Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana

f u n g a l b i o l o g y x x x ( 2 0 1 7 ) 1 e1 2

journal homepage: www.elsevier.com/locate/funbio

Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana Wachiraporn TOOPAANGa, Suranat PHONGHANPOTa, Juntira PUNYAa, Cheerapha PANYASIRIa, Kewarin KLAMCHAOa, Rudsamee WASUWANa, Chettida SRISUKSAMa, Duangjai SANGSRAKRUa, Chutima SONTHIRODa, Sithichoke TANGPHATSORNRUANGa, Morakot TANTICHAROENb, Alongkorn AMNUAYKANJANASINa,* a

National Center for Genetic Engineering and Biotechnology, National Science and Technology Development Agency, 113 Paholyothin Road, Klong 1, Klong Luang, Pathumthani 12120, Thailand b School of Bioresources and Technology, King Mongkut’s University of Technology Thonburi, Bangkok 10140, Thailand

article info

abstract

Article history:

The reducing clade III polyketide synthase genes, including pks15, are highly conserved

Received 12 November 2016

among entomopathogenic fungi. To examine the function of pks15, we used targeted dis-

Received in revised form

ruption to investigate the impact of Beauveria bassiana pks15 on insect pathogenesis. South-

19 April 2017

ern analysis verified that the Dpks15 mutant was disrupted by a single integration of the

Accepted 26 April 2017

transformation cassette at the pks15 locus. The Dpks15 mutant had a slight reduction in

Corresponding Editor:

radial growth, and it produced fewer spores. Our insect bioassays indicated the Dpks15

Richard A. Humber

mutant to be significantly reduced in virulence against beet armyworms compared to wild type (WT), which could be partially accounted for by its markedly decreased ability

Keywords:

to survive phagocytosis. Total haemocyte count decreased sharply by 50-fold from days

Acanthamoeba castellanii

1e3 post-inoculation in insects infected with WT, compared to a 5-fold decrease in the

Entomopathogenic fungi

Dpks15 mutant. The mutant also produced fewer hemolymph hyphal bodies than WT by

Insect pathogenesis

3-fold. In co-culture studies with amoebae that have phagocytic ability similar to that of

Phagocytosis

insect haemocytes, at 48 h the mortality rate of amoebae engulfing Dpks15 decreased by

Polyketides

72 %, and Dpks15 CFU decreased by 83 % compared to co-culture with WT. Thus, the

Reducing clade III polyketide

Dpks15 mutant had a reduced ability to cope with phagocytosis and highly reduced viru-

synthase

lence in an insect host. These data elucidate a mechanism of insect pathogenesis associated with polyketide biosynthesis. ª 2017 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ66 2 5646700; fax: þ66 2 5646707. E-mail address: [email protected] (A. Amnuaykanjanasin). http://dx.doi.org/10.1016/j.funbio.2017.04.007 1878-6146/ª 2017 British Mycological Society. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

2

Introduction Microorganisms synthesize a large and diverse array of secondary metabolites. These secondary metabolites supply defensive or offensive small molecules for the producing microbe to fit their ecological niche. Microorganisms producing these compounds experience remarkable advantages over nonproducing ones by gaining access to and utilizing nutrient sources that are inaccessible to other groups of microorganisms. In fungi, there have been several fungal genomes reported, many of which are rich in secondary metabolite biosynthetic gene clusters (Xiao et al. 2012; Inglis et al. 2013). Nevertheless, few of these clusters have been characterized regarding their roles in their interaction with other organisms, with the surrounding environment, or in normal growth and development. Fungal secondary metabolites include polyketides, which are synthesized by multifunctional enzymes called polyketide synthases (PKSs) using acetyl-CoA as the starter unit and malonyl-CoA as extenders (Khosla 2009). Three major domains of PKSs include b-ketosynthase (KS), acyltransferase (AT), and acyl carrier protein (ACP). Modification of a newly formed polyketide is then mediated by accessory domains including but not limited to b-ketoreductase (KR), dehydratase (DH), and trans-acting enoyl (ER) domains (Schwarzer & Marahiel 2001; Staunton & Weissman 2001). PKSs are classified into three types: types I, II, and III (Hertweck 2009; Weissman 2009). Fungal PKSs are type I and can be further divided into two major classes: nonreducing (NR; subclades I-III) and reducing (R; I-VIII) (Kroken et al. 2003; Punya et al. 2015). Entomopathogenic fungi are a remarkable and beneficial group of microbes used for the biocontrol of insects. In the absence of insect hosts, they can be saprobic in the soil. However, upon contact with the insect cuticle, these fungi switch to their pathogenic phase and cause mycosis (Roberts & Hajek 1992; Bidochka et al. 2000). Beauveria bassiana is one of the most well-known and studied entomopathogenic fungi used for insect control. The fungus has a broad host range and is widely used as a biocontrol agent for agricultural pests, such as white flies, aphids, mealybugs, thrips (Shah & Goettel 1999), corn earworms and beet armyworms (Wraight et al. 2010). Beauveria bassiana infects insect hosts upon contact, distinguishing its group from entomopathogenic bacteria and viruses that require ingestion by the host. After fungal conidia adhere to the host cuticle, they can germinate and grow given favourable environmental conditions such as pH, temperature, and host surface compounds (St Leger et al. 1986). The growing hyphae then penetrate into the haemocoel, form hyphal bodies that can evade the host immune system, and develop into the vegetative stage (Amnuaykanjanasin et al. 2013). Finally, the fungus emerges and hyphae cover the host surface. During invasion into the haemocoel, insect hosts defend themselves using innate immune responses, including humoural and cellular systems (Lavine & Strand 2002). The cellular immune response is associated with production and mobilization of haemocytes (Russo et al. 2001), the phagocytosis ability of which is crucial for removal of infectious microbes (Borges et al. 2008). Two entomopathogenic fungi, Metarhizium anisopliae and B. bassiana, have been examined

W. Toopaang et al.

for susceptibility to phagocytosis by the soil amoeba Acanthamoeba castellanii (Bidochka et al. 2010), which has properties analogous to insect haemocytes. The results showed that these phagocytosed insect pathogens were able to survive and grow within the amoebae, leading to amoeboid death. In hostepathogen interactions, some secondary metabolites produced by fungal pathogens play a major role in successful infection and colonization of a host. In another large family of fungal secondary metabolites, nonribosomal peptides, beauvericin, and bassianolide were shown by targeted gene disruption studies to be crucial insect virulence factors for Galleria mellonella, Spodoptera exigua, and Helicoverpa zea (Xu et al. 2008, 2009). In the polyketide family, some PKS genes have been studied using a similar genetic knockout approach. A functional study of tenellin revealed that this polyketide is not important for infection of G. mellonella (Eley et al. 2007). A more recent study on the oosporein biosynthetic cluster demonstrated that this red polyketide pigment contributes to fungal virulence on an insect host, perhaps by modulating the host’s immunity (Feng et al. 2015). Previously, we reported that fungal PKSs in reducing clade III are extremely conserved among insect-pathogenic fungi, including B. bassiana, Cordyceps pseudomilitaris, Isaria javanica, Metarhizium flavoviride, and Paecilomyces tenuipes (Amnuaykan janasin et al. 2009; Punya et al. 2015). Moreover, PKS genes in this group have been found as single-copy genes in these fungal genomes, in contrast to other reducing clades, of which more than one gene can be found (Amnuaykanjanasin et al. 2009). In this study, the function of the reducing clade III PKS gene pks15 in the fungus B. bassiana BCC2660 was investigated by targeted disruption. Differences in colony morphology, radial growth, sporulation, and gene expression levels between the wild type (WT) and mutant were evaluated. To investigate the influence of the pks15-derived polyketide(s) in virulence against insects, the disrupted mutant was used in bioassays against beet armyworms (S. exigua) and silkworm (Bombyx mori). Moreover, phagocytic survival against A. castellanii was determined to study the potential effect of gene disruption on insect immune response escape. Here, we showed that deletion of pks15 in this fungal entomopathogen reduces its ability to survive phagocytosis and reduces virulence.

