombach et al., 1986), and the eggs and cysts of nematodes (Carneiro, 1992).

Development of Paecilomyces spp. as biopesticides requires proper environmental monitoring and a thorough examination of population dynamics in release experiments. This is required for efficacy estimates, biosafety assurance and regulatory purposes. Such monitoring is complicated by the difficulty in reliable identification of these fungi, conventionally conducted using morphological characters, virulence tests and biochemical characters (Samson, 1974). More recently, molecular markers have been applied to the taxonomy of P. fumosoroseus (Tigano-Milani et al., 1995a) and P. lilacinus (Tigano-Milani et al., 1995b). Such analyses enabled the demonstration of genetic variability among these strains, and in the case of P. fumosoroseus, to divide the species into a number of phenetic groups. However, these methods may not be sensitive enough to differentiate closely related strains and are unsuited to the rapid monitoring of large numbers of isolates in release trials or under routine conditions. An alternative is to apply genetic transformation to introduce one or more marker genes into test strains. The strategy adopted in this study was to use a directly selectable marker, high-level resistance to the fungicide benomyl, to transform strains. This system has the advantage that transformants may be differentiated from non-transformed strains on primary isolation. Moreover, it produces resistant strains that may then be used simultaneously with the integrated application of benomyl-like fungicides for the control of plant pathogens (Goettel et al., 1990).



Fungal strains and vectors

P. fumosoroseus strain CG170 and P. lilacinus strain CG36 were obtained as liquid nitrogen-stored cultures from the Cenargen/Embrapa collection of entomopathogenic fungi. CG170 was originally isolated from a species of Pseudococcus (Homoptera: Pseudococcidae) in Apopka, Florida, USA in 1989. CG36 was originally isolated from an egg of Deois flavopicta (Homoptera: Cercopidae) in Brasília, Federal District, Brazil, in 1992. Working cultures were maintained as monosporic colonies on potato dextrose agar (PDA; Difco) and stored at 4°C until needed. Cultures for preparation of conidia were seeded onto PDA plates and incubated at 28°C for 7-10 days. The minimum inhibitory concentration (MIC) for benomyl was determined by transferring hyphal fragments of CG36 and CG170, and 5 other randomly selected isolates of both Paecilomyces species from the Cenargen collection, to PDA plates containing different concentrations of benomyl.

Plasmid pBT6 was propagated in E. coli strain XL1-blue (Stratagene), and purified using the QIAprep spin miniprep kit (QIAGEN). pBT6 contains the cloned b-tubulin gene from a benomyl-resistant (BenR) mutant of Neurospora crassa (Orbach et al., 1986; McClung et al., 1989).

Preparation of fungal protoplasts

Conidia were harvested from a single plate using sterile distilled water, and inoculated into 200 ml Aspergillus complete medium (ACM) (Pontecorvo et al., 1953) in a 1-l conical flask. Liquid cultures were then incubated for 42 h at 28°C with 200 rpm agitation. Fungal culture (50 ml) was transferred to a sterile centrifuge tube, and mycelium collected by centrifugation at 4000 g. Mycelium was then washed twice in sterile distilled water and once in 0.7 M KCl. Finally, mycelium was resuspended in 10 ml of 0.7 M KCl, as osmotic stabiliser, containing 5 mg/ml Novozym 234 (Novo Biolabs) and 5 mg/ml Cellulase CP (Sturge Enzymes). After 1.5-h digestion at 30°C with gentle agitation, protoplasts were separated from residual mycelium by filtration on 30-µm mesh filters, and recovered from the filtrate by centrifugation at 3000 g for 3 min. Protoplasts were then washed twice in 0.7 M KCl, 50 mM CaCl2 and resuspended in the same buffer to a concentration of approximately 0.5-5 x 108 protoplasts/ml.


Transformations were carried out according to a modification of the method of Herrera-Estrella et al. (1990). Ten micrograms of plasmid pBT6, in a maximum of 10 µl TE buffer, was added to 200 µl protoplast suspension in a 15-ml capacity polypropylene centrifuge tube. Fifty microlitres of a filter-sterilised solution of 25% (w/v) polyethylene glycol (PEG 8000; Sigma), 50 mM CaCl2 and 10 mM Tris-HC1, pH 7.5, was added, and tubes then incubated on ice for 20 min. The same PEG solution (2 ml) was added with gentle mixing, and the tubes incubated at room temperature for 5 min. The PEG was then diluted by adding 4 ml 0.7 M KCl, 50 mM CaCl2, and 0.5-ml aliquots of the transforming mixture then plated on PDA containing 0.7 M KCl and benomyl (10 µg/ml for P. fumosoroseus and 7.5 µg/ml for P. lilacinus). Controls consisting of transforming mixtures lacking DNA were performed twice and plated as before.

