LaeA negatively regulates dothistromin production in the pine needle pathogen Dothistroma septosporum

Accepted Manuscript LaeA negatively regulates dothistromin production in the pine needle pathogen Dothistroma septosporum. Pranav Chettri, Rosie E. Bradshaw PII: DOI: Reference:

S1087-1845(16)30126-8 http://dx.doi.org/10.1016/j.fgb.2016.11.001 YFGBI 3008

To appear in:

Fungal Genetics and Biology

Received Date: Revised Date: Accepted Date:

22 August 2016 30 October 2016 1 November 2016

Please cite this article as: Chettri, P., Bradshaw, R.E., LaeA negatively regulates dothistromin production in the pine needle pathogen Dothistroma septosporum., Fungal Genetics and Biology (2016), doi: http://dx.doi.org/10.1016/ j.fgb.2016.11.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

LaeA negatively regulates dothistromin production in the pine needle pathogen Dothistroma septosporum. Pranav Chettri and Rosie E. Bradshaw. Bio-Protection Research Centre, Institute of Fundamental Sciences, Massey University, Palmerston North 4474, New Zealand Corresponding author: R.E. Bradshaw; E-mail: [email protected]

Abstract In filamentous fungi both pathway-specific and global regulators regulate genes involved in the biosynthesis of secondary metabolites. LaeA is a global regulator that was named for its mutant phenotype, loss of aflR expression, due to its effect on the aflatoxin-pathway regulator AflR in Aspergillus spp. The pine needle pathogen Dothistroma septosporum produces a polyketide virulence factor, dothistromin, that is chemically related to aflatoxin and whose pathway genes are also regulated by an ortholog of AflR. However, dothistromin biosynthesis is distinctive because it is switched on during early (rather than late) exponential growth phase and the genes are dispersed in six loci across one chromosome instead of being clustered. It was therefore of interest to determine whether the function of the global regulator LaeA is conserved in D. septosporum. To address this question, a LaeA ortholog (DsLaeA) was identified and its function analyzed in D. septosporum. In contrast to aflatoxin production in Aspergillus spp., deletion of DsLaeA resulted in enhanced dothistromin production and increased expression of the pathway regulatory gene DsAflR. Although expression of other putative secondary metabolite genes in D. septosporum showed a range of different responses to loss of DsLaeA function, thin layer chromatography revealed increased levels of a previously unknown metabolite in DsLaeA mutants. In addition, these mutants exhibited reduced asexual sporulation, germination and hydrophobicity. Our data suggest that

1

although the developmental regulatory role of DsLaeA is conserved, its role in the regulation of secondary metabolism differs from that of LaeA in A. nidulans and appears to be species specific. This study provides a step towards understanding fundamental differences in regulation of clustered and fragmented groups of secondary metabolite genes that may shed light on understanding functional adaptation in secondary metabolism.

Key words:, gene regulation; global regulator; secondary metabolism; dothistromin,

2

Introduction Filamentous fungi produce a diverse array of secondary metabolites that do not normally have a role in growth and development but can contribute towards fitness factors such as defense, communication, niche adaptation and UV protection (Trienens et al., 2010; Amare and Keller 2014). Fungal secondary metabolite genes are usually clustered and transcriptionally co-regulated by a combination of pathway-specific and global regulator proteins. For example, the pathway-specific transcription factor AflR regulates expression of genes in aflatoxin clusters in Aspergillus flavus and A. parasiticus and the sterigmatocystin cluster of A. nidulans (Chang et al., 1993; Fernandes et al., 1998). Some global regulators orchestrate regulation of both secondary metabolism and fungal development in response to environmental cues (Butchko et al., 2012; Calvo and Cary, 2015) by means of chromatin modification (Strauss and Reyes-Dominguez, 2011, Brakhage, 2013). The global regulator gene laeA (named for loss of aflR expression) is predicted to encode a methyltransferase (Bok and Keller, 2004; Sarikaya Bayram et al., 2010). Identification and characterisation of laeA in A. nidulans was an important milestone to better understand global secondary metabolite gene regulation in filamentous fungi (Bok and Keller, 2004) and its crucial function in secondary metabolism has now been documented for many fungi (reviewed by Jain and Keller, 2013). Most of the initial studies were done in Aspergillus spp., in which laeA deletion mutants showed marked reductions in production of secondary metabolites such as sterigmatocystin, aflatoxin, gliotoxin and endocrocin, (Bok and Keller, 2004; Amaike and Keller, 2009; Bok et al., 2005; Lim et al., 2012). LaeA was also shown to control both asexual and sexual development in A. nidulans (Sarikaya Bayram et al. 2010). A role of LaeA in development was also demonstrated in Fusarium fujikuroi (Wiemann et al., 2010), Cochliobolus heterostrophus (Wu et al., 2012), A. flavus (Amaike

3

and Keller 2009), Trichoderma reesei (Seiboth et al., 2012), Monascus ruber (Liu et al., 2016) and Alternaria alternata (Takao et al., 2016) LaeA occurs as part of a protein complex with the light-regulated developmental regulatory proteins VeA and VelB in A. nidulans (Bayram et al., 2008). In the presence of light, LaeA is required for proper asexual development whereas VeA and VelB are also required for sexual fruiting body development in dark conditions (Bayram and Braus, 2012). A similar type of protein complex interaction involving LaeA, VeA and VelB was shown in other fungi such as Fusarium spp., Penicillium chrysogenum, Botrytis cinerea (Wiemann et al., 2010; López-Berges et al., 2013, Hoff et al., 2010, Schumacher et al., 2015). LaeA has also been shown to have a role in virulence of pathogenic fungi. For example an A. flavus laeA mutant was significantly impaired in the ability to degrade host cell lipid reserves in peanut and also showed reduced host cell colonization compared to wild type strains (Amaike and Keller, 2009). Similarly, laeA mutants of F. fujikuroi showed reduced virulence on rice (Wiemann et al., 2010), and those of the Dothideomycete pathogens C. heterostrophus and A. alternata had reduced pathogenicity in maize and tomato, respectively, compared to the wild type (Wu et al., 2012; Takao et al., 2016). In this study we characterized the LaeA ortholog of another Dothideomycete, the forest pathogen Dothistroma septosporum. This fungus is the main causal agent of Dothistroma needle blight, one of the most devastating diseases of pine trees worldwide (Watt et al., 2009; Mullett and Fraser, 2016). The impact of D. septosporum on forest health has increased substantially in recent years, particularly in the Northern hemisphere (Drenkhan and Hanso, 2009; Woods et al., 2005). Climate change has been implicated in the increased severity of the disease (Woods, 2011), leading to efforts to understand the pathogen and its interaction with its host at the molecular level (Bradshaw et al., 2016). Identification of key molecules such as secondary metabolites that are involved in pathogenesis, and