Materials and methods Fungal, bacterial, and amoeboid strains and culture conditions Beauveria bassiana BCC2660 was obtained from the BIOTEC Culture Collection, Thailand, and grown on half-strength PDA (Difco, USA) at 25  C for 5e7 d to obtain conidia. For production of blastospores, the fungal conidia were shaken in Sabouraud dextrose broth (Difco) supplemented with 1 % yeast extract (SDY) at 150 rpm and 25  C for 2 d. The fungus was grown in PDB for 5e7 d for genomic DNA extraction and in SDY for 3e7 d for total RNA isolation. Agrobacterium tumefaciens EHA105 was obtained from the Enzyme Technology Laboratory (BIOTEC, Thailand) and

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

The polyketide synthase gene pks15

3

Fig 1 e (A) The putative protein of pks15. (B) The Beauveria bassiana BCC2660 pks15 biosynthetic cluster predicted using antiSMASH. The cluster is located in the genomic contig BBA2660_218 and contains 12 genes.

maintained on solid LuriaeBertani (LB; Difco) at 28  C. An A. tumefaciens strain bearing the recombinant binary vector pCAM-PKS15-bar was selected and maintained on LB agar containing 50 mg l1 kanamycin (SigmaeAldrich, USA). Saccharomyces cerevisiae was obtained from the Food and Feed Laboratory (BIOTEC, Thailand) and cultured in yeast extractpeptone-dextrose broth (YPD; 1 % yeast extract, 2 % peptone, and 0.1 M dextrose). Acanthamoeba castellanii was obtained from the Faculty of Tropical Medicine, Mahidol University, Thailand. The amoeba was maintained in 10 ml peptone-yeast extract-glucose broth (PYG; 2 % peptone, 0.2 % yeast extract, 4 mM MgSO4$7H2O, 0.4 mM CaCl2$2H2O, 3.4 mM Na3C6H5O7$2H2O, 0.05 mM Fe(NH4)2(SO4)2$6H2O, 2.5 mM KH2PO4, 2.5 mM Na2HPO4, and 100 mM glucose and adjusted pH to 6.5) in static conditions at 25e30  C. The amoeboid culture was subcultured by transferring 1 ml into 10 ml fresh PYG broth every 15 d.

The reducing clade III PKS gene pks15 of B. bassiana BCC2660 was disrupted by integration of the bialaphos resistance gene bar at the NsiI site of the gene (618 bp after the start codon) (Fig 1A). The disruption vector pPKS15-bar and the Agrobacterium vector pCAM-PKS15-bar were constructed as previously described (Srisuksam et al. 2015). The pks15-disruption cassette from pCAM-PKS15-bar was transformed into B. bassiana BCC2660 using Agrobacterium tumefaciens EHA105 and a modification (Srisuksam et al. 2015) of a previously described method (de Groot et al. 1998). Transformants were selected on fungal minimal medium (MM; 2 % w/v dextrose, 0.51 % w/v (NH4)2SO4 (SigmaeAldrich), 0.17 % w/ v yeast nitrogen base without amino acid (Difco), 1.8 % w/v agar) supplemented with 200 mg l1 glufosinate ammonium (Zhejiang Yongnong Chem, China).

Draft genome sequencing of Beauveria bassiana BCC2660

Verification of bar integration by PCR and Southern analysis

The genomic DNA of B. bassiana BCC2660 was extracted as previously described (Reader & Broda 1985). Ten mg of B. bassiana BCC2660 genomic DNA was used to prepare one SMRTbell library with insert size of approximately 10 000 bp according to the manufacturer’s instruction (Pacific Biosciences, USA), and sequencing was performed using the P6-C4 polymerase and chemistry with 360-min movie times on a PacBio RSII system (Pacific Biosciences). PacBio sequencing reads from 2 cells were corrected and de novo assembled using Canu version 1.3 (Koren et al. 2016). The assembled contigs were polished with Quiver program in SMRT Analysis version 2.3.0 (Pacific Biosciences) to obtain the highest quality genome draft.

The genomic DNA of glufosinate-resistant transformants was extracted as previously described (Reader & Broda 1985). PCR and Southern analyses were performed to verify bar integration in the transformants’ genomes. For PCR analysis, two primer pairs were used. The first primer pair Bar-100F (50 -AAGCACGGTCAACTTCCGTAC-30 ) and Bar-360R (50 -CTTCAGCAGGTGGGTGTAGA-30 ) was designed to amplify a 260-bp bar fragment. The second primer pair PKS15-minus 925F (50 and PKS15-2200R (50 TCAAGCTTGCCCGTCACTTG-30 ) 0 GCTTATCAATGTGAGCCTCGTC-3 ) was designed to amplify a 3151-bp pks15 fragment in wild-type Beauveria bassiana and a 4206-bp fragment in disruptants with the correct bar integration (Fig 1A). PCR reactions were conducted using DreamTaq

Disruption vector construction and Agrobacterium-mediated transformation of Beauveria bassiana BCC2660

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

4

W. Toopaang et al.

Fig 2 e The bar integration in pks15. (A) A schematic diagram of bar-disrupted pks15 integration in B. bassiana BCC2660. The bar cassette was integrated into pks15 at the NsiI site. (B) PCR analysis of the Dpks15 mutants of B. bassiana, A19 and A21, compared to the ectopic transformant B10 and wild type (WT). (C) PCR amplification of the bar cassette with the primer pair Bar-100F and Bar-360R. (D) Southern analysis of the pks15-knockout mutant A21 and wild type. Lane C indicates a 4.2 kb fragment amplified from the disruption vector pCAM-PKS15-bar. Sizes of DNA standards are shown in kb on the left.

Green PCR Master Mix (Thermo Scientific) and the following thermal cycling program: 5 min at 95  C; 35 cycles of 30 s at 94  C, 1 min at 55  C, and 4 min (for pks15) or 1 min (for bar) at 72  C; and 10 min at 72  C. For Southern analysis, 15 mg genomic DNA from the WT and the transformant A21 was digested to completion with the restriction enzyme EcoRI or PstI, subjected to electrophoresis in a 1 % agarose gel, then transferred and cross-linked to a nylon membrane (Hybond Nþ; GE Healthcare Bio-Sciences, U.S.A.) according to the manufacturer’s instructions. To prepare a probe for Southern blot analysis, 100 ng of a 3-kb pks15 fragment was non-radioactively labelled using an alkaline phosphatase-based system (CDP-Star; GE Healthcare BioSciences). The membrane was hybridized with the CDP-Starlabelled pks15 probe at 55  C overnight. After a high stringency wash at 55  C, the membrane was incubated with CDP-Star detection solution and exposed to X-ray film (Hyperfilm ECL; GE Healthcare Bio-Sciences) for 15e30 min.