Selection and stabilisation of transformants

Benomyl-resistant transformants started to appear between 7 and 24 days after plating for both CG36 and CG170. These colonies were transferred to PDA plates, lacking osmotic stabiliser, containing the appropriate concentration of benomyl used for selection, and were allowed to grow at 28°C until colonies were observed to conidiate. Transformants were sub-cultured in this fashion for 5 generations, after which transformants were transferred to PDA plates lacking benomyl selection pressure. Transformants were sub-cultured a further 4 times on non-selective medium, and conidia then collected in order to compare germination and development rates on PDA containing selective concentrations of benomyl with rates on non-selective PDA.

Molecular analysis of transformants

Genomic DNA was extracted from liquid-grown (48 h, 28°C in 200 ml Aspergillus complete medium (Pontecorvo et al., 1953) transformants and wild-type fungi using the method of Rogers and Bendich (1988), modified by the addition of a phenol extraction step to remove residual nuclease activity. Southern blots were prepared on Zeta-probe nylon membrane (Bio-Rad), using approximately 5 µg fungal genomic DNA in each lane/digest. Probes were labelled using the RediPrime kit (Amersham) and 32P-dCTP and following hybridisation using standard procedures, blots were exposed to X-ray film.



Protoplasts were efficiently isolated within 2 h from both Paecilomyces spp. using combined treatment with Novozyme 234 and Cellulase CP. Protoplasts regenerated at a frequency of 3-10% and 5-12% for CG170 and CG36, respectively. For both species, PEG treatment reduced this frequency approximately 2-fold. Transformation experiments were carried out twice. In the first experiment, two transformants were obtained per µg of transforming DNA for CG170 (selected on 10 µg/ml benomyl; MIC 7.5 µg/ml benomyl). For CG36, 0.4 transformants were obtained per µg of transforming DNA (selected on 7.5 µg/ml benomyl; MIC 5 µg/ml benomyl). In the second transformation experiment, 15 transformants were obtained per µg of transforming DNA for CG170 and 23 transformants were obtained per µg of transforming DNA for CG36. In both experiments, benomyl-resistant colonies of both species appeared between 7 and 21 days following transformation. No spontaneously growing colonies were observed in control experiments lacking DNA.

Nine CG170 transformants and four CG36 transformants were selected for further study. On benomyl-containing media, the P. lilacinus transformants were identical in appearance and growth rate to the parental strain growing on non-selective media. All P. fumosoroseus transformants, however, grew slowly and irregularly on benomyl-selective plates, even after 5 serial transfers. Conidial germination tests (Table I) proved that the four selected CG36 transformants were completely mitotically stable and were thus probably homokarotic for the BenR phenotype. In contrast, only 4 out of 9 CG170 transformants were stable, after 5 transfers on non-selective medium, when results were examined statistically. The benomyl MICs of the 4 stable P. lilacinus CG36 transformants were all in excess of 30 µg/ml, the highest concentration tested. MICs of all 4 of the stable P. fumosoroseus CG170 transformants were found to be 20 µg/ml. Benomyl MICs of the randomly selected wild-type fungi are shown in Table II.


Table I - Mitotic stability of Paecilomyces transformants, inferred by conidial germination frequencies.