4

understanding how the production of these molecules is regulated may lead to new methods of disease management. D. septosporum produces dothistromin, a polyketide that is chemically similar to a precursor of aflatoxin (AF) and sterigmatocystin (ST) (Gallagher and Hodges, 1972). Dothistromin is a virulence factor required for expansion of necrotic disease lesions (Kabir et al., 2015a) and certain aspects of the genetics of dothistromin biosynthesis are unusual. Whilst fungal secondary metabolite genes are generally clustered (Keller and Hohn, 1997), dothistromin biosynthetic genes are instead arranged at six separate loci spread across a 1.3Mb chromosome (Chettri et al., 2013; Bradshaw et al 2013). Furthermore, dothistromin is mainly produced during early exponential phase in culture (Schwelm et al., 2008), instead of during late exponential and stationary phases as seen for other fungal secondary metabolites, in keeping with high levels of dothistromin production during periods of rapid fungal biomass accumulation in planta (Kabir et al., 2015b). Due to its unique gene arrangement and timing of biosynthesis we were interested to determine whether LaeA has a role in regulation of the fragmented dothistromin gene cluster. To date the importance of LaeA proteins in regulation and coordination of secondary metabolism and development has not been examined in a forest foliar pathogen. As a first step towards achieving our goal we genetically characterized the LaeA ortholog in D. septosporum and showed its role in governing morphology and development. We provide evidence that LaeA could also act as a negative regulator of secondary metabolism in this fungus and also showed the presence of a previously undescribed metabolite that is suppressed by LaeA under normal in vitro conditions.

2. Materials and methods 2.1 Fungal strain and culture conditions

5

Wild-type D. septosporum NZE10 was used throughout this work. Cultures were grown at 22 °C on Dothistroma medium (DM), Dothistroma sporulation medium (DSM) or pine needle extract minimal medium (PMMG) (Chettri et al., 2012). For growth in liquid media the cultures were inoculated with 1x105 spores per ml in 125 ml flasks containing 25 ml of media and shaken at 180 rpm for seven days. To study the effect of light and dark the growth chamber had three Sylvania GRO-LUX (F30w/ GRO-TB) and two Philips-lifemax (TLD-30 W/840 cool white) lights, used with continuous illumination (light conditions) or with culture flasks double wrapped in foil (dark conditions).

2.2 Identification of Dothistroma septosporum LaeA (DsLaeA). To confirm the identity of the previously reported D. septosporum LaeA ortholog (de Wit et al., 2012), BLASTP and reciprocal BLASTP were performed using the LaeA protein sequence from A. nidulans (accession number AAQ95166). To further confirm its authenticity phylogenetic analysis was done using additional putative or confirmed LaeA orthologs from other species. Phylogenetic analysis was done using MEGA 5.0 with default parameters (Tamura et al., 2011).

2.3 Manipulation of the DsLaeA gene in Dothistroma septosporum. Genomic DNA was isolated from D. septosporum using a CTAB method (Moller et al., 1992). A DsLaeA gene knockout construct, pOscLaeA, was made via One Step Construction of Agrobacterium-Recombination-ready-plasmids (OSCAR) using PCR-based methods described previously (Paz et al., 2011). Vectors pA-HYG OSCAR and pOSCAR were purchased from the Fungal Genetics Stock Center (http://www.fgsc.net). The pOscLaeA knockout construct was made such that 1035 bp of DsLaeA coding region (i.e. nucleotides 1803893

–

1805158

of

D.

6

septosporum

Scaffold

3;

http://genome.jgi.doe.gov/Dotse1/Dotse1.home.html)

was replaced

by a hygromycin

resistance gene, flanked with 1250 bp and 1850 bp of 5' and 3' DsLaeA flanking regions to guide targeted integration. All primers used for PCR are listed in Supplementary Table 1. D. septosporum NZE10 was transformed with pOscLaeA using methods described previously (Bradshaw et al., 2006) and transformants were single-spore purified. To complement the phenotype of the D. septosporum LaeA replacement strain KO1 with DsLaeA, co-transformation was done with a 1:1 ratio of DsLaeA coding sequence flanked by 1 kb upstream and downstream sequence (i.e. nucleotides 1802651 to 1806194 of D. septosporum scaffold 3), and the linearized vector pBCphleo (pR224) that confers phleomycin resistance. Transformants were selected on 10 µg/ml phleomycin and were single-spore purified. Confirmation of DsLaeA gene replacement was determined by PCR and Southern hybridisation of AvaI-digested DNA with a digoxigenin (DIG)-labeled probe encompassing DsLaeA 3’flank and hph gene regions, following the protocol described earlier (Bradshaw et al., 2006). The copy number of DsLaeA gene in the complemented strains was determined as previously described (Chettri et al., 2015). To determine if CfLaeA, the LaeA ortholog of Cladosporium fulvum, a close relative of D. septosporum, is functionally orthologous to DsLaeA, the D. septosporum DsLaeA KO1 mutant was transformed with a plasmid (pR224) containing the phleomycin resistance gene and a 3.1 kb genomic DNA fragment containing the CfLaeA gene (nucleotides 11511–14650 of C. fulvum scaffold scf7180000128058; http://genome.jgi.doe.gov/Clafu1/Clafu1.home.html), using methods outlined above. PCR primers used for amplification and for verification of constructs and transformants are shown in Supplementary Table 1. To construct DsPksA and DsHps1 knockout plasmids (R412 & R413) used to make double mutants in a ΔDsLaeA strain, entry clones were made by amplifying a 1.945 kb

7

fragment from a pII99-based vector containing a geneticin resistance cassette (Namiki et al., 2013) then combined with an entry vector carrying ~ 1 kb of 5’ and 3’ flank of either the DsPksA or DsHps1 gene using a Multisite Gateway system (Invitrogen, Carlsbad, CA) as described previously for the DsVeA gene knockout (Chettri et al., 2012). The gene knockout schemes are shown in Supplementary Figures 9 and 10, and primer sequences in Supplementary Table 1.

2.4 Radial growth, sporulation, germination and hydrophobicity tests. Radial growth rates of the DsLaeA knockout mutant, DsLaeA complemented and wild type strains were determined in both light and dark conditions in DM agar media at 22°C. Approximately 5 mm diameter mycelium plugs from ten day old fungal colonies were plated onto DM (3 biological x 2 technical replicates per strain). Radial colony growth was measured using an Electronic Digital Caliper (Chicago Brand, Medford, OR) at 2-3 day intervals for 31 days and radial growth was calculated as mm/day. To assess sporulation, 100 μl of freshly grown spores (1 x 105/ml from 10 d DSM light-grown cultures) were spread over three PMMG plates for each strain. After ten days of growth, 5 ml of sterile water was added to the plate, spread with a sterile glass spreader and allowed to stand for 10 min with intermittent spreading. The spore suspension was poured into a 15 ml centrifuge tube and the same spore harvest steps were repeated once. The concentration of spores in the resulting suspension was quantified using a haemocytometer (Weber Scientific, Middlesex, England). The germination rate was determined by spreading 5 x 104 spores, grown as above, onto a water agar plate overlaid with cellophane membrane and sampling over a period of 48 hours with incubation at 22°C. At twelve hour intervals the percentage germination was calculated from samples of approximately 200 conidia.