One of the transformants verified for bar integration in pks15 was purified by isolation of single conidia (Harris 2001).

Gene expression analysis of pks15 in the disruptants and wild type Total RNA was extracted from Beauveria bassiana WT and the Dpks15 mutant using TRI Reagent (SigmaeAldrich) and treated with DNase I (Thermo Scientific, USA) to remove contaminating genomic DNA according to the manufacturers’ instructions. cDNA was generated using RevertAid Reverse Transcriptase and random hexamers (Thermo Scientific) and used as the template for PCR amplification with two pks15-specific primer pairs, PKS15-Start (50 -ATGCTCATCGACAAAA TGGAG-30 ) and PKS15-483R (50 -CAAGCTTCCGGTACGATAGTC30 ), located upstream of the bar insertion site, and PKS15-Start and PKS15-1182R (50 -CAGCATGAGAGTAGACTTGATGAC-30 ), located downstream of the bar insertion site (Fig 2A). PCR

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

The polyketide synthase gene pks15

amplification of the b-tubulin gene was performed with the primer pair Bt2a (50 - GGTAACCAAATCGGTGCTGCTTTC-30 ) and Bt2b (50 -ACCCTCAGTGTAGTGACCCTTGGC-30 ; Glass & Donaldson 1995) using DreamTaq Green PCR Master Mix (Thermo Scientific) and the following thermal cycling program: 2 min at 95  C; 30 cycles of 1 min at 94  C, 1 min at 55  C, and 1 min at 72  C; and 10 min at 72  C.

Fungal growth and developmental analysis Radial growth of Dpks15 mutants was compared to that of WT. For each strain, a few hyphae were excised at the margin of a 7-d-old colony under a stereo microscope. This agar piece that had a few hyphae was transferred to fresh PDA and incubated at 25  C. The average diameter of the colonies was measured in millimetres at 3, 5, 7, and 9 d after inoculation. There were five replicates for each strain, and the experiment was repeated twice. Conidial and blastospore yields were determined in 5-d-old half-strength PDA and in 2-d-old SDY cultures, respectively. For PDA cultures, a conidial suspension of 1  107 conidia in 100 ml sterile water was spread onto each culture, and the culture was incubated at 25  C. For SDY cultures, a conidial suspension of 1  107 conidia in 100 ml sterile water was inoculated into each culture, and the culture was shaken at 150 rpm at 25  C. There were three replicates for each strain, and the experiment was repeated twice. Germination analysis was performed by incubation of conidia in 200 ml of 5 % (v/v) PDB for 20 h. A conidial suspension of 2.5  106 conidia was inoculated for each replicate. Conidia were considered to have germinated when the germinating tube was longer than the diameter of the conidium. There were three replicates for each strain, and the experiment was repeated twice.

Bioassay of virulence against insects Fungal conidia were harvested in saline (0.85 % NaCl), and the density was adjusted to 1  105 or 107 cells$ml1 using a haemocytometer. Fourth-instar beet armyworm (Spodoptera exigua) and silkworm (Bombyx mori) larvae were injected with 3 ml conidial suspension of each fungal strain using a specialized 33-gauge needle-syringe set (Hamilton, USA). The two densities of conidial suspension above were used for inoculation of beet armyworm larvae, which were equivalent to 300 and 30 000 conidia per larva, respectively, but only the density of 1  107 cells$ml1 was used for injection of silkworm larvae. Injected larvae were transferred individually into a 24-well plate and fed with armyworm medium (in 1 L: (w/ v) 13 % mung bean, 1 % brewer’s yeast, 0.3 % ascorbic acid, 0.125 % sorbic acid, 0.25 % methyl paraben, 0.3 % casein, 0.15 % vitamin E, 0.05 % choline chloride, and 1.2 % agar; (v/ v) 0.3 % vitamin stock and 0.08 % formalin) (Abdullah et al. 2000) or mulberry leaves for beet armyworms and silkworms, respectively. Saline-injected worms were used as controls. Ten insect larvae were treated for each fungal strain and the saline control, and the experiment was repeated three times. Larval mortality was determined on d 1e7 postinoculation.

5

Total haemocyte count and Beauveria bassiana hyphal bodies in the hemolymph Spodoptera exigua larvae were inoculated with the conidial suspension of either B. bassiana WT or Dpks15 mutant, as described above. For total haemocyte count, five ml of the hemolymph of each treated insect larva was withdrawn using an 18-gauge needle-syringe set and mixed with 45 ml of phosphate buffer saline (PBS). The total haemocyte density in the cell suspension was then determined using a haemocytometer. The total haemocyte count was performed on days 1, 2, and 3 after inoculation. For CFU counts of B. bassiana hyphal bodies, the cell suspension above was diluted to 1/100 and 1/ 1000. At the most diluted suspension, 50 out of 100 ml of each sample was spread on PDA and incubated at 28  C for 3e7 d or until fungal colonies appeared. The fungal CFU count was conducted on day 2 after inoculation. Ten insect larvae were treated for each fungal strain and the saline control, and the experiment was repeated twice.

Phagocytic survival assay Blastospores of the Dpks15 mutant and WT were determined for phagocytic survival with Acanthamoeba castellanii as previously described (Bidochka et al. 2010), and the yeast Saccharomyces cerevisiae was used as a control. The Beauveria bassiana WT and mutant were grown in SDY broth. Saccharomyces cerevisiae and A. castellanii were grown in YPD broth and PYG broth, respectively. Two days after inoculation, fungal cultures were filtered through sterile cheesecloth to remove mycelia. Both B. bassiana blastospores and S. cerevisiae cells were collected by centrifugation at 6729 g for 5 min, while the amoebae were collected by centrifugation at 30 g for 5 min. All harvested cells were resuspended in PBS buffer, counted by haemocytometer, and adjusted to 1  105 cells$ml1. Each fungal strain was co-cultured with A. castellanii in 96-well plates at a 1:1 ratio (100 ml each) for 0, 24, 48, and 72 h in order to determine postchallenge amoeboid survival rates and fungal CFUs. For determination of amoeboid survival rates, co-cultured samples were centrifuged at 30 g for 5 min, and the supernatant was discarded carefully. Fifty ml of each co-cultured sample was mixed with 10 ml of trypan blue dye. Living amoebae are able to exclude the dye, whereas the dead ones cannot and are stained blue. At least 50 amoebae were determined as dead or alive in four randomly selected areas on a slide at 400 magnification. For determination of CFUs, samples were collected from all co-cultures at each time point, mixed with RIPA buffer at a 1:1 ratio, and vortexed for 7 s to lyse amoeboid cells. Then, 100 ml of each sample was spread on PDA and incubated at 28  C for 3e7 d or until fungal colonies appeared. Both the experiments were repeated three times.

