Cell wall traits as potential resources to improve resistance of durum wheat against Fusarium graminearum

Growing condition of wheat and pathogenicity tests

Wheat seeds were surface-sterilized in Sodium hypochlorite and transferred on petri dishes containing 3MM paper soaked with water. Plates were stored at 4°C in the dark for 24-48 h and transferred in a growth chamber at 23°C in the dark for 15 days. Plants grown in a controlled environmental chamber maintained at 22°C, 70% humidity with a 16 hours photoperiod (300μE m-2 s-1).

Pathogenicity tests were conducted using the Fusarium resistant common wheat line, accession n. 02-5B-318 (a breeding line derived from cv. Sumai3, kindly provided by dott. Stefano Ravaglia, S.I.S., Bologna, Italy) and on the susceptible durum wheat cv. Saragolla. Uniform inoculum pressure was applied during flowering by using the Fusarium graminearum PH 1 isolate (kindly provided by prof. Quirico Migheli, University of Study of Sassari, Italy). Plants were artificially inoculated by spraying on each plants 100 mL of a suspension containing a mixture of conidia of F. graminearum (about 1.0 × 105 conidia per mL). Fusarium strain was grown for one week on PDA (Potato Dextrose Agar) and conidia were isolated by growing pieces of mycelium in shaking cultures in 2 L PIREX flasks containing 1 L sterile CMC (Carboxyl-methyl-cellulose) medium (15gr CMC, 1gr NH₄NO₃; 1gr KH₂PO₄; 0.5 gr MgSO₄*7H₂0; 1gr yeast extract; 50 ug/mL chloramphenicol). After 5-day incubation in the dark at 25°C shaking at 150 rpm, flasks content was filtered through two layers of cheesecloth by centrifugation at 3,000 rpm for 10 min; pellet was re-suspended in sterile water and centrifuged again. Filtered conidia were finally re-suspended in 10 mL of sterile water. The concentration of the inoculum was measured with a Burker camera (HBG Henneberg-Sander GmbH, Lutzellinden, Germany) using a light-microscope.

Twenty plants for 02-5B-318 and Saragolla line were artificially spray-inoculated during anthesis with a 10⁶/mL distilled-water macroconidia suspension, for each plant 5 spikes were chosen for a total of 100 spikes per lines. Fusarium head blight (FHB) incidence and severity were recorded five, ten and twenty days after inoculation on both infected and mock-inoculated (controls) wheat plants: FHB severity was averaged as the percentage of infected spikelets per plant, while FHB incidence was averaged as the number of infected spikes per plant; a mean value of at least 20 plants per genotype was assessed. Infection experiments were statistically evaluated by performing analysis of variance followed by the Student’s t test.

Alcohol-insoluble solids (AIS) extraction

Wheat spikes were collected at anthesis stage and infected spikes were collected after 24, 48 and 72 hours post Fusarium inoculation. Tissues excised from the central part of each spike, including rachis and spikelets were ground to a fine powder with a mortar and pestle in presence of liquid nitrogen. Milled tissue (200 mg) was washed twice in a pre-warmed (70°C) 70% ethanol, vortexed, and pelleted by centrifugation at 25,000 g for 10 min. The pellet was suspended with a chloroform:methanol mixture (1:1, v/v) and shaked for 30 min at room temperature. Samples were pelleted by centrifugation at 25,000 g for 10 min. Pellets were re-suspended in 1 ml 80% acetone and spin at 25,000 g for 5 min. Supernatants were discarded and pellets were dried at room temperature over-night.. Starch was removed by treating the AIS with the porcine Type I-A α-amylase (100 U g-1 AIS; product number A4268; Sigma-Aldrich) in a 100 mM potassium phosphate buffer pH 7.5 mM NaCl and 0.02% (w/v) NaN3 for 24 hours at 37°C. The suspension was centrifuged at 25,000 × g for 20 minutes, and pellet was then washed with distilled water and 80% acetone.