8

To compare hydrophobicity, 0.2% bromophenol blue (aqueous solution) was placed on the surface of a fungal colony grown on DM or PDA media and visualised at 10, 30 and 60 min. Fungal colonies with reduced hydrophobicity showed absorption of the bromophenol blue solution. Expression of the hydrophobin Hdp1 gene (JGI: protein ID 75009) was determined by relative quantitative PCR relative to the beta-tubulin Tub1 gene as described previously (Chettri et al., 2012). Statistical significance was determined using a two-tailed T-test based on the null hypothesis of no significant difference between mutant and wild type phenotypes.

2.5 Expression analysis and dothistromin assays D. septosporum mutants and wild-type controls were grown in liquid DM or PMMG medium for 7 days at 22°C with shaking (180 rpm). Secondary metabolites were extracted from the media and the secreted DOTH quantified by HPLC using methods described previously (Chettri et al., 2012). For qualitative assessment of dothistromin and other metabolites, TLC was performed using glass-backed, 200 µm analytical Silica Gel 60 plates (Merck, Kenilworth, NJ), with a toluene:acetone (80:20) solvent containing 0.1% formic acid, as described previously (Chettri et al., 2012). RNA was extracted from mycelia grown in the same flasks as those for dothistromin analysis and gene expression assessed by quantitative real time (qRT)-PCR using methods published earlier (Chettri et al., 2013). Sequences of primers used for qRT-PCR are shown in Supplementary Table 1. Statistical significance in comparisons to the wild type where required, was determined on normalized expression data using two-tailed T-tests based on the null hypothesis of no significant difference.

3. Results

9

3.1 Identification of the Dothistroma septosporum LaeA ortholog An ortholog of the well-characterised A. nidulans laeA gene was previously predicted in the D. septosporum genome (de Wit et al 2012). To confirm its identity reciprocal BLAST with the A. nidulans LaeA protein sequence was performed. The D. septosporum LaeA candidate (JGI: Protein ID 148869) was then aligned with known LaeA protein sequences from other species, and a tree produced whose branching was consistent with taxonomic relationships between fungi (Schoch et al., 2009) (Figure 1). Phylogenetic analysis of LaeA proteins along with a broader set of putative SAM methyltransferases, including eight predicted LaeA-like methyltransferases from D. septosporum, and similar orthologs from other sequenced Dothideomycete fungi, confirmed that the D. septosporum LaeA candidate was most closely related to characterised LaeA from other fungi. (Supplementary Figure 1). In contrast other SAM methyltransferases grouped in separate clades, with some showing similarity to Llm1 and LlmF LaeA-like methyltransferases of Cochliobolus heterostrophus (Bi et al., 2013) and A. nidulans (Palmer et al., 2013). Together these results confirmed that the D. septosporum methyltransferase (JGI: Protein ID 148869) is a LaeA ortholog and the gene was named DsLaeA. The gene model of DsLaeA was verified using RNAseq data (Bradshaw et al., 2016) and revealed an open reading frame (ORF) of 1035 bp interrupted by three introns with a predicted protein of 345 amino acid residues. DsLaeA shows less than 40% amino acid identity with LaeA proteins from A. nidulans, A. parasiticus and C. heterostrophus, but 87.7% identity with the functionally characterised LaeA from C. fulvum (Griffiths et al., 2015) (Supplementary Figure 2). In A. nidulans LaeA was shown to be automethylated at a specific methionine residue 207 (Patananan et al., 2013); however, methionine 207 is absent in LaeA proteins from D. septosporum, C. fulvum and T. reesei (Supplementary Figure 2).

10

3.2 DsLaeA knockout mutants are impaired in asexual sporulation, growth rate and spore germination. Deletion of LaeA genes has been shown to affect morphological development and secondary metabolism in some fungi, therefore we deleted DsLaeA from the D. septosporum genome and determined the mutant phenotype. Protoplast-mediated transformation of D. septosporum with the gene knockout construct pR324 yielded four independent transformants that were confirmed to have targeted replacement of DsLaeA (Supplementary Figure 3). Of these, two independently isolated gene knockout strains ΔDsLaeA KO1 and ΔDsLaeA KO2 were used for further analysis. Reintroduction of the full-length DsLaeA gene into knockout strain ΔDsLaeA KO1 yielded six DsLaeA complemented strains, designated C1 to C6, which were confirmed by PCR (Supplementary Figure 3) using primers specific to the coding sequence of DsLaeA gene. Complemented strains had random integration of the introduced full-length DsLaeA gene as all six strains showed resistance to both hygromycin and phleomycin antibiotic. To investigate whether LaeA has a conserved function in its close relative C. fulvum, a biotroph pathogen of tomato (De Wit et al. 2012), the same ΔDsLaeA strain (KO1) was transformed with a functional copy of C. fulvum LaeA (CfLaeA; JGI: protein ID 186126). Based on real-time PCR determination of copy numbers, strains C1 (one copy of DsLaeA), CfC2 (six copies of CfLaeA) and CfC3 (one copy of CfLaeA) were selected for analysis; no multi-copy DsLaeA transformants were obtained. Sporulation and spore germination were impaired in ΔDsLaeA mutants. Both mutant strains produced significantly fewer spores than the wild type and their sporulation levels were unaffected by light. This was in contrast to the significantly higher levels of conidia produced in dark compared to light conditions by the wild type and complemented strains (Figure 2a). Analysis of radial growth rates revealed that ΔDsLaeA mutant strains grew more

11

slowly than the wild type under light conditions (Figure 2b) but there was no significant difference in growth rates in the dark. Similarly, the germination of ΔDsLaeA spores was significantly lower than that of wild type spores irrespective of whether spores were obtained from cultures grown in either light or dark conditions (Supplementary Figure 4). Examination of spores by light microscopy showed that those of ΔDsLaeA were slightly stunted and differed in shape compared to those of the wild type strain (Supplementary Figure 5). Expression of D. septosporum StuA (JGI: Protein ID 45737), a putative ortholog of StuA genes that regulate asexual sporulation, mycotoxin production and effector gene expression in S. nodorum and F. graminearum (Ipcho et al., 2010; Lysoe et al., 2011) was reduced in the ΔDsLaeA strain (Supplementary Figure 6), concordant with reduced sporulation in this mutant.

