Profiling the transcriptome of Sclerotium cepivorum Berk related to white rot on garlic ( Allium sativum Linnaeus )

1 Instituto Tecnológico de Celaya. Departamento de Ingeniería Bioquímica. Avenida Tecnológico y A. García-Cubas S/N. Colonia Alfredo V. Bonfil. Apartado postal 57, Celaya, Guanajuato, CP 38010, México. 2 Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias. Campo Experimental Bajío. Unidad de Biotecnología del Bajío. Carretera Celaya-San Miguel de Allende km 6.5. Apartado postal 112. Celaya, Guanajuato, CP 38010, México. 3 CA de Ingeniería de Biosistemas. Facultad de Ingeniería. Universidad Autónoma de Querétaro. C.P 76010. Santiago Zhang de Querétaro, Qro, México. 4 Departamento de Genética, Facultad de Biología, Universidad de Sevilla, Apartado 1095, 41080 Sevilla, Spain.


INTRODUCTION
White rot, caused by Sclerotium cepivorum Berk, is the predominant disease of garlic crops worldwide.Once *Corresponding author.E-mail: lorenzogo@yahoo.com.Tel: +52 461 6117575 ext.324.Fax: +52 461 6117575 ext.402.introduced into the field, the sclerotia can survive for over 20 years without the presence of an Allium sativum Linnaeus host.The sclerotia are specifically induced to germinate by Allium root exudates, in particular alkyl cysteine sulfoxides (ACSOS) (Coley et al., 1990;Davies et al., 2007;Ulacio-Osorio et al., 2006).Current control methods for white rot involve intensive fungicide application and crop production on uninfected land (Miñambres et al., 2010;Melero-Vara et al., 2000).However, this approach is becoming less practical as soil microbes become more efficient at degrading the active fungicidal chemicals thus reducing the efficacy of many fungicides.Several alternative strategies for white rot control are now being utilized as part of an integrated control program that includes cultural practices such as soil solarisation and biological control (Davies et al., 2007).Unfortunately, fungicide application is still required for effective management with these systems.
The molecular mechanisms that participate in the regulation of the interaction fungal-garlic are still unclear.Studies of gene expression profiles during fungal-garlic interaction can provide clues about pathogenesis regulation and may lead to the discovery of molecular targets for novel antifungal drugs.Suppression subtractive hybridization (SSH) technique is now well known and has been used by many laboratories (Diatchenko et al., 1996;Yang et al., 1999;Guevara-Olvera and Acosta-Garcia, 2011).It also permits the identification of differentially expressed genes without the need to obtain previously cloned cDNA.
The aim of the present study was to use a genomic approach to identify and clone all the fragments of genes differentially expressed in S. cepivorum Berk during pathogenesis on garlic using SSH method and cDNA arrays.

Fungus and plant material
S. cepivorum Berk, strain C2 an isolated from Cortazar, Guanajuato, México and Allium sativum Linnaeus Texcoco provided by the Instituto Nacional de Investigaciones Forestales Agricolas y Pecuarias (INIFAP) was used for experimental work.

Fungal-garlic interaction
One square centimeter of S. cepivorum Berk young mycelium grown in potato dextrose agar (PDA) was used to inoculate garlic clove during 72 h in a 50 ml glass flask containing 20 ml of double distilled water.As control, the fungus was inoculated in absence of the garlic.

Isolation of total RNA
RNeasy plant mini kit (Qiagen, Hilden, Germany) was used to extract the total RNA from mycelium in absence (driver) and presence (tester) of garlic.RNA purified by RNeasy column was analyzed for integrity and size by formaldehyde agarose gel electrophoresis, and quantification and purity of RNA by A260/280 value, using a Jenway 6405UV/vis spectrophotometer (Jenway, Dunmow, UK).

Synthesis, amplification and purification of cDNA
One micro-gram of total RNA of each conditions of fungal Medina et al. 2753 development was used as template to synthesize the first strand of cDNA using the Superscript II Reverse Transcriptase (Life Technologies, Rockville, MD, USA) and the SMART TM PCR cDNA synthesis kit (Switch Mechanism At the 5´ end of RNA Transcript) (Clontech, Palo Alto, CA, USA), then amplified by LD-PCR with 15, 18, 21, and 24 cycles separately (Davies et al., 2007) and analyzed through 1.2% agarose gel electrophoresis in order to identify the optimal cycle number for which a suitable amount of PCR product is obtained rather than to build the SSH library.For SSH control, 1 µg placental total RNA was used to synthesize cDNA driver and placental cDNA with Hae III-digest øX174 DNA was used as cDNA tester.CROMA-SPIN 1000 column (CLONTECH, Palo Alto, CA, USA) was used to purify cDNA.

