Expression Analysis of PKS13, FG08079.1 and PKS10 Genes in Fusarium graminearum and Fusarium culmorum

1. Background
Mycotoxins are fungal secondary metabolism’s products with low-molecular weight. Fungi producing mycotoxins are well-defined at genera level (1). Fusarium is well known genus infecting considerable plant species especially small-grain cereals (2, 3). Fusarium graminearum and F. culmorum are common mycotoxin producers. Their mycotoxins can be divided into two classes as: (i) Major consisting of trichothecenes, fumonisins, zearalenones, and (ii) minor including beauvercin and enniatins, equisetin, fusarins, butenolide (4).
Both F. graminearum and F. culmorum have the ability to produce zearalenone (ZEN), butenolide (BUT) and fusarin C (FUS C) mycotoxins (4). ZEN is a nonsteroidal estrogenic endotoxin synthesized from gene cluster that of 25 kb. Cluster consists of nine genes. PKS4 and PKS13 encoding two different polyketide synthases (PKS) are essential for ZEN production. Disruption of these genes resulted in loss of ZEN production (5, 6). BUT is a water-soluble endotoxin and produces by a gene cluster with eight genes on a contig, namely 1.324. Gene disruption and add-back approaches indicated that FG08079.1 gene in the cluster is essential for BUT production. According to sequence similarity, FG08079.1 appears to be a cytochrome P450 monooxygenase, while the function of other 7 genes remained elusive (7). FUS C is one of the fusarin stereoisomers. FUS C biosynthetic pathway is controlled by cluster of nine genes reported in F. fujikuroi. The fus1 is essential in production and its product is a PKS. It was shown that PKS10 in F. graminearum was homologous to fus1 of F. fujikuroi (8, 9).
Detection of mycotoxins within crops and their products is very important for human, animal and plant health. Several conventional methods are effectively used for determination of mycotoxins. High-performance liquid chromatography/mass spectrometry (HPLC/MS), gas chromatography-mass spectrometry (GC-MS), liquid chromatography with tandem mass spectrometry (LC-MS/MS) and enzyme linked immunosorbent assay (ELISA) are methods regularly being used for mycotoxin determination and sometimes quantification. However, these techniques are labour intensive, time consuming and expensive. Instead of using above direct methods in identification and determination of the presence of mycotoxins, one may evaluate the presence of genes involved in biosynthesis of such substances via PCR and further investigate their expression via Real-Time PCR (10).

2. Objectives
    Expression of PKS13, FG08079.1 and PKS10, the key players of zearalenon, butenolide and fusarin C (FUS C) production, were analysed. It was aimed to determine mycotoxin production potential of F. graminearum and F. culmorum isolates and relative quantification of expression of the genes.

3. Materials and Methods

3.1. Fungal Isolation and Culture
    F5, F8, F9, 1F, 2F, F2, F12, F15, F16 and F19 isolates was obtained from single-spore cultures (11). Isolates (Table 1) were kindly provided from Prof. Berna Tunali, at Samsun Ondokuz Mayis University, Plant Protection Department of Agriculture Faculty. Isolates were grown at 25°C on potato dextrose agar (PDA).
3.2. Polymerase Chain Reaction
    Genomic DNA was extracted from 100 mg of 7-day-old fresh mycelium using CTAB method with minor modifications (12). Total RNA was extracted using Roche TriPure Reagent (Roche, Switzerland) that used for cDNAs synthesis using Roche cDNA synthesis kit (Roche, Switzerland) following to manufacturer’s protocol.
    UBC85F/R (5¢-ctccggatatgttgcgtcaa-3¢/5¢-ggtaggtatccgacatggcaa-3¢) and OPT18F/R (5¢-atggtgaactcgtcgtggc-3¢/5¢-cccttcttacgccaatctcg-3¢) primer sets were used in amplification of SCAR85 marker specific for F. graminearum and SCAR2-35 for F. culmorum, respectively (13). PCR cycling and conditions were maintained as described by Schilling et al. (13) in a thermocycler (Bio-Rad-T100, USA).
   ZENF/R (5¢-ggtagcgataacgtggagga-3¢/5¢-gactcaaaggtgctcggttc-3¢), BUTF/R (5¢-tcattgaccgagctttctga-3¢/5¢-ctccatcgtgtcgtctctca-3¢) and FUSCF/R (5¢-ttccccctgtacgattcaac-3¢/5¢-tcattgaccgagctttctga-3¢) primers were used for amplification of PKS13 (Acc. no: AY495638.1), FG08079.1 (Acc. no: FG08079.1) and PKS10 (Acc. no: AY495635.1) genes, respectively. PCR mixing were carried out as previously described by Yörük and Albayrak (14). The thermal cycling conditions were applied as 40 cycles of 94°C for 1 min, 55-57°C for 1 min and 72°C for 2 min. Prior to cycling, PCR tubes were incubated at 95°C for 5 min, and the cyclings were finished with at 72°C for 10 min Standard Taq DNA polymerase enzyme kit (Promega, USA) was used in each PCR sets.