Numbers of colonies per plate ± SDa



PDA + Benomyl

P. fumosoroseus

Wild type

323.0 ± 45.8

0 ± 0



86.0 ± 15.0

35.3 ± 4.3



291.0 ± 14.6

100.7 ± 6.2



385.3 ± 43.2

310.3 ± 10.9



421.3 ± 30.7

351.7 ± 31.1



101.0 ± 3.6

16.0 ± 3.1



492.0 ± 19.5

447.7 ± 22.4



485.3 ± 6.4

469.3 ± 11.4



465.3 ± 24.6

643.7 ± 7.8



594.0 ± 21.0

522.0 ± 35.6


P. lilacinus

Wild type

165.3 ± 18.7

0 ± 0



36.7 ± 2.3

36.0 ± 8.3



108.3 ± 6.4

108.3 ± 9.4



91.7 ± 9.0

71.3 ± 21.6



58.0 ± 11.0

71.0 ± 3.6


aConidia were collected from transformants, diluted by a factor of 106 and plated. Results are reported as mean ± SD (standard deviation), (N = 3) numbers of colonies forming on potato dextrose agar (PDA) or PDA containing selective concentrations of benomyl (10 µg/ml for CG170 and 7.5 µg/ml for CG36), after 5-day incubation at 28oC.
bOne tail t-test, germination and colony development frequency on PDA = frequency on PDA + benomyl (P > 0.01; N = 3).



Table II - Benomyl minimum inhibitory concentration (MIC) of selected wild-type Paecilomyces fumosoroseus and P. lilacinus isolates and respective pBT6 transformants.


MICa (µg/ml benomyl)

P. fumosoroseus











(transformant) MIC


P. liacinus











pBT6/CG36 (transformant) MIC


aTests were conducted using PDA (potato dextrose agar) incorporating benomyl, and by incubating plates at 28oC for 7 days. Results are the mean of triplicate tests, with standard deviations of less than 10%.


Genomic Southern blots (Figures 1 and 2) were first hybridised using the 1.4-kb EcoRI fragment of pBT6, carrying the structural gene for b-tubulin. A simple hybridisation pattern was obtained for transformants of both species of Paecilomyces. In CG36, digestion with PstI, which cuts pBT6 once (though not in the b-tubulin gene), produced a signal at approximately 6.4 kb in all transformants and also in the host strain. Digestion with EcoRV, which does not cut pBT6, produced bands estimated at 7.5 kb in all transformants, and, also in the untransformed host strain. In CG170 transformants, the b-tubulin probe hybridised to a band estimated at 2.3 kb in PstI digested genomic DNA, and to a band of approximately 2.1 kb in EcoRV-digested DNA. These hybridised bands were also found in the genomic DNA of the untransformed host strain. Minor hybridising bands were also observed in all lanes, with molecular weights greater and less than the main hybridised bands.


pg122f1.GIF (63767 bytes)

Figure 1 - Genomic Southern hybridisation analysis of P. fumosoroseus strain CG170 pBT6 transformants, probed with the 1.4-kb EcoRI fragment (b-tubulin) of pBT6. Lane 1: plasmid pBT6 EcoRI digest; lane 2: CG170 wild-type DNA PstI digest; lanes 3-11: CG170 BenR transformants (T1-T9) DNA digested with PstI; lane 12: CG170 wild-type DNA EcoRV digest; lanes 13-21: CG170 BenR transformants (T1-T9) DNA digested with EcoRV. Molecular weight markers were l DNA digested with HindIII and the 100-bp ladder (Gibco BRL).



pg122f2.GIF (61555 bytes)

Figure 2 - Genomic Southern hybridisation analysis of P. lilacinus strain CG36 pBT6 transformants, probed with the 1.4-kb EcoRI fragment (b-tubulin) of pBT6. Lane 1: plasmid pBT6 EcoRI digest; lane 2: CG36 wild-type DNA PstI digest; lanes 3-6: CG36 BenR transformants (T1-T4) DNA digested with PstI; lane 7: CG36 wild-type DNA EcoRV digest; lanes 8-11: CG36 BenR transformants (T1-T4) DNA digested with EcoRV. Molecular weight markers were l DNA digested with HindIII and the 100-bp ladder (Gibco BRL).


Filters were subsequently stripped and re-probed using the 4.2-kb EcoRI fragment of pBT6, in order to determine the possible integration of vector sequences in transformants. No hybridisation signals were detected in this case, ruling out complete vector integration in the transformants (data not shown).



The comparatively low frequencies of transformation obtained in this study are typical of entomopathogenic fungal PEG-mediated and benomyl-selective systems. For example, Bernier et al. (1989) obtained 4 transformants using 50 µg DNA in M. anisopliae. Goettel et al. (1990) obtained 9 transformants using 50 µg of DNA and 2 x 106 viable transformants, again in M. anisopliae. Preliminary observations suggest that transformation frequencies may be significantly increased by allowing protoplasts an 18 h-recovery period, following transformation, in osmotically stabilised liquid medium, thereby diluting the toxic effect of PEG and allowing expression of the benomyl-resistant phenotype (data not shown).