3.3 DsLaeA is required for normal colony hydrophobicity LaeA is known to regulate hydrophobicity in many filamentous fungi (Dagenais et al., 2010; Chang et al., 2012), thus the effect of DsLaeA deletion on hydrophobicity in D. septosporum was assessed. Bromophenol blue solution applied to colony surfaces stayed as spherical droplets on wild type (WT) and complemented (C1) strains while it was gradually absorbed by ΔDsLaeA mutants, suggesting a reduction in hydrophobicity (Figure 3a). To follow up on this observation, the expression of one of the hydrophobin genes DsHdp1 (JGI: protein ID 75009) was investigated. The D. septosporum genome contains three genes encoding hydrophobin domain-containing proteins; however, only DsHdp1 is highly expressed in planta and was previously shown to have a role in surface hydrophobicity (Bradshaw et al., 2016). Quantitative RT-PCR analysis showed that the expression of DsHdp1 was significantly reduced in the ΔDsLaeA strain compared to levels in the wild type and complemented strains (Figure 3b).

12

3.4 D. septosporum LaeA negatively regulates dothistromin biosynthesis Because LaeA is known to regulate secondary metabolite biosynthesis in many fungi we looked at the effect of DsLaeA mutation on dothistromin production. Dothistromin was produced under both light and dark conditions in all strains (Table 1). Unexpectedly, both ΔDsLaeA mutant strains (KO1 and KO2) produced significantly more (5 - 8 fold) dothistromin than the WT, irrespective of light or dark conditions in PMMG (pine extract minimal) media. The levels of dothistromin were also significantly higher than the wild type (but only ~2 fold) in the DsLaeA complementation strain C1. The experiment was repeated using a nutrient-rich growth medium (DM instead of PMMG) and again the ΔDsLaeA mutants showed significantly higher levels of dothistromin than the wild type (Supplementary Figure 7). To further confirm the role of DsLaeA as a regulator of dothistromin production, expression of key dothistromin biosynthetic and regulatory genes was examined in WT, ΔDsLaeA mutant and DsLaeA complemented strains. Apart from DsVbsA, all dothistromin biosynthesis and regulatory genes tested were up-regulated in the mutants. Dothistromin regulatory genes DsAflR and DsAflJ were up regulated by 2.5 – 7 fold, and dothistromin biosynthetic genes DsPksA and DsVer1 by ~1.6 fold in the ΔDsLaeA mutant, compared to the wild type (Figure 4). Whilst the mutant phenotypes were complemented by addition of DsLaeA (ie. in complemented strain C1) the complementation was not complete in all cases; expression of four of the five dothistromin genes tested (Figure 4), and production of dothistromin (Table 1), did not completely revert to the wild type phenotype. However, it was noted that expression of DsLaeA itself only reached 76% of wild type levels in the complemented strain (Figure 4). There was no significant difference in dothistromin biosynthesis gene expression

in light or dark conditions, for any of the strains

13

(Supplementary figure 8), as expected based on the similar levels of dothistromin produced in the presence or absence of light (Table 1). Together these results suggest that DsLaeA is a negative regulator of dothistromin production and gene expression, independent of light conditions.

3.5 Partial complementation of the ΔDsLaeA phenotype was obtained with multiple copies of Cladosporium fulvum LaeA. To determine if the increased dothistromin production observed in a ΔDsLaeA mutant could be reverted to wild type levels by LaeA from C. fulvum, the closest well-characterised relative of D. septosporum, CfLaeA complementation strains were assessed for their toxin production in light conditions. A ΔDsLaeA strain with a single copy of CfLaeA (Cf3) did not show complementation of the DsLaeA mutation but instead produced dothistromin at levels (55.9 ± 1.9 ng/mg DW mycelium) similar to those of the DsLaeA mutant (p = 0.14). The multicopy CfLaeA strain (Cf2) showed partial complementation with a significant reduction in dothistromin level (25.8 ± 4.3 ng/mg DW mycelium) compared to the DsLaeA mutant (p = 0.005); however, this was still 3-fold higher than wild type levels. In CfLaeA-containing strains the level of expression of dothistromin genes (Figure 4) was congruent with the dothistromin levels produced. The promoter of CfLaeA was partially functional in D. septosporum as indicated by the expression of CfLaeA in the complemented strains Cf2 (33% of wild type DsLaeA expression levels) (Figure 4). The expression of dothistromin genes DsPksA DsAflR, DsAflJ and DsVer1 was reduced in the multi-copy complemented strain Cf2 compared to the knockout strain but did not revert to wild type levels. The expression levels did not vary significantly in the CfLaeA single copy complemented strain Cf3 compared to the knockout strain.

14

3.6 DsLaeA regulates the expression of other secondary metabolite genes in Dothistroma septosporum LaeA is known to regulate a diverse range of secondary metabolites in fungi (Jain and Keller, 2013). A thin layer chromatograph of organic solvent extracts from DsLaeA strains KO1 and KO2 growth cultures clearly showed the presence of an extra compound compared to the wild type (Figure 5a). As the compound was likely to be a secondary metabolite based on the extraction procedure used, secondary metabolite backbone genes that were identified previously (De Wit et al 2012; Chettri et al., 2012) were selected for expression analysis in the DsLaeA mutant by qRT-PCR. The results revealed reduced expression of DsPks2, DsNps1 and DsNps2, no change in DsNps3, but 7 and 2 fold increased expression of DsHps1 and DsHps2 respectively in the ΔDsLaeA strain compared to the wild type (Figure 5b). Expression between replicates of DsPks3 was highly variable so no reliable trends could be observed for this gene. Based on the results from the expression studies, we conducted a functional analysis of the DsHps1 gene as well as the previously characterised DsPksA dothistromin gene (Bradshaw et al., 2006) in a ∆DsLaeA background. For this we generated double knockout mutants ∆DsLaeA ∆DsHps1 and ∆DsLaeA ∆DsPksA (Supplementary Figures 9 and 10). Thin layer chromatography of the double mutants revealed no difference in metabolite profiles between the ∆DsLaeA ∆DsHps1 and ∆DsLaeA mutants. However, the ∆DsLaeA ∆DsPksA mutant showed no detectable amount of the unknown extra compound on a TLC, suggesting that DsPksA is involved in production of that metabolite (Supplementary fig 11).