Transmission electron microscopy (TEM) of Beauveria bassiana cells and their interaction with amoebae Each fungal strain was co-cultured with Acanthamoeba castellanii at a 1:1 ratio (105 cells$ml1 each) in a 6-well plate for 4 h, and Saccharomyces cerevisiae was used as the control. Co-

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

6

culture samples were harvested by centrifugation at 30 g for 5 min, and the supernatant was carefully discarded. Each sample was fixed as described above, embedded in epoxy resin (EPON Resin; Epoxy Resins, U.S.A.), and sliced into 70e90 nm sections using an Ultracut E microtome (Reichert, Austria) with a diamond knife (Pelco, U.S.A.). Sections were subsequently stained with aqueous uranyl acetate and lead citrate (Electron Microscopy Sciences, USA). Transmission electron microscopy (TEM) was performed using the JEM-1230 (Jeol, Japan) transmission electron microscope at an 80-kV accelerating voltage.

Statistical analysis of experimental data All the experiments in this study were repeated two or three times. There were three replicates for each fungal strain in a given experiment. Pairwise comparisons of the data between the WT and the Dpks15 mutant were analyzed for statistical significance using Student’s t-test. Using the SPSS package version 11.5 (SPSS Inc., USA), LD50 and LT50 were determined using Probit analysis, and the data from the amoeba assay were analyzed by ANOVA.

Results The reducing clade III PKS gene pks15 and its cluster have unique sequence features Some differences existed in the sequence of a PKS gene between our Beauveria bassiana strain, BCC2660, and another strain whose genome has been sequenced, ARSEF2860 (Xiao et al. 2012). The B. bassiana BCC2660 tenellin synthase gene had a 53-bp longer intron than that in ARSEF2860 (Punya et al. 2015). Therefore, we performed the whole genome sequencing of B. bassiana BCC2660, which generated 2.83 Gb of polymerase read length. After removing adaptor sequences, the total length of subreads was 2.44 Gb, which is 72 coverage of the 35-Mb estimated B. bassiana genome size. We selected the 83 622 longest reads (30 coverage) as seed reads for base-correction and de novo assembly. After de novo assembly and genome polishing, we obtained 36 contigs covering 34.6 Mb with N50 length of 2.64 Mb and an average contig length of 0.96 Mb. The draft whole genome sequence of B. bassiana BCC2660 was deposited at GenBank under the accession number MWYT00000000. The genomic contig BBA2660_218 where pks15 is located was analyzed for the pks15 open reading frame (ORF) and its 50 and 30 noncoding sequences as well as the pks15 biosynthetic cluster. Sequence analysis of pks15 in B. bassiana BCC2660 showed that this reducing clade III PKS gene has an ORF of 7188 bp in length (GenBank accession number KU990876). We compared its genomic and cDNA sequences and determined that the ORF is interrupted by one intron (57 bp). The intron has the characteristic 50 (GT) and 30 (AG) intron splicing sites of eukaryotic genes (Jacobs & Stahl 1995). The presence of this single intron in B. bassiana BCC2660 pks15 contrasts with the two putative introns (57 and 228 bp) reported in a pks15 homologue in B. bassiana ARSEF2860 (Xiao et al. 2012) (GenBank accession number EJP70141). In addition, the 50 noncoding

W. Toopaang et al.

sequence of B. bassiana BCC2600 pks15 has a unique feature: multiple repeats of CATA/CATT, located at positions 26 to 81 upstream of the start codon (Supplementary Fig S1). The nucleotide sequence of B. bassiana BCC2660 pks15 had 98e99 % identity to those of B. bassiana ARSEF2860 and D1-5. The putative PKS15 protein consists of 2376 amino acids and has the KS, AT, DH, ER, KR, and ACP domains arranged in order from the N-terminus to the C-terminus (Fig 1A). The KR-DH-KR domains form a complete set of reducing domains in any PKS, making PKS15 a highly reducing PKS enzyme. In addition, we found a domain that is similar to the aspartate aminotransferase (AAT) superfamily (fold type I) of pyridoxal phosphate (PLP)-dependent enzymes (Milano et al. 2013) (Fig 1A). The pks15 cluster in B. bassiana BCC2660 has been predicted using antiSMASH (Medema et al. 2011) and has 12 genes spanning over 46 770 bp of genomic contig BBA2660_218 (Fig 1B). This biosynthetic cluster arrangement is similar to that in B. bassiana ARSEF2860 (Xiao et al. 2012). The B. bassiana BCC2660 pks15 cluster consists of a bromodomain-containing protein, a small nucleolar ribonucleoprotein complex component, three hypothetical proteins, an amino acid permease, PKS15, an EF-hand calcium-binding domain protein, a uridine diphosphate (UDP)-glucosyltransferase (relatively rare among fungal biosynthetic clusters), a fungal and bacterial-specific protein, a Zn(II)2Cys6 (C6) transcription factor, and a major facilitator superfamily (MFS) transporter (Fig 1B).

The gene pks15 was disrupted in Beauveria bassiana BCC2660 using Agrobacterium-mediated transformation To study the importance of pks15 in B. bassiana BCC2660 virulence, the gene was disrupted by transformation with the Agrobacterium vector pCAM-PKS15-bar. Two Dpks15 mutants out of twenty-eight transformants were obtained after screening for bar integration in pks15. Among these, PCR amplification of genomic DNA using primers binding outside the construct gave a PCR product of 4116 bp for transformants A19 and A21, while transformant B10 and the WT yielded a product of 3156 bp (Fig 2A and B), suggesting that A19 and A21 had correct integration of bar in the targeted PKS gene. Internal amplification of the bar cassette in the transformants demonstrated that all three transformants (Fig 2C) as well as 25 others (data not shown) had bar, while the gene was absent in the wild type (Fig 2C). Southern analysis of the WT and transformant A21, which was selected for further study, verified that pks15 in A21 has been interrupted with the selectable marker gene bar, and this bar insertion was a single integration event (Fig 2D). Because of these PCR and Southern analyses, we concluded that B. bassiana strains A19 and A21 are Dpks15 mutants. There were no noticeable differences in cell shape or colony colour between the Dpks15 mutants and the WT. The Dpks15 mutant A21 was subjected to isolation of single conidia, and its single conidial isolate was used in all experiments for characterization of the pks15-disrupted mutant.

The Dpks15 mutant has no functional copy of pks15 To compare pks15 gene expression between Dpks15 mutants and WT, two pks15-specific primer pairs were used to amplify

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

The polyketide synthase gene pks15

a region downstream of the NsiI site (PKS15-start and PKS151182R) and a region upstream of the bar integration site (NsiI) (PKS15-start and PKS15-483R) (Fig 3A). Amplification of cDNA from two single conidial isolates of the Dpks15 mutant, A21.2 and A21.4, from the pks15 start codon to the region downstream of the bar integration site, yielded no PCR product at either 3 or 7 d post-incubation. However, 1125-bp and 1182-bp PCR products were seen as expected with WT cDNA and genomic DNA, respectively, with the length difference due to the presence of an intron (Fig 3B). PCR amplification from the start codon to the region upstream of the bar integration site generated the expected 426-bp PCR product from A21.2, A21.4 and WT cDNA, and a 483-bp PCR product from WT genomic DNA (Fig 3C). These results together demonstrate that the mutants A21.2 and A21.4 cannot express a functional copy of pks15 as a result of bar integration in the NsiI site of the gene.

Growth and developmental phenotypes of the Dpks15 mutant The Dpks15 mutant had slightly reduced radial growth on MM supplemented with glufosinate compared to the WT on MM (data not shown). On PDA (without any selective drugs), the Dpks15 mutant had a slight reduction (13e15 %) in radial growth, but only at 9 d after inoculation (Table 1). The mutant produced significantly fewer spores: 62 % fewer ( p ¼ 0.01) aerial conidia (PDA) and 76 % fewer ( p ¼ 0.02) blastospores (SDY) (Table 1). In addition, the mutant conidia germinated significantly ( p ¼ 0.0015) less by 52 % than the wild type (Table 1).