3.3. Real-Time PCR
    Real-time PCR assays were carried out in LightCycler® 480 II (Roche, Switzerland) system using SYBR Green qPCR kit (Thermo, USA). The qPCR experiments were conducted in a reaction volume of 16 mL containing 1× SYBR Green mix, 4 pmol of each primer, cDNA amount corresponding to 500 ng RNA. b-tubulin gene was amplified with TUBF/R primers (5¢-gaagccattgatgttgttcgt-3¢/5¢-tccgaccatgaagaagtgaag-3¢) as internal control. Cycling conditions were followed as; 95°C for 5 min, 40 cycles of 95°C for 10 s, 55°C for 10 s, 72°C for 10 s, and final extension at 72°C for 10 min. Also, melting curve conditions were carried out to analyse the accuracy of real time PCR. Each experiment sets were replicated at least thrice. Standard curves were formed by cDNA series of 1:1/1:4/1:16/1:64 for each target genes and also for b-tubulin. According to the software, crossing point (SCp) values and target/reference ratios (SDDCT) were calculated.

4. Results
    According to species specific DNA marker amplifications, five isolates (F1, F2, F5, F8 and F9) produced a common band of 332 bp with UBC85 primer pair (Figure 1 A). They were confirmed as F. graminearum. Gene fragment of 472 bp in length were produced with OPT18 primers in remaining isolates (Figure 1 B). It was revealed that they (F2, F12, F15, F16 and F19) belonged to F. culmorum.
    PKS13 was amplified in all isolates as a band of 1213 bp long (Figure 1 C) except for F. graminearum 2F (850 bp). They yielded a common fragment of 1551 bp belonging to FG08079.1 gene region (Figure 1 D). Amplification products of PKS10 of 1296 bp in length were obtained from all isolates (Figure 1 E). It was demonstrated that three gene sequences were carried by F. graminearum and F. culmorum isolates.
    Different expression levels of PKS13 and FG08079.1 were detected in all isolates whereas no signal peaks were recorded for PKS10 gene (Table 1). Also, b-tubulin expression levels were calculated for all isolates (Table 1). SCp values of FG08079.1 and PKS13 were found as 29.86 ± 0.22-32.18 ± 1.14 and 29.80 ± 0.07-32.07 ± 0.009, respectively. Values were between 21.78 ± 0.20-23.55 ± 0.38 for b-tubulin. After normalization, the highest gene expressions were recorded in 2F isolate of F. graminearum for FG08079.1 and F16 isolate of F. culmorum for PKS13. While SDDCT values for PKS13 were ranged from 1.79E-03 to 3.97E-03, the values calculated between 0.25E-03 and 6.02E-03 for FG08079.1.