Lignin content and monolignol composition

Acetyl bromide lignin in de-strached AIS from the spikes of both wheat varieties was determined according to [77] with some modifications. Briefly, 3 mg of AIS were placed in glass vials, and then 200 μl 25% acetyl bromide in acetic acid and 600 μl of acetic acid (glacial) were added. Mixtures were incubated at 50°C for 2 h, with occasional shaking. 15 μl of reaction mixture after cooling was transferred to 96-well plate (UV transparent), and 15 μl 0.3 M NaOH, 5 μl 0.5 M hydroxylamine hydrochloride and 65 μl acetic acid (glacial) were added. After shaking, optical density at 280 nm against blanks (all reagents without AIS samples) was measured using plate reader. Lignin concentration was determined using the following equation: % lignin content = (absorbance × 100)/SAC × AIS concentration (g⁻¹) where SAC is the specific absorption coefficient of lignin [78]. Specific monolignol composition was determined using Pyrolysis-GC-MS. De-starched AIS (3 mg) were single-shot pyrolized at 500°C and the volatile compounds were separated on HP-5 MS column (30 m × 0.25 mm, Agilent Technologies Inc, USA) using GC system (6890 N GC-system interfaced to 5975B inert MSD, Agilent Tech., USA). Oven temperature was initially set at 50°C and ramped to 280°C over a period of 53 min. Helium was the carrier gas for the volatile compounds and the split ratio was set at 50:1. Peak identification was performed by comparison of sample spectra with those published by [79]. The monolignol composition was calculated as %, combining the peak areas of similar type of lignin.

Determination of the degree of methylesterification

De-starched AIS (4 mg) were saponified by suspending them in 60 μl H₂O up and 20 μl of 1 M NaOH. The solution was incubated at room temperature for 1 h and afterward neutralized with HCl. After centrifugation at 25,000 × g, aliquots of the supernatant (50 μl) of 02-5B-318_R and Saragolla_S were loaded in microtiter plates (96-well cod.9018 from Costar, Cambridge, MA, U.S.A.). Alcohol oxidase (50 μl) was added to each well (0.03 units in 0.1 M sodium phosphate, pH 7.5) (Sigma, St. Louis), and this mixture was incubated at room temperature for 15 min on a shaker. Thereafter, 100 μl of a mixture containing 0.02 M 2,4-pentanedione in 2 M ammonium acetate and 0.05 M acetic acid was added to each well. After 10 min of incubation at 68°C, samples were cooled on ice and absorbance was measured at 412 nm in a microplate reader (ETI-System reader; Sorin Biomedica Cardio S.p.A., Saluggia, Italy. The amount of methanol was estimated as described [80]. For uronic acid quantification, 4 mg saponified AIS samples were incubated in 200 μl of 2 M Trifluoracetic acid (TFA) at 121°C. After 1.5 hours, 200 μl of isopropanol was added and the mixtures evaporated at 40°C with a stream of N₂ gas. This step was repeated twice and samples were dried at room temperature overnight. The TFA hydrolyzed monosaccharides were suspended in 200 μl of water and the Uronic acid content in the supernatant was quantified colorimetrically using the automated sulfamate/m-hydroxy diphenyl assay [81] and galacturonic acid (Fluka 48280) as standard. The degree of methylesterification was expressed as methanol to uronic acid molar ratio (%).

Cell wall fractionation and monosaccharides composition

To isolate fractions enriched in various cell wall components, AIS were subjected to sequential extraction buffers (at final concentration of 30 mg/ml) in constant mixing for 24 hours at room temperature. The following order was followed: 50 mM ammonium oxalate (Chelating Agent Soluble Solid, ChASS) pH 5.2 with 0.02% sodium azide; 1 M KOH, 1% (w/v) of sodium borohydride with 0.02% sodium azide and 4 M KOH with 1% (w/v) of sodium borohydride with 0.02% sodium azide. The 1MKOH and 4 M KOH fractions were neutralized using glacial acetic acid. All of the extracts were dialyzed against four changes of 4 L of deionized water and then lyophilized. For each genotype six independent replicates were analyzed. The monosaccharide composition of destarched AIS, the ChASS, 1 M KOH, 4 M KOH fractions and of residue, all hydrolysed with TFA was determined by HPAEC-PAD using a PA20 column (Dionex, CA, USA). Peaks were identified and quantified by comparison to a standard mixture of rhamnose (Rha), arabinose (Ara), fucose (Fuc), galactose (Gal), glucose (Glc), xylose (Xyl), mannose (Man), galacturonic Acid (GalUA),and glucuronic acid (GlcUA) (Sigma-Aldrich).

The crystalline cellulose was determined as previously described [82]. The cellulose derived glucose content in destarched AIS was determined by an anthrone colorimetric assay [83] with glucose (Sigma G8270) as a standard.