Attenuation of virulence against insects in the Dpks15 mutant Changes in virulence against insects between the Beauveria bassiana WT and Dpks15 mutant were tested in insect larvae.

7

At the lower conidial inoculum of 300 conidia per larva, injection of WT resulted in low initial insect mortality with only 10 % by third day post injection, but it then increased to 43 % mortality on day 4 and 80 % insect death by day 7. In contrast, injection of the mutant resulted in neither mycosis nor insect death during the first 5 d of the experimental period, but it slightly increased to 13 % mortality on days 6 and 7 (Fig 4A). At the higher conidial inoculum of 30 000 conidia per larva, the mutant exhibited a marked reduction in virulence by 100, 44, and 15 % at 2, 3, and 4 d post inoculation (Fig 4A). Interestingly, none of the Dpks15-injected larvae developed the white external colonization by B. bassiana hyphae, whereas the WT injection clearly resulted in external colonization. The median lethal dose (LD50) of the Dpks15 mutant was 30 269 conidia per larva, 12X more than required for the WT’s LD50 of 2386 conidia per larva. The mutant also had a higher median lethal time (LT50) at 4.0 d, compared to the WT’s LT50 at 2.8 d, at the inoculum dose of 3000 conidia per larva. Similarly, we determined fungal virulence against silkworms using injections of 30 000 conidia per larva (107 conidia$ml1). The Dpks15 mutant caused lower mortality than the WT throughout the observation period (Supplementary Fig S2). Reduction of hyphal body formation in Dpks15 mutant and a sharp decline in total haemocytes in B. bassiana WT challenge. Insects inoculated at high dosage (30 000 conidia per larva) with the WT or Dpks15 mutant were further examined for the impact of pks15 disruption on hyphal body formation and total haemocyte count. Because fungal conidia switch growth phases to form yeast-like hyphal bodies (Pendland & Boucias 1996) after encountering the insect immune response (Russo et al. 2001), we extracted hemolymph on day 2 postinoculation to quantify the number of hyphal bodies. The mutant produced fewer hyphal bodies than the WT, with our quantitative determination of these cells indicating that the number of these yeast-like cells significantly decreased by 3fold compared to the wild type ( p ¼ 0.04) (Fig 4B and C). The total haemocyte count data indicated that the insect challenge with B. bassiana WT resulted in a sharp decrease in the total number of haemocytes by 3.3-fold (from day 1 to day 2 post inoculation) and by 21.9-fold to 3.4  105 cells$ml1 (from day 2 to day 3) (Fig 4C). In contrast, challenge with the Dpks15 mutant led to a smaller decline by 3.8-fold from day 2 to day 3 post inoculation to 5.1  106 cells$ml1 (Fig 4C).

The Dpks15 mutant exhibits deficiency in phagocytic survival

Fig 3 e Comparative gene expression analysis of pks15 between Dpks15 mutants (two single conidial isolates A21.2 and A21.4) and wild type (WT) in culture. (A) Locations of the two primer pairs used in this analysis. (B) PCR amplification with PKS15-start and PKS15-1182R. (C) PCR amplification with PKS15-start and PKS15-483R. (D) PCR amplification of the tubulin gene as a reference.

Amoebae, which have the same phagocytic property as haemocytes, were used in a phagocytic survival assay to assess the impact of the Dpks15 mutation on immune escape. Using microscopy, we observed that amoebae began phagocytosing both Beauveria bassiana blastospores and Saccharomyces cerevisiae cells within 1e2 h of exposure. These fungal cells adhered to the amoebae and were subsequently engulfed, with most being contained in lysosomes. Several S. cerevisiae cells were engulfed per amoeba, compared to 1e2 B. bassiana blastospores in each amoeba. After 24 h, nearly all phagocytosed S. cerevisiae cells disappeared, whereas most of the B. bassiana WT blastospores germinated, leading to amoeba death. In

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

8

W. Toopaang et al.

Table 1 e Effect of pks15 disruption on growth and development of the fungus Beauveria bassiana using comparison analysis between the Dpks15 mutant and the wild type. Data shown are averages ± s.e.m. Radial growtha,b (mm) at day

Fungal strain

Wild type Dpks15

Sporulation yieldb

3

5

7

9

7.7  0.3 7.0  0.2

10.3  0.0 9.8  0.2

16.1  1.3 13.8  0.3

27.9  0.1 24.3**  0.2

Conidial germinationb (%)

Number of blastospores Number of conidia (108 conidia per (1010 spores per 50 ml culture) petri dish) 6.1  0.2 2.3*  0.4

12.9  0.6 3.1*  0.1

81  4.7 39**  0.7

a Fungal cells were grown on PDA. b Student’s t-test was used to compare values from the wild type and Dpks15 mutant:*, p < 0.05; and**, p < 0.01. Significance p-values were as follows: conidial yield ( p ¼ 0.0105); blastospore yield ( p ¼ 0.0240); germination ( p ¼ 0.0015); and 9-d radial growth ( p ¼ 0.0058).

contrast, amoebae that phagocytosed Dpks15 mutants survived. At 48 h post-incubation, B. bassiana WT replicated to produce extensive mycelial networks and numerous blastospores (data not shown). In contrast, the Dpks15 mutant formed considerably smaller hyphal networks and noticeably fewer blastospores only (data not shown). Results from the amoeba killing assay indicated that coculture of amoebae with the Dpks15 mutant led to significantly lower amoeba mortality rates than that with WT throughout the experimental period; 66 (t-test, p ¼ 0.03), 72 ( p ¼ 0.03) and 52 % ( p ¼ 0.02) lower at 24, 48, and 72 h post-incubation, respectively (Fig 5A). In addition, the mortality rates of Dpks15 mutant-co-cultured amoebae were not significantly (ANOVA, a ¼ 0.05) distinguishable from those of S. cerevisiaeco-cultured amoebae or the pure amoeba culture. The fungal CFU data were correlated with the amoeboid mortality rates. Co-cultures of amoebae with the Dpks15 mutant and S. cerevisiae had significantly (ANOVA, p ¼ 0.0005) lower fungal CFUs than those co-cultured with the wild-type B. bassiana (Fig 5A). At 48 h post-incubation, for instance, the fungal CFU in the Dpks15-amoeba co-culture was 17 % of that in the WT co-culture.

Interaction between Beauveria bassiana and Acanthamoeba castellanii The cellular interaction between amoebae and WT or Dpks15 mutant B. bassiana was observed by TEM. Four hours after amoeba co-culture, all fungal blastospore walls were lysed, resulting in leaky cell walls (Fig 5B). Lysosomes, which are involved in digesting foreign bodies and degrading organelles, were abundant in the cytoplasm. The images showed 4e6 Dpks15 blastospores phagocytosed and digested by each amoeba, similar to Saccharomyces cerevisiae, and the amoebae appeared to be healthy (Fig 5B). Noticeably, only one to two of the wild-type blastospores were phagocytosed and digested per amoeba, and the amoebae appeared degraded. Furthermore, WT-challenged amoebae looked apoptotic with chromatin condensation (white arrowheads), membrane blebbing, and apoptotic bodies (black arrows in Fig 5B).