Discussion

15

LaeA proteins are regulators of secondary metabolite biosynthesis and fungal development (Keller et al., 2006; Brakhage, 2013; Griffiths et al., 2015). Elucidating the role of DsLaeA in regulating the biosynthesis of dothistromin in D. septosporum was a main aim of this study. The DsLaeA gene was identified in Dothistroma septosporum using BLAST and phylogenetic approaches, and characterised using gene knockout mutants. One of the most striking phenotypes in the ∆DsLaeA strain was its enhanced ability to produce dothistromin, compared to the wild type, in two different media and under light or dark conditions. Similarly, gene expression analysis showed up regulation of dothistromin genes DsPksA, DsVer1, DsAflJ and DsAflR in the mutant. Our results contrast with the effects of Aspergillus spp. laeA mutations on AF/ST biosynthesis that led to its name (loss of aflR expression). In A. nidulans, both toxin production and aflR expression were abolished in a laeA mutant (Bok et al., 2004). Similarly, in an A. flavus laeA mutant, expression of the aflatoxin biosynthetic genes nor1 and aflJ was reduced to approximately 23-37% of wild type levels while that of the late-stage biosynthetic gene ver1 was less than 0.01% of the wild type (Chang et al., 2012). The higher level of dothistromin production and gene expression seen in D. septosporum is most likely due to increased levels of DsAflR in the DsLaeA mutant, as DsAflR is known to activate dothistromin gene expression (Chettri et al., 2013). Thus our data suggest that, in D. septosporum, DsLaeA negatively regulates dothistromin production. This could be an indirect mechanism involving chromatin level regulation, which is known to be important in secondary metabolism (Bayram and Braus, 2012). In the DsLaeA complemented strain (C1) both the expression of LaeA and levels of dothistromin indicated incomplete complementation as wild type levels were not restored. A similar phenomenon was reported for complementation of a velvet gene mutation in Trichoderma virens (Mukherjee and Kenerley, 2010). A possible explanation for lower expression in C1 could be integration of the complementation construct at a non-homologous

16

locus causing lower expression due to a position effect, as demonstrated for other genes in Aspergillus spp. (Liang et al., 1997; Chiou et al., 2002). However, in D. septosporum five additional independent single-copy transformed DsLaeA complementation strains were obtained and tested, but these also showed incomplete complementation similar to that shown for C1 (results not shown). Attempts to complement the DsLaeA mutation with the CfLaeA gene from C. fulvum only achieved partial complementation when many copies (6) of the CfLaeA gene were introduced. Although C. fulvum does not produce dothistromin it contains a complete set of dothistromin genes (de Wit et al., 2012) and the CfAflR regulatory gene partially complemented a DsAflR mutant of D. septosporum (Chettri et al., 2013). Lack of interspecific LaeA complementation was also shown for P. chrysogenum laeA in an attempt to complement F. fujikuroi laeA deletion mutants (Wiemann et al., 2010) and T. reesei lae1 in an attempt to complement an A. nidulans ΔlaeA strain (Karimi-Aghcheh et al., 2013). Together these results strongly suggest species specificity in LaeA function. In addition to the effects on dothistromin, other putative secondary metabolite backbone genes in D. septosporum showed responses to deletion of DsLaeA. Of these, two showed enhanced expression (DsHps1 and DsHps2) three reduced (DsPks2, DsNps1 and DsNps2) and two others had no effect (DsPks1 and DsNps3) compared to the wild type. These results show that DsLaeA acts as a positive as well as a negative regulator of secondary metabolism in D. septosporum. It is possible that, in D. septosporum, other LaeAlike proteins such as those included in the phylogenetic tree (Supplementary fig 1) could also regulate secondary metabolism in the absence of DsLaeA. Recently a LaeA like methyl transferase was identified in A. nidulans (Palmer et al., 2013) and C. heterostrophus (Bi et al., 2013). In these cases the ΔlaeA-like mutants showed opposite phenotypes to those exhibited by the ΔlaeA strains in both fungi with respect to secondary metabolite levels.

17

The TLC results indicated production of high levels of a previously unnoticed metabolite by ΔDsLaeA mutants. The double knockout mutant ∆DsLaeA∆DsPksA however showed no detectable amount of this unknown metabolite. This indicated that the compound could be a dothistromin variant or some derailed stable metabolite of the dothistromin pathway. Intermediates derailed into other pathways are not uncommon in fungal polyketide biosynthesis (Sorensen et al., 2003) and similar types of derailed products have been reported in aflatoxin biosynthesis (Ehrlich, 2009). Other studies have shown negative as well as positive regulation of secondary metabolism by LaeA. In Fusarium verticillioides deletion of laeA caused increased expression of 3110 out of 14,196 genes (~22 %), of which two were PKS genes; one of these showed co-regulation with flanking genes that indicated a previously uncharacterised secondary metabolite biosynthetic gene cluster (Butchko et al., 2012). In the Dothideomycete pathogens C. heterostrophus and C. fulvum, deletion of LaeA caused enhanced production of melanin (Wu et al., 2012) and cladofulvin (Griffiths et al., 2015) respectively. Fungal secondary metabolism helps determine the virulence and lifestyle of fungal plant pathogens (Scharf et al., 2014; Pusztahelyi, et al., 2015) and in many cases regulation by LaeA has been implicated. In A. flavus deletion of laeA reduced host colonisation in peanut (Amaike and Keller, 2009) and F. oxysporum and F. fujikuroi laeA mutants showed reduced virulence in tomato and rice (Lopez-Berges et al., 2013; Wiemann et al., 2010). In the Dothideomycete maize pathogen C. heterostrophus, ChLae1 regulated production of Ttoxin, a host-selective pathogenicity factor (Wu et al., 2012). In D. septosporum it was recently demonstrated that dothistromin is a virulence factor (Kabir et al., 2015a) and we showed here that DsLaeA regulates dothistromin production, thus DsLaeA indirectly affects virulence in this pathogen. It is not yet known whether end-products of other DsLaeAregulated secondary metabolite genes have a role in disease development.

18

The effect of DsLaeA mutation on light-regulated asexual development (conidiation) was similar to those of LaeA homologs in other fungi, in which the sporulation decreased (Sarikaya Bayram et al., 2010; Wu et al., 2012; Kim et al., 2013; Griffiths et al., 2015). However, the light-regulated sporulation (light repression or dark induction) was disturbed in ∆DsLaeA strains, which implies that DsLaeA is involved in a light-sensing pathway for sporulation in D. septosporum. A similar result was shown in P. chrysogenum where deletion of laeA also caused light independent reduction of sporulation compared to wild type (Hoff et al., 2010). In D. septosporum, ∆DsLaeA mutants also produced spores with abnormal morphology, similar to those reported in A. flavus and A. fumigatus (Chang et al., 2012; Bok et al., 2005). Expression of two D. septosporum putative development related genes DsNsdD and DsStuA required DsLaeA, as well as requiring DsVeA as reported previously (Chettri et al., 2012). The reduced sporulation compared to wild type as seen in the ΔDsLaeA strain could be due to an indirect effect of reduced DsStuA levels. The ∆DsLaeA mutants also exhibited a reduced rate of spore germination rate compared to that of wild type. Similar observations were reported in A. fumigatus (Soukup et al., 2012), where a microarray study revealed that aberrant germination in ∆laeA mutants was mediated by NosA, a developmental regulator. An ortholog of A. fumigatus NosA is present in D. septosporum (JGI PID 138917) and bears 55.5 % amino acid identity, but further work is required to determine if this NosAlike protein has a similar role in D. septosporum. Another phenotype observed in the ΔDsLaeA strain was decreased surface hydrophobicity compared to wild type. Reduced hydrophobicity was also seen in laeA mutants of A. flavus, P. chrysogenum, F. graminearum and F. verticillioides (Chang et al., 2012; Kosalkova et al., 2009; Jiang, et al., 2011; Kim et al., 2013). The decreased hydrophobicity was associated with decreased expression of the hydrophobin gene DsHdp1. Hydrophobins have been shown to assist in surface adhesion in C. fulvum (Lacroix et al.,