Immunodot assay

For each experiment, ChASS and KOH 1 M fractions were applied as 1 μL aliquots to nitrocellulose membrane (0.45 μm pore size; Bio-Rad, Hercules, CA, USA) in a threefold dilution series. Arrays were incubated for 1 hour in 5% (w/v) milk protein (MP; Bio-Rad) in PBS pH 7.8 (MP-PBS), and probed for 1.5 hours with primary LM7 and LM11 monoclonal antibodies (purchased from PlantProbes, Paul Knox Cell Wall Lab, University of Leeds,Leeds, UK) diluted 1:20 in 3% MP-PBS. After extensive washes in PBS, arrays were incubated with anti-rat conjugated to horseradish peroxidase (A7058; Sigma-Aldrich) diluted 1:1000 in MP-PBS buffer. After washing in PBS, LM11 arrays was developed using 4-chloro-1-naphthol [84] and, due to a weak signal, LM7 was developed using ECL detection reagent (Amersham).

Bioinformatic analysis

In order to identify homologous proteins to wheat methylesterase enzyme, a bioinformatic analysis was carried out on grass species (Sorghum bicolor, Zea mays, Setaria italica, Panicum virgatum, Oryza sativa and Brachypodium distachyon) annotated in Phytozome v.9.1 database (http://www.phytozome.net). Brachypodium BdPME1 complete genomic sequence was used as the initial query in a BLAST-search against wheat EST (Expressed Sequence Tags) database at NCBI (http://blast.ncbi.nlm.nih.gov), with the aim to retrieve sequences with a high similarity score (>80%). Each suitable EST was finally searched for similarity in the Chinese Spring database at Cereal DB (http://www.cerealsdb.uk.net/search_reads.htm), to extract 454 reads and obtain larger consensus contigs of the hexaploid reference cultivar using an e-value cut-off of e⁻⁵.

Isolation and characterization of WheatPME1 sequence in wheat lines

WheatPME1 gene isolation was conducted in the 02-5B-318 accession of T. aestivum and in the durum wheat cv. Saragolla, respectively FHB-resistant and susceptible. Genomic DNA was isolated from the two wheat lines according to the extraction protocol by [85] starting from 0.1 gr of fresh leaves, then checked for quality and concentration at a Nanodrop device (Thermo Scientific, Walthman, MA, USA). Purity of extracted DNA was assessed by measuring 260 nm/280 nm ratio, with a value of approximately 1.8-2 indicating a good quality.

Genomic DNA was PCR-amplified with several primer pairs opportunely designed by OligoExplorer software on Brachypodium genomic sequence, Chinese Spring ESTs and consensus contigs, in order to cover the entire gene sequence. All the amplification reactions were initially carried out in a gradient of annealing temperature in order to check for primer specificity and identify the optimal annealing conditions for each primer combination. PCR reactions were conducted in a total volume of 25 μl containing 100 ng of template gDNA, 250 nmol/L of each primer, 1X reaction Buffer (10 mmol/L Tris–HCl, pH 8.3; 10 mmol/L KCl), 200 μmol/L of each dNTP, 2.5 mmol/L of MgCl₂, and 1 unit of Taq DNA polimerase (EuroTaq, Euroclone®). Amplifications were run in a MyCycler™ Personal Thermal Cycler (Bio-Rad®) according to the following protocol: 5 min at 95°C, followed by 32 cycles of: 1 min at 95°C, 1 min at the given annealing temperature, and 2 min at 72°C, followed by a final extension step of 15 min at 72°C. Finally, PCR products were checked for the expected molecular size by visualization on 1.5-2% agarose gel stained with Gel-Red® dying solution (Biotium, Inc., Hayward, CA).

For the chromosomal localization of WheatPME1 genes, nulli-tetrasomic lines (NTs) of Triticum aestivum cv. Chinese Spring [86,87] were used to physically localize PME markers to chromosomes. Chinese Spring di-telosomic lines [88] were used for the assignment of markers to each chromosomal arm. Physical location on chromosome bins of each PCR fragment was obtained using a set of common wheat deletions lines dividing genome chromosomes into bins (kindly provided by B. S. Gill, USDA-ARS, Kansas State University) [89]. Single-band PCR products were directly purified from a volume of about 100 μl using the EuroGold Cycle Pure Kit (Euroclone®) following the manufacturer instructions, with the only exception of using sterile deionized water rather than the supplied elution buffer, to increase the efficiency of following sequencing reactions. Purified DNA fragments were checked on 1.5-2% agarose gel stained with Gel-Red® dye solution, then evaluated for concentration by detecting absorbance at a 260 nm wave length at a Nano Drop device (Thermo Scientific®). Sequencing analyses were performed for each fragment in both strands by BMR Genomics S.r.l (Padova). Sequence assembly was obtained with Codone Code Aligner and Geneious softwares. Multi-alignments of gene sequences between 02-5B-318 and Saragolla were carried out by ClustalW (http://www.ebi.ac.uk) and BLAST (http://blast.ncbi.nlm.nih.gov). Gene structure prediction was performed by the FGENESH on-line tool (http://linux1.softberry.com/cgi-bin/programs/gfind/bestorf.pl).