Discussion Among several entomopathogenic fungi, the reducing clade III PKS gene pks15 is a member of a highly conserved group in the

PKS family. Here, we demonstrated using the genetic knockout approach that pks15 in Beauveria bassiana BCC2660 is critical for virulence against insects. The Dpks15 mutant has lost nearly all virulence against Spodoptera exigua as shown by intrahemocoelic injection of 300 conidia per larva. A 100-fold increase in fungal inoculum per host resulted in higher death rates of insects infected with Dpks15, but insect mortality remained markedly and consistently lower than that from the WT in the early stages of infection, showing delayed insect pathogenicity and virulence. Interestingly, the Dpks15 mutant-injected larvae that died did not develop external white fungal colonization as seen with the WT injections. Therefore, these results indicate that the pks15 metabolite is a crucial virulence factor for B. bassiana pathogenesis in its insect host. The pks15 metabolite could be among a relatively small group of secondary metabolites that are associated with fungal growth and development. The expression of pks15 was constitutive in nearly all the culture conditions (Punya et al. 2015). Thus, it could be speculated that this secondary metabolite contributes to growth and development in this entomopathogenic fungus, in addition to its roles in insect pathogenesis. The impairment in sporulation of conidia and blastospores found in the Dpks15 mutant supports this hypothesis. Among other growth and development-associated secondary metabolites, melanin is a notable example, which is synthesized by a non-reducing polyketide synthase in several fungi, including Colletotrichum lagenarium (Takano et al. 1995) and Pestalotiopsis microspora (Yu et al. 2015). Apart from the albino appearance of these mutants, the C. lagenarium Dpks1 mutant was also drastically weakened in penetration of host plant tissue (Takano et al. 1995), whereas the P. microspora Dpks1 mutant’s conidia formed three main-body cells instead of the normal five-celled conidia as in the wild type (Yu et al. 2015). Thus, the deletion of a PKS gene can affect the normal growth and development of a fungal pathogen as well as its ability to infect a host. We set out to investigate the mechanisms by which PKS15 mediates virulence against insects. We focused on the first line of host defense, the cellular immune response by phagocytosis, which is a principal determining factor for the outcome of infection by an invader. The soil-dwelling amoeba Acanthamoeba castellanii has been used as a model in studies of interactions between this protozoa and different pathogens, including Legionella pneumophila (the bacterial causal

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

The polyketide synthase gene pks15

9

Fig 4 e (A) Virulence of Dpks15 mutants and wild type (WT) against S. exigua using a low fungal inoculum (1 3 105 conidia$mlL1) or a high dosage (1 3 107 conidia$mlL1). Mortality rates (%) of beet armyworm larvae were determined after injection. (B) In vivo hyphal bodies (indicated with arrows). Bar, 10 mm. (C) Quantification of total haemocyte count at 1, 2, and 3 d post-inoculation and the hyphal bodies at 2 d post-inoculation in the hemolymph of the wild type and Dpks15 mutant. Total haemocyte count data are shown in a log10 scale. Data shown are mean ± s.e.m. Means with different letters are significantly different (Scheffe’s, p < 0.05). The asterisks indicate statistical significance ( p < 0.05).

agent of pneumonia; reviewed in Escoll et al. 2013), Cryptococcus neoformans (the opportunistic fungal pathogen; Steenbergen et al. 2001), and Metarhizium anisopliae and B. bassiana (two fungal entomopathogens; Bidochka et al. 2010). In L. pneumophila-related studies, amoebae undergo molecular mechanisms similar to those in macrophages throughout the phagocytosis process (attachment, internalization, inclusion into phagosomes, avoidance and survival in lysosomes, and pathogen exit of the host cell). The actin-binding protein coronin is conserved between amoebae and macrophages in

phagocytosis of this bacterial pathogen (Yan et al. 2005). In addition, the mitogen-activated protein kinase (MAPK) signalling cascade is activated in both macrophage and amoeba upon L. pneumophila infection (Li et al. 2009; Clarke et al. 2013). Lastly, analysis of the A. castellanii genome suggested that this amoeboid species has a diverse array of pattern-recognition receptors (PRRs) for recognition of various microbial invaders, which are highly similar to the innate immune responses of higher eukaryotes (Clarke et al. 2013). Although the shared properties between amoebae and macrophages in innate

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

10

W. Toopaang et al.

Fig 5 e Phagocytic survival of B. bassiana blastospores by A. castellanii. (A) Mortality rates of amoebae (%) after incubation with wild type B. bassiana (A D Bb), the Dpks15 mutant (A D Dpks15), and S. cerevisiae cells (A D Sc). A pure culture of A. castellanii (A) was used as the control. (B) Fungal CFU of B. bassiana or S. cerevisiae cells after incubation with A. castellanii. Data shown are mean ± s.e.m. Means with different letters are significantly different (Scheffe’s, p < 0.05). (C) TEM micrographs of amoebae-fungal cells interaction at 4 h incubation time. Chromatin condensation is indicated by white arrowheads; lysed wall sites (black arrowheads); and membrane blebbing (black arrows). Bars, 2 mm.

immunity are widely established, the link between amoebae and insect haemocytes is notably less studied. However, it is tempting to assume that because arthropods are in the middle of the evolutionary tree between more primitive protozoa and more advanced vertebrate animals, insect haemocytes could share properties with those of amoebae and macrophages. In this study, our amoeboid phagocytic assay yielded results consistent with a previous report where M. anisopliae and B. bassiana spores were adhered to and engulfed by amoebae within 2 h and subsequent fungal germination and growth caused amoeboid death (Bidochka et al. 2010). We found that B. bassiana WT-induced death of phagocytic amoebae peaked 48 h after challenge, whereas the Dpks15 mutant was unable to cause any lethal effect on the amoebae and was comparable to the yeast Saccharomyces cerevisiae in this measure of phagocytic survival ability. Beauveria bassiana WT cells also survived and replicated to high levels 48 h after the challenge, in contrast to the low fungal CFUs for the Dpks15 mutant. Furthermore, our hemolymph observation data strongly correlated with the amoeboid phagocytic assay data. The drastic decline in total haemocyte count in the insects infected with B. bassiana WT was consistent with the high amoeboid death when challenged with the WT. These findings suggest that the B. bassiana hyphal bodies interacted with and likely killed the phagocytic haemocytes and amoebae. Additionally, a notable reduction of hyphal body number occurred in insects inoculated with the Dpks15 mutant, consistent with the marked decrease of Dpks15 hyphal bodies mixed with the amoebae compared to those of WT. The pks15 deletion attenuated the ability of fungal cells to cope with insect haemocytes, presumably in the hemolymph, or phagocytic amoebae.