19

2008) and virulence in Magnaporthe oryzae (Kim et al., 2005), so may also have a role in virulence in D. septosporum, although no loss of virulence was seen in DsHdp1 mutants (Bradshaw et al., 2016). However other hydrophobic molecules like lipids, waxes and adhesins can provide complementary functions by contributing to fungal surface hydrophobicity (Ishiga et al., 2013; Tronchin et al., 1997). Our key findings were that DsLaeA negatively regulates the synthesis of dothistromin and expression of the DsAflR pathway specific regulatory gene during growth in vitro. In addition the ∆DsLaeA strain produced an unknown metabolite that is suppressed in the wild type strain. Hitherto the role of LaeA has been studied mostly in fungal species where the genes for secondary metabolites are clustered and generally located in sub-telomeric regions of the genome. The fragmented nature of the dothistromin cluster and presence of the DsAflR regulatory gene in a central chromosomal location could be a contributing factor that makes its response to LaeA different from that of AF/ST in Aspergillus spp.. The current study also highlights the species-specific nature of LaeA regulation, with lack of DsLaeA mutant phenotype complementation by a LaeA gene from a closely-related species, C. fulvum. Further study of global secondary metabolite regulation with fragmented gene clusters will shed more light on mechanisms of LaeA function.

Acknowledgements: Financial support from Massey University and the New Zealand Bio-Protection Research Centre is acknowledged. We thank Trevor Loo (Massey University) for his assistance with HPLC, Dr Pierre-Yves Dupont for help with bioinformatics data analysis and Mrs Carole Flyger for technical assistance. Melissa Guo and Kutay Ozturk are thanked for critical evaluation of the manuscript.

20

References: Amaike, S., Keller, N.P., 2009. Distinct roles for VeA and LaeA in development and pathogenesis of Aspergillus flavus. Eukaryotic Cell 8, 1051–1060. Amare, M.G., Keller, N.P., 2014. Molecular mechanisms of Aspergillus flavus secondary metabolism and development. Fungal Genetics and Biology 66, 11-18. Bayram, Ö. et al., 2008. VelB/VeA/LaeA complex coordinates light signal with fungal development and secondary metabolism. Science 320, 1504–1506. Bayram, O., Braus, G.H., 2012. Coordination of secondary metabolism and development in fungi: the velvet family of regulatory proteins. FEMS Microbiology Reviews 36, 1–24. Bi, Q. et al., 2013. Cochliobolus heterostrophus Llm1 - a Lae1-like methyltransferase regulates T-toxin production, virulence, and development. Fungal Genetics and Biology 51, 21–33. Bok, J. et al., 2005. LaeA, a regulator of morphogenetic fungal virulence factors. Eukaryotic Cell 4, 1574–1582. Bok, J.W., Keller, N.P., 2004. LaeA , a Regulator of secondary metabolism in Aspergillus spp.. Eukaryotic Cell 3, 527-535. Bradshaw, R.E. et al., 2006. A polyketide synthase gene required for biosynthesis of the aflatoxin-like toxin, dothistromin. Mycopathologia 161, 283–294. Bradshaw, R.E. et al., 2013. Fragmentation of an aflatoxin-like gene cluster in a forest pathogen. The New Phytologist 198, 525–535. Bradshaw, R.E. et al., 2016. Genome-wide gene expression dynamics of the fungal pathogen Dothistroma septosporum throughout its infection cycle of the gymnosperm host Pinus radiata. Molecular Plant Pathology 17, 210-224. Brakhage, A.A., 2013. Regulation of fungal secondary metabolism. Nature Reviews Microbiology 11, 21–32.

21

Butchko, R.A.E. et al., 2012. Lae1 regulates expression of multiple secondary metabolite gene clusters in Fusarium verticillioides. Fungal Genetics and Biology 49, 602–612. Calvo, A.M., Cary, J.W., 2015. Association of fungal secondary metabolism and sclerotial biology. Frontiers in Microbiology 6, 62.. Chang, P. et al., 1993. Cloning of the Aspergillus parasiticus apa-2 gene associated with the regulation of aflatoxin biosynthesis. Applied and Environmental Microbiology 59, 3273–3279. Chang, P.K. et al., 2012. Effects of laeA deletion on Aspergillus flavus conidial development and hydrophobicity may contribute to loss of aflatoxin production. Fungal Biology 116, 298–307. Chettri, P. et al., 2012. The veA gene of the pine needle pathogen Dothistroma septosporum regulates sporulation and secondary metabolism. Fungal Genetics and Biology 49, 141–151. Chettri, P. et al., 2013. Dothistromin genes at multiple separate loci are regulated by AflR. Fungal Genetics and Biology 51, 12–20. Chettri, P. et al., 2015. Regulation of the aflatoxin-like toxin dothistromin by AflJ. Fungal Biology 119, 503–508. Chiou, C.H. et al., 2002. Chromosomal location plays a role in regulation of aflatoxin gene expression in Aspergillus parasiticus. Applied and Environmental Microbiology 68, 306–315. Dagenais, T.R.T. et al., 2010. Aspergillus fumigatus LaeA-mediated phagocytosis is associated with a decreased hydrophobin layer. Infection and Immunity 78, 823–829. de Wit, P.J.G.M. et al., 2012. The genomes of the fungal plant pathogens Cladosporium fulvum and Dothistroma septosporum reveal adaptation to different hosts and lifestyles but also signatures of common ancestry. PLoS Genetics 8, e1003088.