Gene expression analysis

Total RNA was isolated from spikes of infected and mock-inoculated (control) plants of both resistant 02-5B-318 and susceptible Saragolla at 24, 48 and 72 hours post inoculation. For each sample three biological replicates were collected from different plants. Tissues were harvested in each phase, immediately frozen in liquid nitrogen and stored at −80°C until RNA extraction. Total RNA was extracted using the RNeasy Plant Mini Kit (Qiagen®) and checked on 1.5% denaturing agarose gel; amount and purity were determined with a Nano-Drop spectrophotometer. All RNA samples were led to the same concentration (1 μg/μl) and reverse-transcribed into double stranded cDNA by using the Quanti-Tect Reverse Transcription Kit (Qiagen®) following the manufacturer instructions, after a prior treatment with a DNA Wipeout Buffer for the removal of gDNA contamination.

Primer pairs were designed by using OligoExplorer software on a conserved pme nucleotide region between the three wheat genomes, in order to determine the total pectin methyl-esterase gene expression in the two wheat lines. As shorter amplicons work more efficiently, primers were designed to amplify small DNA fragments in the range of 50–200 bp. Actin, CDC (Cell Division Control), ADP-RF (ADP-Ribosilation Factor) and RLI (RNase L Inhibitor-like protein) genes were used as internal references to normalize PME expression data. Specific primers for Fusarium β-tubulin 2 (βTUB2) gene were used to assay fungal infection in both inoculated and non-inoculated wheat samples (Additional file 4: Table S1).

In order to identify the best temperature to ensure primer specificity, standard PCR on cDNA were performed with a gradient of annealing temperatures (ranging between 55°C and 65°C) for both target and reference primer pairs, by using high fidelity MyTaq DNA polymerase (BioLine). Amplicon specificity was confirmed for each primer pair by checking the presence of single PCR products of expected molecular size on 2% (w/v) agarose gel stained with Gel Red® dying solution, and by direct sequencing of the amplified fragments (BMR Genomics, Padova, Italy).

Primer concentration was optimized for each gene in preliminary Real-Time amplification experiments by running reactions with different combinations of forward and reverse primers in the final mix (100, 300, 500 and 900 nM), then choosing those giving the highest endpoint fluorescence and a low Cq value. Primer specificity was also checked by performing melting curves of PCR products following Real Time amplifications.

qRT-PCR reactions were performed using EvaGreen® chemistry in the CFX96™ Real-time PCR System (Bio-Rad) following these conditions: 95°C for 3 min, followed by 40 cycles of: 95°C for 10 sec and 60°C for 30 sec. In each qPCR experiment 1 μl of a 1:10 dilution of cDNA was used in a final volume of 10 μl containing 5 μl of SsoFast EvaGreen® SuperMix 10X (Bio-Rad) and a primer concentration of 500 nM for WheatPME1, and 100 nM for Actin, CDC, ADP-RF and RLI. Three independent amplification reactions (technical replicates) were carried out for each biological replicate.

PCR reaction efficiency was calculated for both target and reference genes by generating six-point standard curves of three-fold serial dilutions of cDNA. Standards were run in the same amplification plate of the unknown samples. All experiments were performed in Hard-Shell 96-well skirted PCR plates (HSP9601) with Microseal® ‘B’ Adhesive Seals (MSB-1001) from Bio-Rad®.

Data analyses were performed with the CFX Manager™ 3.1 software, using the Normalized Expression mode (ΔΔC_q) which calculated the relative quantity of target (WheatPME1) normalized to the relative quantity of internal references (geometric mean of multiple reference genes). For both target and reference genes, relative expression was calculated as fold-change respect to the mock-inoculated controls at each harvesting stage, and determining the standard deviation (SD) for the relative quantity. All the results were analyzed by ANOVA.

Availability of supporting data

All the supporting data are included as additional files in this manuscript.