The mitotic stability of benomyl-resistant transformants of P. fumoroseus CG170 and P. lilacinus CG36 differed significantly. This may also be reflected in the differences observed in the growth patterns of the transformants on selective medium. Growth irregularities were previously observed in Trichoderma harzianum, transformed to benomyl resistance by pBT6 (Ulhoa et al, 1992). The contrast between the two Paecilomyces species could be a result of differences in expression of the benomyl-resistant phenotype, possibly brought about by different recombinational events during transformation. The mitotically unstable P. fumosoroseus is probably a result of the persistence of heterokaryotic nuclei in the mycelium, even after 5 serial transfers on selective medium.

Southern hybridisation analysis revealed that the most likely mechanism of transformation in all transformant was Type 3 or homologous gene replacement (Fincham, 1989). This was inferred by the fact that single hybridising bands were detected for each digest, and by the fact that vector sequences were not detectable by hybridisation in any of the transformant genomic DNA. Moreover, the hybridising bands were observed to have the same molecular weight as the bands corresponding to the untransformed parental strains b-tubulin genes, which evidently share high homology with the cognate gene from N. crassa carried on the transforming plasmid. Previously, homologous bands were not detected by hybridisation in wild-type Metarhizium anisopliae, when probed with the Aspergillus nidulans benA3 b-tubulin gene (Goettel et al., 1990). However, Ulhoa et al. (1992) did detect b-tubulin homology with species of Trichoderma, when probed with the b-tubulin gene carried on pBT6. Differences in probe specificity and hybridisation conditions may explain this variation. Some minor variation was observed in the molecular weights of the hybridised bands of the P. lilacinus CG36 PstI-digested DNA compared with the untransformed parent and between transformants. This may be explained by variation in the border sequences involved in the specific recombinational events. Several weakly hybridising bands were also detected on the P. fumosoroseus CG170 blots, when hybridised with the b-tubulin probe. These minor bands, also detected in the parental DNA, most likely correspond to heterologous sequences, cross-hybridising with the probe.

The results confirm that P. fumosoroseus and P. lilacinus are stably transformable using a gene from another fungus. Moreover, these data suggest that these species possess an efficient system for recombination with homologous genes, leading to gene replacement and elimination of non-homologous vector sequences. Such a system would enable further genetic manipulation of these species, leading to possible improvements in their efficacy as biological control agents.



The authors thank Irene Martins for technical assistance. This work was conducted as Cenargen/Embrapa sub-project, and was supported by a grant from the International Centre for Genetic Engineering and Biotechnology (ICGEB). P.W.I. was supported by IICA/PROMOAGRO fund. M.S.T. was partially supported by the Brazilian National Research Council (CNPq).




Os fungos entomopatogênicos Paecilomyces fumosoroseus e P. lilacinus foram transformados com o plasmídio pBT6, contendo o gene de b-tubulina modificado, isolado do fungo Neurospora crassa, o qual confere resistência ao fungicida benomil. Transformantes resistentes ao benomil, obtidos da P. lilacinus, apresentaram tolerância a concentrações superiores a 30 mg/ml, enquanto transformantes da P. fumosoroseus apresentaram tolerância a 20 mg/ml. Após 5 passagens seriadas dos transformantes em meio contendo benomil e 5 passagens seriadas em meio não seletivo, 100% dos transformantes da P. lilacinus apresentaram estabilidade mitótica, através do teste de germinação de conídios. Em contraste, somente 4 dos 9 transformantes obtidos para P. fumosoroseus foram mitoticamente estáveis. Análise de Southern blot dos DNA genômicos de ambas as espécies sugere que o mecanismo de transformação foi por substituição alélica do gene de b-tubulina, para todos os transformantes obtidos. As seqüências não homólogas do vetor não foram detectadas nos genomas dos transformantes.