22

Drenkhan, R., Hanso, M., 2009. Recent invasion of foliage fungi of pines (Pinus spp) to the Northern Baltics. Forestry Studies 49–64. Ehrlich, K.C., 2009. Predicted roles of the uncharacterized clustered genes in aflatoxin biosynthesis. Toxins 1, 37–58. Fernandes, M. et al., 1998. Sequence-specific binding by Aspergillus nidulans AflR, a C6 zinc cluster protein regulating mycotoxin biosynthesis. Molecular Microbiology 28, 1355–1365. Gallagher, B.R.T., Hodges, R., 1972. The chemistry of dothistromin, a difuroanthraquinone from Dothistroma pini. Australian Journal of Chemistry 25, 2399–2407. Griffiths, S. et al., 2015. Regulation of secondary metabolite production in the fungal tomato pathogen Cladosporium fulvum. Fungal Genetics and Biology 84, 52–61. Hoff, B. et al., 2010. Two components of a velvet-like complex control hyphal morphogenesis, conidiophore development, and penicillin biosynthesis in Penicillium chrysogenum. Eukaryotic Cell 9, 1236–1250. Ipcho, S.V.S. et al., 2010. The transcription factor StuA regulates central carbon metabolism, mycotoxin production, and effector gene expression in the wheat pathogen Stagonospora nodorum. Eukaryotic Cell 9, 1100–1108. Ishiga, Y. et al., 2013. Expression analysis reveals a role for hydrophobic or epicuticular wax signals in pre-penetration structure formation of Phakopsora pachyrhizi. Plant Signaling and Behavior 8, e26959. Jain, S., Keller, N., 2013. Insights to fungal biology through LaeA sleuthing. Fungal Biology Reviews 27, 51–59. Jiang, J. et al., 2011. Involvement of a velvet protein FgVeA in the regulation of asexual development, lipid and secondary metabolisms and virulence in Fusarium graminearum. PloS One 6, e28291.

23

Kabir, M.S. et al., 2015 (a). Dothistromin toxin is a virulence factor in dothistroma needle blight of pines. Plant Pathology 64, 225–234. Kabir, M.S. et al., 2015(b). The hemibiotrophic lifestyle of the fungal pine pathogen Dothistroma septosporum. Forest Pathology 45, 190–202. Karimi-Aghcheh, R. et al., 2013. Functional Analyses of Trichoderma reesei LAE1 reveal conserved and contrasting roles of this regulator. Genes, Genomes, Genetics 3, 369– 378. Keller, N. et al., 2006. LaeA, a global regulator of Aspergillus toxins. Medical Mycology 44, 83–85. Keller, N., Hohn, T., 1997. Metabolic pathway gene clusters in filamentous fungi. Fungal Genetics and Biology 21, 17–29. Kim, H.K. et al., 2013. Functional roles of FgLaeA in controlling secondary metabolism, sexual development, and virulence in Fusarium graminearum. PloS One 8, e68441. Kim, S. et al., 2005. MHP1, a Magnaporthe grisea hydrophobin gene, is required for fungal development and plant colonization. Molecular Microbiology 57, 1224–1237. Kosalková, K. et al., 2009. The global regulator LaeA controls penicillin biosynthesis, pigmentation and sporulation, but not roquefortine C synthesis in Penicillium chrysogenum. Biochimie 91, 214–225. Lacroix, H. et al., 2008. Localization of Cladosporium fulvum hydrophobins reveals a role for HCf-6 in adhesion. FEMS Microbiology Letters 286, 136–144. Liang, S.H. et al., 1997. Analysis of mechanisms regulating expression of the ver-1 gene, involved in aflatoxin biosynthesis. Applied and Environmental Microbiology 63, 1058–1065.

24

Lim, F.Y. et al., 2012. Genome-based cluster deletion reveals an endocrocin biosynthetic pathway in Aspergillus fumigatus. Applied Environmental Microbiology 78, 4117– 4125. Liu, Q. et al., 2016. Inactivation of the global regulator LaeA in Monascus ruber results in a species-dependent response in sporulation and secondary metabolism. Fungal Biology 120, 297–305. López-Berges, M.S. et al., 2013. The velvet complex governs mycotoxin production and virulence of Fusarium oxysporum on plant and mammalian hosts. Molecular Microbiology 87, 49–65. Lysøe, E. et al.,2011. The transcription factor FgStuAp influences spore development, pathogenicity, and secondary metabolism in Fusarium graminearum. Molecular PlantMicrobe Interactions 24, 54–67. Moller, E.M. et al., 1992. A simple and efficient protocol for isolation of high molecular weight DNA from filamentous fungi, fruit bodies and infected plant tissues. Nucleic Acids Research 20, 6115–6116. Mukherjee, P.K., Kenerley, C.M., 2010. Regulation of morphogenesis and biocontrol properties in Trichoderma virens by a VELVET protein, Vel1. Applied and Environmental Microbiology 76, 2345–52. Mullett, M.S., Fraser, S., 2016. Infection of Cedrus species by Dothistroma septosporum. Forest Pathology. doi 10.1111/efp.12214 Namiki, F. et al., 2001. Mutation of an arginine biosynthesis gene causes reduced pathogenicity in Fusarium oxysporum f. sp. melonis. Molecular Plant-Microbe Interactions 14, 580–584.

25

Palmer, J.M. et al., 2013. Secondary metabolism and development is mediated by LlmF control of VeA subcellular localization in Aspergillus nidulans. PLoS Genetics 9, e1003193. Patananan, A.N. et al., 2013. A novel automethylation reaction in the Aspergillus nidulans LaeA protein generates S-methylmethionine. The Journal of Biological Chemistry 288, 14032–14045. Paz, Z. et al., 2011. One step construction of Agrobacterium-recombination-ready-plasmids (OSCAR), an efficient and robust tool for ATMT based gene deletion construction in fungi. Fungal Genetics and Biology 48, 677–684. Pusztahelyi, T. et al., 2015. Secondary metabolites in fungus-plant interactions. Frontiers in Plant Science 6, 573. Sarikaya Bayram, O., et al., 2010. LaeA control of velvet family regulatory proteins for lightdependent development and fungal cell-type specificity. PLoS Genetics 6, e1001226. Scharf, D.H. et al., 2014. Human and plant fungal pathogens: The role of secondary metabolites. PLoS Pathogens 10, e1003859. Schoch, C.L. et al., 2009. The ascomycota tree of life: A phylum-wide phylogeny clarifies the origin and evolution of fundamental reproductive and ecological traits. Systematic Biology 58, 224–239. Schumacher, J. et al., 2015. The VELVET complex in the gray mold fungus Botrytis cinerea: Impact of BcLAE1 on differentiation, secondary metabolism and virulence Molecular Plant-Microbe Interactions 28, 659–674. Schwelm, A. et al., 2008. Early expression of aflatoxin-like dothistromin genes in the forest pathogen Dothistroma septosporum. Mycological Research 112, 138–146. Seiboth, B. et al., 2012. The putative protein methyltransferase LAE1 controls cellulase gene expression in Trichoderma reesei. Molecular Microbiology 84, 1150–1164.