The sequence analysis of B. bassiana BCC2660 pks15 and its biosynthetic cluster led to two notable findings. First, a UDPglucosyltransferase is found in the pks15 genetic cluster. The presence of glycosyltransferase(s) is common in bacterial polyketide synthase clusters as an immune strategy (e.g., the oleandomycin biosynthetic cluster of Streptomyces antibioticus (Bolam et al. 2007) and the incednine biosynthetic gene cluster of Streptomyces sp. ML694-90F3 (Takaishi et al. 2013). Thus, this is, to the best of our knowledge, the first report of a glycosyltransferase gene in a fungal polyketide synthase cluster. It is noted that fewer than ten glycosylated secondary metabolites have been reported so far (Krasnoff et al. 2014; Schueffler & Anke 2014; Zhang et al. 2016). We speculate that the glycosyltransferase in the pks15 cluster may be important for regulating the potency of the PKS15-catalyzed product. Second, our sequence analysis of up to 500 bp of promoters of pks15 and its homologues in other fungi revealed that the CATT repeats are conserved only in B. bassiana strains, not in the other two fungi (Supplementary Fig S1). The CATT tetranucleotide repeats have also been found in the promoter of the macrophage migration inhibitory factor gene (Nimer et al. 1990), with the number of CATT repeats modulating promoter activity. The presence and the number of CATT repeats may be important for expression of this reducing clade III PKS gene. Nonetheless, further study is necessary for determining the function of these CATT repeats in B. bassiana reducing clade III PKS genes. In conclusion, the Dpks15 mutant had a reduction in fungal ability to cope with phagocytosis, severely attenuating its virulence against insects. More experiments are needed to strengthen the concept of these pathogenic mechanisms mediated by PKS15.

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

The polyketide synthase gene pks15

Acknowledgements This research was financially supported by the International Foundation for Science’s First and Second Grants, NSTDA’s CPM Program, and BIOTEC. We greatly thank Dr Thareerat Kalabaheti, Department of Microbiology and Immunology, Faculty of Topical Medicine, Mahidol University, Thailand, for kindly providing Acanthamoeba castellanii. We are grateful to the Enzyme Technology Laboratory and the Food and Feed Laboratory (BIOTEC) for providing Agrobacterium tumefaciens strain EHA105 and Saccharomyces cerevisiae respectively. We highly appreciate Dr Samaporn Teeravechyan for critical reading of the manuscript. We are indebted to Dr Lynn Epstein for her advice on statistical analyses and for critical reading of this manuscript.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.funbio.2017.04.007.

references

Abdullah MD, Sarnthoy Q, Tantakom S, Isichaikul S, Chaeychomsri S, 2000. Monitoring insecticide resistance de€ bner) velopment in beet armyworm, Spodoptera exigua (Hu (Lepidoptera: Noctuidae). Kasetsart Journal (Natural Science) 34: 450e457. Amnuaykanjanasin A, Jirakkakul J, Panyasiri C, Panyarakkit P, Nounurai P, Chantasingh D, Eurwilaichitr L, Cheevadhanarak S, Tanticharoen M, 2013. Infection and colonization of tissues of the aphid Myzus persicae and cassava mealybug Phenacoccus manihoti by the fungus Beauveria bassiana. BioControl 58: 379e391. Amnuaykanjanasin A, Ponghanphot S, Sengpanich N, Cheevadhanarak S, Tanticharoen M, 2009. Insect-specific polyketide synthases, potential PKS-NRPS hybrids, and novel PKS clades in tropical fungi. Applied and Environmental Microbiology l75: 3721e3732. Bidochka MJ, Clark DC, Lewis MW, Keyhani NO, 2010. Could insect phagocytic avoidance by entomogenous fungi have evolved via selection against soil amoeboid predators? Microbiology 156: 2164e2171. Bidochka MJ, Kamp AM, De Croos AMN, 2000. Insect pathogenic fungi: from genes to populations. In: Kronstad JW (ed.), Fungal Pathology. Kluwer Academic Publishers, The Netherlands, pp. 171e193. Bolam DN, Roberts S, Proctor MR, Turkenburg JP, Dodson EJ, Martinez-Fleites C, Yang M, Davis BG, Davies GJ, Gilbert HJ, 2007. The crystal structure of two macrolide glycosyltransferases provides a blueprint for host cell antibiotic immunity. Proceedings of the National Academy of Sciences of the United States of America 104: 5336e5341. Borges AR, Santos PN, Furtado AF, Figueiredo RCB, 2008. Phagocytosis of latex beads and bacteria by hemocytes of the triatomine bug Rhodnius prolixus (Hemiptera: Reduvidae). Micron 39: 486e494. Clarke M, Lohan AJ, Liu B, Lagkouvardos I, Roy S, Zafar N, € rglin TR, Frech C, Bertelli C, Schilde C, Kianianmomeni A, Bu Turcotte B, Kopec KO, Synnott JM, Choo C, Paponov I, Finkler A, Heng Tan CS, Hutchins AP, Weinmeier T, Rattei T, Chu JS, Gimenez G, Irimia M, Rigden DJ, Fitzpatrick DA,

11

Lorenzo-Morales J, Bateman A, Chiu CH, Tang P, Hegemann P, Fromm H, Raoult D, Greub G, Miranda-Saavedra D, Chen N, Nash P, Ginger ML, Horn M, Schaap P, Caler L, Loftus BJ, 2013. Genome of Acanthamoeba castellanii highlights extensive lateral gene transfer and early evolution of tyrosine kinase signaling. Genome Biology 14: R11. de Groot MJ, Bundock P, Hooykaas PJ, Beijersbergen AG, 1998. Agrobacterium tumefaciens-mediated transformation of filamentous fungi. Nature Biotechnology 16: 839e842. Eley KL, Halo LM, Song Z, Powles H, Cox RJ, Bailey AM, Lazarus CM, Simpson TJ, 2007. Biosynthesis of the 2-pyridone tenellin in the insect pathogenic fungus Beauveria bassiana. Chembiochem 8: 289e297. Escoll P, Rolando M, Gomez-Valero L, Buchrieser C, 2013. From amoeba to macrophages: exploring the molecular mechanisms of Legionella pneumophila infection in both hosts. Current Topics In Microbiology and Immunology 376: 1e34. Feng P, Shang Y, Cen K, Wang C, 2015. Fungal biosynthesis of the bibenzoquinone oosporein to evade insect immunity. Proceedings of the National Academy of Sciences of the United States of America 112: 11365e11370. Glass NL, Donaldson GC, 1995. Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes. Applied and Environmental Microbiology 61: 1323e1330. Harris SD, 2001. Genetic analysis of ascomycete fungi. In: Talbot N (ed.), Molecular and cellular biology of filamentous fungi: A Practical Approach. Oxford University Press, UK, pp. 47e58. Hertweck C, 2009. The biosynthetic logic of polyketide diversity. Angewandte Chemie International Edition in English 48: 4688e4716. Inglis DO, Binkley J, Skrzypek MS, Arnaud MB, Cerqueira GC, Shah P, Wymore F, Wortman JR, Sherlock G, 2013. Comprehensive annotation of secondary metabolite biosynthetic genes and gene clusters of Aspergillus nidulans, A. fumigatus, A. niger and A. oryzae. BMC Microbiology 13: 91. Jacobs M, Stahl U, 1995. Gene regulation in mycelial fungi. In: Kueck U (ed.), The Mycota, II; Genetics and Biotechnology. Springer-Verlag, New York, USA, pp. 155e167. Khosla C, 2009. Structures and mechanisms of polyketide synthases. The Journal of Organic Chemistry 74: 6416e6420. Koren S, Walenz BP, Berlin K, Miller JR, Phillippy AM, 2016. Canu: scalable and accurate long-read assembly via adaptive k-mer weighting and repeat separation. bioRxiv. http: //dx.doi.org/10.1101/071282. Krasnoff SB, Keresztes I, Donzelli BG, Gibson DM, 2014. Metachelins, mannosylated and N-oxidized coprogen-type siderophores from Metarhizium robertsii. Journal of Natural Products 77: 1685e1692. Kroken SN, Glass L, Taylor JW, Yoder OC, Turgeon BG, 2003. Phylogenomic analysis of type I polyketide synthase genes in pathogenic and saprobic ascomycetes. Proceedings of the National Academy of Sciences of the United States of America 100: 15670e15675. Lavine MD, Strand MR, 2002. Insect hemocytes and their role in immunity. Insect Biochemistry and Molecular Biology 32: 1295e1309. Li Z, Dugan AS, Bloomfield G, Skelton J, Ivens A, Losick V, Isberg RR, 2009. The amoebal MAP kinase response to Legionella pneumophila is regulated by DupA. Cell Host and Microbe 6: 253e267. Medema MH, Blin K, Cimermancic P, de Jager V, Zakrzewski P, Fischbach MA, Weber T, Takano E, Breitling R, 2011. antiSMASH: rapid identification, annotation and analysis of secondary metabolite biosynthesis gene clusters in bacterial and fungal genome sequences. Nucleic Acids Research 39: W339eW346. Milano T, Paiardini A, Grgurina I, Pascarella S, 2013. Type I pyridoxal 5’-phosphate dependent enzymatic domains embedded