5. Discussion
    Detection of phytopathogenic fungal species and their mycotoxins is crucial in development of plant disease control strategies. Conventional and molecular marker techniques are currently used in species-specific diagnostics and determination of mycotoxin types. Fusarium isolates (10) was identified by SCAR marker amplification and were determined as F. graminearum and F. culmorum.
    Mycotoxin types and their quantity are important in the controlling of Fusarium diseases. To get information about genes associated with mycotoxin production is also essential. Since genes involved in biosynthesis of ZEN, BUT and FUS C are known (4). It is possible to predict possibility of the production of these toxins via expression analysis of the genes. PKS13, FG08079.1 and PKS10 genes were investigated at genomic and transcriptomic levels. Routine tests for trichothecene and zearealenone detection in cereals have been used. There is no any kit for butenolide and FUS C detection.
    ZEN production can increase by high temperature as well as plant-pathogen interaction. But, this effect may vary depending on strain. The mycotoxin generally accumulates during growth phase and stays stabile on crops and their products. Our results showed that isolates included in the study produced ZEN mycotoxin under optimal culture conditions without any plant infection and/or stress factor. Moreover, F. graminearum 2F isolate contained PKS13 gene in a rather different amplicon size (850 bp). This may suggest the presence of other variants of this gene as seen for trichothecene (14). Slight differentiation among gene expression profiles of isolates may support this argument. Nearly the same expression patterns were obtained among isolates even when the differences were statistically significant (p < 0.05)
    The FG08079.1 is expressed during cell growth phase and in early stage of plant infection (7, 15). Here, FG08079.1 was determined in all isolates under optimal culture conditions. It was noted an almost 24-fold increase in BUT production in 1F as opposed to 2F, isolates of F. graminearum. However, in isolates of F. culmorum, F2 was greater than F12-by only 2 fold. Expression levels of both PKS13 and FG08079.1 could provide information about aggressiveness of Fusarium isolates.
    Despite the fact that PKS10 present in isolates of both fungal species, no expression of the gene was detected under optimal culture conditions. It was reported that FUS C was actively synthesized during growing mycelia, under high nitrogen concentrations or acidic conditions (8, 16). Therefore, the provided condition most likely was not suitable to induce the expression of PKS10. Investigation other genes involved in fusarin gene cluster could be useful in FUS C production analysis. The polyketide synthase, encoded by fusA, contain ten domains and four are characteristics of NRPS (nonribosomal peptide synthase) domain (16). To select the NPRS instead of PKS10 could be useful for detection of fusarin production.
    Relative quantification values of PKS13 and FG08079.1 were under “1”, which was accepted as standard gene expression value for positive calibrator. These expression values were about ×10-3, smaller than the reference gene. These findings are indicative of low abundance of PKS13 and FG08079.1 associated with production of related mycotoxins. Moreover, these results obtained from optimal culture conditions would indirectly indicate mycotoxins levels produced by pathogens in non-inducing condition. Differentiation of isolates as mycotoxin producer or non-producer is carried out in a short time by using qPCR analysis. Our data showed that molecular approaches used in determination of mycotoxin type were efficient as much as conventional approaches. At the same time, it was seen that mycotoxin amount could be sensitively determined and compared by using real-time PCR.
    Occurrence of various endotoxins (aflatoxin, T-2 toxin, DON, ZEN, and fumonisin) except BUT and FUS C have been reported by using conventional methods like HPLC and ELISA in Turkey (17, 18). Chemotyping studies of F. graminearum and F. culmorum isolates are focused on class B trichothecene detection (14, 19). But, there is no studies for ZEN, BUT and FUS C chemotyping in Turkey. Number of chemotyping investigation on these three mycotoxins has been also restricted worldwide when compared to class B chemotyping. To our knowledge, this is the first report to determine the potential of Fusarium isolates to produce ZEN, BUT and FUS C. Finding obtained from similar studies could contribute to evaluate of infection grade and to improve of disease control strategies.

Acknowledgements
The project was partially supported by Research Found of Istanbul University by project number 23655