26

Sorensen, J.L. et al., 2003. Transformations of cyclic nonaketides by Aspergillus terreus mutants blocked for lovastatin biosynthesis at the lovA and lovC genes. Organic & Biomolecular Chemistry 1, 50–59. Soukup, A. et al., 2012. NosA, a transcription factor important in Aspergillus fumigatus stress and developmental response, rescues the germination defect of a laeA deletion. Fungal Genetics and Biology 49, 857–865. Strauss, J., Reyes-Dominguez, Y., 2011. Regulation of secondary metabolism by chromatin structure and epigenetic codes. Fungal Genetics and Biology 48, 62–69. Takao, K. et al., 2016. The global regulator LaeA controls biosynthesis of host-specific toxins, pathogenicity and development of Alternaria alternata pathotypes. Journal of General Plant Pathology 82, 121–131. Tamura, K. et al., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Molecular Biology and Evolution 28, 2731–2739. Trienens, M., et al., 2010. Fruit, flies and filamentous fungi - experimental analysis of animal-microbe competition using Drosophila melanogaster and Aspergillus mould as a model system. Oikos 119, 1765–1775. Tronchin, G. et al., 1997. Expression and identification of a laminin-binding protein in Aspergillus fumigatus conidia. Infection and Immunity 65, 9–15. Watt, M.S. et al., 2009. The hosts and potential geographic range of Dothistroma needle blight. Forest Ecology and Management 257, 1505–1519. Wiemann, P. et al., 2010. FfVel1 and FfLae1, components of a velvet-like complex in Fusarium fujikuroi, affect differentiation, secondary metabolism and virulence. Molecular Microbiology 77, 972–994.

27

Woods, A. et al., 2005. Is an unprecedented Dothistroma needle blight epidemic related to climate change? BioScience 55, 761. Woods, A., 2011. Is the health of British Columbia’s forests being influenced by climate change? If so, was this predictable? Canadian Journal of Plant Pathology 33, 117–126. Wu, D. et al., 2012. ChLae1 and ChVel1 regulate T-toxin production, virulence, oxidative stress response, and development of the maize pathogen Cochliobolus heterostrophus. PLoS Pathogens 8, e1002542.

28

Figure: 1 Phylogenetic tree of D. septosporum LaeA protein and its orthologs in other species. Phylogenetic tree constructed by neighbor-joining with 1000 bootstrap replicates using MEGA 5.0 (Tamura et al., 2011). Numbers on the branches represent the percentage of replicates supporting each branch. Labels on the right indicate the percentage (%) aa identity with D. septosporum LaeA (JGI: protein ID 148869) and their corresponding NCBI or JGI protein ID (PID). The shaded box indicates fungi belonging to order Capnodiales, and the Dothideomycetes class is marked with a border. The scale bar represents approximately 20% sequence divergence.

Figure 2: Sporulation and growth characteristics of ΔDsLaeA mutants (A) Sporulation and (B) growth rate of wild type (WT), ΔDsLaeA mutants (KO1, KO2) and DsLaeA complemented (C1) strains in D. septosporum in light (grey bar) and dark (black bar) conditions (mean ± SD, n=3). Values significantly different from the WT (p ≤ 0.05) are marked with an asterisk. There was no significant difference in sporulation in between light and dark conditions in the ΔDsLaeA mutant strains.

Figure 3: Hydrophobicity and hydrophobin gene expresson of DsLaeA mutants (A) 30 μl aliquots of bromophenol blue solution were spotted on colonies, which were photographed after 20 min. Spherical droplets were maintained on the surface of wild type (WT) and complemented strains (C1) but the droplet was readily absorbed by the ΔDsLaeA strains (KO1, KO2) indicating loss of hydrophobicity in the latter. (B) Expression of the DsHdpI hydrophobin gene in ΔDsLaeA mutants.. Normalized gene expression ratios for DsLaeA knockout (KO1 and KO2) and complemented (C1) strains,

29

relative to wild type (WT); mean ± SD, n=3. Significant differences from WT (p ≤ 0.05) are shown by an asterisk.

Figure 4: Expression of dothistromin genes in DsLaeA mutant strains. Gene expression was evaluated by quantitative real-time PCR and is shown as normalized gene expression ratios (mean ± SD; n=3 relative to the β-tubulin gene) for DsLaeA, dothistromin biosynthetic (DsPksA, DsVer1, DsVbsA)) and regulatory (DsAflR, DsAflJ) genes in WT, ΔDsLaeA (KO1 and KO2) mutants and DsLaeA complemented (C1) or CfLaeAcontaining CF2 (multicopy) and CF3 (single copy) strains grown in PMMG media under light conditions. Significant differences from WT (p ≤ 0.05) are shown by an asterisk.

Figure 5: Secondary metabolite production and gene expression in DsLaeA mutants (A) TLC of extracts from ΔDsLaeA strains (KO1 and KO2), DsLaeA complemented (C1) or wild-type (WT) strains. The extra spot in ΔDsLaeA KO1 and KO2 has been indicated with a white arrow. (B) Gene expression compared to WT was evaluated by quantitative real-time PCR and is shown as log2-fold differences in expression (mean ± SD, n = 3). Expression ratios for polyketide synthase DsPks1, DsPks2 and DsPks3 non ribosomal peptide synthase DsNps1, DsNps2 and DsNps3 and PKS-NRPS hybrids (DsHps1and DsHps2) (in ΔDsLaeA mutants (KO1 and KO2) and complemented (C1) strains, normalised to wild type (WT) levels (WT = 1); Significant differences from WT (p ≤ 0.05) are shown by an asterisk.

30

31

32

33

Table 1: Dothistromin production in PMMG medium Straina

Dothistromin b (ng/mg of DW mycelium) Light Dark WT 9.0 ± 1.2 7.7 ± 1.4 ∆DsLaeA KO1 71.6 ± 11.2* 64.6 ± 2.8* ∆DsLaeA KO2 46.3 ± 9.8* 59.8 ± 3.4* DsLaeA C1 17.8 ± 2.1* 18.8 ± 2.7* a WT, wild-type; ΔDsLaeA KO1, KO2, knockout mutants; DsLaeA C1, complementation strain. b

Dothistromin (mean ± SD); n = 3.

*

Significantly different to WT (P < 0.05).

34

Highlights

LaeA mutants of the pine pathogen Dothistroma septosporum were produced LaeA negatively regulates production of the virulence factor dothistromin. Expression of the dothistromin pathway regulator AflR was increased in LaeA mutants. LaeA mutants showed high levels of an unknown metabolite not seen in the wild type. Incomplete complementation by Cladosporium fulvum LaeA suggests species specificity.

35