Isolation of differentially expressed cDNA fragments
Suppression subtractive hybridization was conducted using the CLONTECH PCR-Select TM cDNA subtraction kit (CLONTECH, Palo Alto, CA, USA).The tester (garlic presence) and driver (garlic absence) cDNAs were partially digested with Rsa I, a four basecutting restriction enzyme that yields blunt ends.The tester cDNA fragments were divided into two aliquots, and each was ligated separately with adapter 1 and adapter 2 resulting in two populations of tester cDNA.A small amount of each tester population (600 ng) and driver in excess (2 μg) were mixed, heat-denatured, and allowed to anneal 8 h at 68°C.The two samples from the first hybridization were combined and annealed with additional freshly denatured driver cDNA (1 μg), overnight at 68°C.A primary PCR was conducted to amplify those cDNAs that represented differentially expressed genes.A secondary PCR amplification was conducted using nested primers 1 and 2R to reduce background levels (CLONTECH PCR-Select TM cDNA subtraction kit).The secondary PCR amplification products were electrophoresed and fragments longer than 500 bp were sliced using a scalpel and purified by QIAEXII Gel extraction kit (QIAGEN, Hilden, Germany).

Cloning and screening of subtraction fragments
PCR fragments (0.2 µg) were ligated to the pCR2.1-TOPOcloning vector (1 μl) according to the manufacturer´s instructions (INVITROGEN, Carlsbad, CA, USA). 2 µl ligation reaction solutions were transformed into 50 μl of Escherichia coli chemically competent cells strain TOP 10.The transformation culture was plated on Petri dishes containing LB/kamamycin/IPTG/X-gal, and white colonies were screened for insert fragment.Individual white bacterial transformants were cultured into LB/ampicillin/kanamycin medium and then it was shaken at 37°C overnight, and plasmid was screened for inserts presence using the restriction enzyme EcoRI (Invitrogen, Carlsbad, CA, USA).

Storage of library
Selected white colonies containing recombinant plasmid were inoculated separately into 5 ml LB/ampicillin/kanamycin solution, and it was shaken at 37°C overnight.Then 500 μl of each culture were added into 2 ml cryogenic vial (Corning, Acton, MA, USA) containing 500 μl 100% glycerol and kept at -80°C.

Construction of cDNA arrays
Five micro-grams of each recombinant plasmid were spotted into 7X10 cm BrightStar TM -Plus positively charged nylon membrane (Ambion Inc, Austin, TX, USA) to construct 6X4 (usually 12X4) clones array using a Slot Blot Manifold Hoefer PR 648 (Amersham Biosciences, Buckinhamshire, UK).As negative control, 5 μg of pCR2.1-TOPOcloning vector and 5 μg of recombinant plasmid containing an internal fragment of ScGpdh gene (369 bp; Accession number DQ522162) from S. cepivorum Berk as internal control were spotted, respectively.100 ng for each driver and tester cDNA were added as positive controls.

Preparation and labeling of cDNA probes and membrane hybridization
Replicates of the SSH library were hybridized by Southern analysis (Sambrook et al., 1989) with the driver or tester cDNA probes.These probes were generated by incorporating fluorescein-11-dUTP using Gene Images CDP-Star random prime labeling module according to the manufacturer´s instructions (Amersham Pharmacia Biotech Inc, Piscataway, NJ, USA).Detection of fluorescein-labelled probes in Southern dot blots was performed using Gene Images CDP-star detection module (Amersham Pharmacia Biotech Inc, Piscataway, NJ, USA), employing anti-fluorescein alkaline phosphates conjugate and CDP-Star detection reagent.

DNA sequencing and database comparison
The nucleotide sequences of differentially expressed fragments were determined using the ABI PRISM 310 genetic analyzer (Perkin Elmer, Norwalk, CT, USA).On-line database comparisons were performed using blastx algorithm (Altschul et al., 1990) from National Center for Biotechnology Information (NCBI).

Differential expression of ScOah
Total RNA isolation from S. cepivorum Berk mycelium in absence (driver) and presence (tester) of garlic followed of reverse transcription were conducted as described above.PCR amplification of the ScOah cDNA (175 bp) was performed using specific sense and antisense primers whose sequences were as follows: 5´-CTCTTGAATAGCCAACATAGCCG-3´ and 5´-AAAGGAGATGGCTGCGAAGACT-3´, respectively.As internal control, PCR amplification of a constitutive gene ScGdph cDNA of 369 bp (Accession number DQ522162) was obtained using specific sense and antisense primers whose sequence were as follows: 5´-GGTGTCAACAACGAGACCTACA-3´ and 5´-GCGGACAGTCAAGTCAACAAC-3´, respectively.One hundred nano-grams of each driver and tester cDNA were used as a template.PCR products were separated by agarose gel electrophoresis, (1.2%) and the optical density of the EtBr-stained bands was recorded using a Vilber Lourmat gel documentation system (Marne-La-Valée Cedex, France) equipped with an ultraviolet light transiluminator.