Bernier, L., Cooper, R.M., Charnley, A.K. and Clarkson, J.M. (1989). Transformation of the entomopathogenic fungus Metarhizium anisopliae to benomyl resistance. FEMS Microbiol. Lett. 60: 261-266.         [ Links ]

Carneiro, R.M.D.G. (1992). Princípios e tendências de controle biológico de nematóides com fungos nematófagos. Pesq. Agropec. Bras. 27: 113-121.         [ Links ]

Domsch, K.H. and Gams, W. (1980). Compendium of Soil Fungi. Vol 1. Academic Press, New York, pp. 529-532.         [ Links ]

Fincham, J.R.S. (1989). Transformation in fungi. Microbiol. Rev. 53: 148-170.         [ Links ]

Goettel, M.S., Leger, R.J.S., Bhairi, S., Jung, M.K., Oakley, B.R., Roberts, D.W. and Staples, R.C. (1990). Pathogenicity and growth of Metarhizium anisopliae stably transformed to benomyl resistance. Curr. Genet. 17: 129-132.         [ Links ]

Gupta, S.C., Leathers, T.D. and Wicklow, D.T. (1993). Hydrolytic enzymes secreted by Paecilomyces lilacinus cultured on sclerotia of Aspergillus flavus. Appl. Microbiol. Biotechnol. 39: 99-103.         [ Links ]

Herrera-Estrella, G.H., Goldman, G.H. and Van Montagu, M. (1990). High-efficiency transformation system for the biocontrol agents, Trichoderma spp. Mol. Microbiol. 4: 839-843.         [ Links ]

Lacey, L.A., Kirk, A.A. and Hennessey, R.D. (1993). Foreign exploration for natural enemies of Bemisia tabaci and implementation in integrated control programs in the United States. In: A.N.P.P. Third International Conference on Pests in Agriculture, Montpellier, France, pp. 351-360.         [ Links ]

McClung, C.R., Phillips, J.D., Orbach, M.J. and Dunlap, J.C. (1989). New cloning vectors using benomyl resistance as a dominant marker for selection in Neurospora crassa and in other filamentous fungi. Exp. Mycol. 13: 299-302.         [ Links ]

Orbach, M.J., Porro, E.B. and Yanofski, C. (1986). Cloning and characterization of the gene for b-tubulin from a benomyl-resistant mutant of Neurospora crassa and its use as a dominant selectable marker. Mol. Cell Biol. 6: 2452-2461.         [ Links ]

Osborne, L.S. and Landa, Z. (1992). Biological control of whiteflies with entomopathogenic fungi. Fla. Entomol. 75: 456-471.         [ Links ]

Pontecorvo, G., Roper, J.A., Hemons, L.M., MacDonald, K.D. and Bufton, A.W.J. (1953). The genetics of Aspergillus nidulans. Adv. Genet. 5: 141-238.         [ Links ]

Rogers, S.O. and Bendich, A.J. (1988). Extraction of DNA from plant tissues. In: Plant Molecular Biology Manual (Gelvin, S.B., Schilperpoort, R.A. and Verma, D.P.S., eds.). Kluwer Academic Publishers, Dordrecht, pp. 1-10.         [ Links ]

Rombach, M.C., Aguda, R.M., Shepard, B.M. and Roberts, D.W. (1986). Infection of rice brown planthopper, Nilaparvata lugens (Homoptera: Delphacidae), by field application of entomopathogenic hyphomycetes (Deuteromycotina). Env. Entomol. 15: 1070-1073.         [ Links ]

Samson, R.A. (1974). Paecilomyces and some allied hyphomycetes. In: Studies in Mycology. Vol. 6. Centraalbureau voor Schimmelcultures, Baarn, The Netherlands.         [ Links ]

Tigano-Milani, M.S., Honeycutt, R.J., Lacey, L.A., Assis, R., McClelland, M. and Sobral, B.W.S. (1995a). Genetic variability of Paecilomyces fumosoroseus isolates revealed by molecular markers. J. Invertebr. Pathol. 65: 274-282.         [ Links ]

Tigano-Milani, M.S., Samson, R.A., Martins, I. and Sobral, B.W.S. (1995b). DNA markers for differentiating isolates of Paecilomyces lilacinus. Microbiology 141: 239-245.         [ Links ]

Ulhoa, C.J., Vainstein, M.H. and Peberdy, J.F. (1992). Transformation of Trichoderma species with dominant selectable markers. Curr. Genet. 21: 23-26.         [ Links ]


(Received February 13, 1998)

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License