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007

12

within multimodular nonribosomal peptide synthetase and polyketide synthase assembly lines. BMC Structural Biology 13: 26. Nimer S, Fraser J, Richards J, Lynch M, Gasson J, 1990. The repeated sequence CATT(A/T) is required for granulocytemacrophage colony-stimulating factor promoter activity. Molecular and Cellular Biology 10: 6084e6088. Pendland JC, Boucias DG, 1996. Phagocytosis of lectin-opsonized fungal cells and endocytosis of the ligand by insect Spodoptera exigua granular hemocytes: an ultrastructural and immunocytochemical study. Cell & Tissue Research 285: 57e67. Punya J, Swangmaneecharern P, Pinsupa S, Nitistaporn P, Phonghanpot S, Kunathigan V, Cheevadhanarak S, Tanticharoen M, Amnuaykanjanasin A, 2015. Phylogeny of type I polyketide synthases (PKSs) in fungal entomopathogens and expression analysis of PKS genes in Beauveria bassiana BCC 2660. Fungal Biology 119: 538e550. Reader U, Broda P, 1985. Rapid preparation of DNA from filamentous fungi. Letters in Applied Microbiology 1: 17e20. Roberts DW, Hajek AE, 1992. Entomopathogenic fungi as bioinsecticides. In: Leatham GF (ed.), Frontiers of Industrial Mycology. CRC Press, Kentucky, USA, pp. 144e159. Russo J, Brehelin M, Carton Y, 2001. Haemocyte changes in resistant and susceptible strains of D. melanogaster caused by virulent and avirulent strains of the parasitic wasp Leptopilina boulardi. Journal of Insect Physiology 47: 167e172. Schueffler A, Anke T, 2014. Fungal natural products in research and development. Natural Product Reports 31: 1425e1448. Schwarzer D, Marahiel MA, 2001. Multimodular biocatalysts for natural product assembly. Naturwissenschaften 88: 93e101. Shah PA, Goettel MS, 1999. Directory of Microbial Control Products and Services 2nded, Microbial Control Division, Society for Invertebrate Pathology. Srisuksam C, Toopaang W, Phonghanpot S, Punya J, Wattanachaisareekula S, Tanticharoen M, Cheevadhanarak S, Amnuaykanjanasin A, 2015. Enhanced efficiency of Agrobacterium-mediated transformation in Beauveria bassiana. In: Proceedings of the 4th Burapha University International Conference, Chonburi, Thailand, pp. 61e69. St Leger RJ, Cooper RM, Charnley AK, 1986. Cuticle-degrading enzymes of entomopathogenic fungi: cuticle degradation in vitro by enzymes from entomopathogens. Journal of Invertebrate Pathology 47: 167e177. Staunton J, Weissman KJ, 2001. Polyketide biosynthesis: a millennium review. Natural Product Reports 18: 380e416.

W. Toopaang et al.

Steenbergen JN, Shuman HA, Casadevall A, 2001. Cryptococcus neoformans interactions with amoebae suggest an explanation for its virulence and intracellular pathogenic strategy in macrophages. Proceedings of the National Academy of Sciences of the United States of America 98: 15245e15250. Takaishi M, Kudo F, Eguchi T, 2013. Identification of the incednine biosynthetic gene cluster: characterization of novel b-glutamate-b-decarboxylase IdnL3. Journal of Antibiotics (Tokyo) 66: 691e699. Takano Y, Kubo Y, Shimizu K, Mise K, Okuno T, Furusawa I, 1995. Structural analysis of PKS1, a polyketide synthase gene involved in melanin biosynthesis in Colletotrichum lagenarium. Molecular & General Genetics 249: 162e167. Weissman KJ, 2009. Introduction to Polyketide Biosynthesis. In: Hopwood DA (ed.), Methods in Enzymology, 1st edn., vol. 459, Elsevier Inc, California, USA, pp. 3e16. Wraight SP, Ramos ME, Avery PB, Jaronski ST, Vandenberg JD, 2010. Comparative virulence of Beauveria bassiana isolates against lepidopteran pests of vegetable crops. Journal of Invertebrate Pathology 103: 186e199. Xiao G, Ying SH, Zheng P, Wang ZL, Zhang S, Xie XQ, Shang Y, Leger RJ, Zhao GP, Wang C, Feng MG, 2012. Genomic perspectives on the evolution of fungal entomopathogenicity in Beauveria bassiana. Scientific Reports 2: 483.  r I, Xu Y, Orozco R, Wijeratne EM, Gunatilaka AA, Stock P, Molna 2008. Biosynthesis of the cyclooligomer depsipeptide beauvericin, a virulence factor of the entomopathogenic fungus Beauveria bassiana. Chemistry & Biology 15: 898e907. Xu Y, Orozco R, Wijeratne EM, Espinosa-Artiles P, Gunatilaka AA,  r I, 2009. Biosynthesis of the cyclooligomer Stock P, Molna depsipeptide bassianolide, an insecticidal virulence factor of Beauveria bassiana. Fungal Genetics and Biology 46: 353e364. Yan M, Collins RF, Grinstein S, Trimble WS, 2005. Coronin-1 function is required for phagosome formation. Molecular Biology of the Cell 16: 3077e3087. Yu X, Huo L, Liu H, Chen L, Wang Y, Zhu X, 2015. Melanin is required for the formation of the multi-cellular conidia in the endophytic fungus Pestalotiopsis microspora. Microbiological Research 179: 1e11. Zhang S, Zhu J, Liu T, Samra S, Pan H, Bai J, Hua H, Bechthold A, 2016. Myrothecoside, a novel glycosylated polyketide from the terrestrial fungus Myrothecium sp. GS-17. Helvetica Chimica Acta 9: 215e219.

Please cite this article in press as: Toopaang W, et al., Targeted disruption of the polyketide synthase gene pks15 affects virulence against insects and phagocytic survival in the fungus Beauveria bassiana, Fungal Biology (2017), http://dx.doi.org/10.1016/ j.funbio.2017.04.007