RNA isolation and cDNA synthesis for SSH
To detect genes involved in S. cepivorum Berk pathogenesis, fungal mycelium was harvested 72 h post inoculation and before any adhesion to the garlic tissues surface (not shown) for RNA isolation.The amount of RNA extracted from mycelium in absence (driver) and mycelium in presence (tester) of garlic was 15 and 20 μg, respectively.The first strand and double-stranded cDNA were obtained.For cDNA subtraction, optimization of the number of PCR cycles was done to ensure that cDNA will remain in exponential phase of amplification.Over-cycled cDNA is a very poor template, on the other hand, undercycling results in a lower yield of PCR product.The products obtained with 15, 18, 21, and 24 cycles were separated on 1.2% agarose gel electrophoresis.For tester (Figure 1) and driver (not shown), the optimal cycles number were 18 while for placental RNA, the 21 cycle sample was selected (Figure 1).

PCR-selected cDNA subtraction
cDNA before digestion with Rsa I, appeared as a smear of 0.5 to 10 kb on 1% agarose gel electrophoresis, and after digestion, the cDNA size was smaller (0.1 to 2 kb) (not shown).After SSH, a primary PCR was conducted to amplify cDNAs, which represented differentially expressed genes.A secondary PCR amplification was performed using nested PCR primers 1 and 2R to reduce background, and several bands could be clearly seen among these smears (Figure 2, Lane 1), while for SSH control øX174/Hae III DNA fragments are seen (Figure 2, Lane 2).PCR products were isolated and cloned into pCR 2.1TOPO TA vector.120 White colonies were selected for plasmid DNA isolation and analyzed for inserts presence using EcoRI restriction enzyme.96 Recombinant plasmids, containing fragments with estimated size around 500 bp, were selected.

Differential expression of genes identified by SSH
Differential expression was assayed by Southern blot analysis.Ninety six clones were spotted in duplicate on nylon membrane and hybridized with both tester and driver probes.Forty six clones hybridized exclusively with tester probe.Screen from 20 clones are shown (Figure 3).Clone HR4, HR35, HR46, HR49 and HR57 are not included in Table 1 because they are redundant in sequence.Only clones HR9 and HR40 were expressed at a slightly increased level in tester cDNA, while 18 clones were expressed at a much higher level in tester cDNA, confirming the differential gene expression during pathogenesis of S. cepivorum Berk.

Differential expression of ScOah transcript during pathogenesis of S. cepivorum Berk
ScOah is expressed at a much higher level during the compatible reaction with garlic than when grown in absence of a host (Figure 4B).Although non-quantitative, this technique did provide evidence for differential expression of ScOah transcript during pathogenic stage of S. cepivorum Berk.

DISCUSSION
White rot disease of onion, garlic and other Allium spp.results from the attack by the soilborne fungus S. cepivorum Berk and is a continuing concern for worldwide garlic production.Garlic producers are really concerned about the big losses caused by S. cepivorum Berk pathogenesis.Several control methods have been employed; however, these become less effective as the pathogen is able to degrade the fungicides.One alternative still unexplored is the genetic engineering.In order to identify genes involved in S. cepivorum Berk pathogenesis on garlic, Suppression subtractive hybridization method was used.Blast analysis indicated that thirteen of the reported above fourteen sequences had strong homology to genes of known function or sequences present in GenBank database and represent at least 5 classes of putative genes.Blast analysis showed that clone HR51 had strong homology to Nsf 1p, a cysteine desulfurase of Magnaporthe grisea (Score, 250.00;E-value, 1e-76) (Lu et al., 2004), which could be involved in alkyl cysteine sulfoxides cleavage contained in Allium sativum Linnaeus exudates In addition, clone HR23 is related to Rds1p, a regulator of drug sensitivity from Saccharomyces cerevisiae (Score, 32.30; Evalue 4.10) (Wolfe et al., 1999).It is a putative zing-finger transcriptional activator of genes involved in multistress response, so both genes could play a pivotal role avoiding host response.
Oxalic acid plays an important role in a compatible pathogen-host interaction since, in several cases its secretion has been shown to be required for pathogenesis (Dutton and Evans, 1996).Oxalate is produced by a variety of fungi, including saprophytic and phytopathogenic species.Additionally, the role of oxalic acid secreted as a pathogenicity factor by the ubiquitous phytopathogenic Ascomycete fungus Sclerotinia sclerotiorum in inhibiting the oxidative burst from host plants is strongly documented (Cessna et al., 2000).The oxidative burst is the controlled release of O 2 and H 2 O 2 at the site of pathogen invasion (Wojtaszek, 1997;Ebel and Mithofer, 1998), which is one of the earliest responses against microbial invasion.It have been shown that Oxalic acid (OA) secreted by the S. sclerotiorum is a key pathogenicity factor; moreover, transgenic oilseed rape plants constitutively expressing TaOxo from wheat (Triticum aestivum) display considerably increased oxalate oxidase (OXO) activity and enhanced resistance to S. sclerotiorum (Dong et al., 2008).Speculation regarding the mechanism or mechanisms by which oxalate secretion might enhance Sclerotinia virulence currently centers on three modes of action (Dutton and Evans, 1996).First, several of the fungal enzymes secreted during invasion of plant tissues such as endopolygalacturonase (EP) have maximal activities at low pH.Several researchers have postulated that oxalate might aid virulence by decreasing the apoplastic pH to a value better suited for enzymatic degradation of plant cell walls (Bateman and Beer, 1965).Second, oxalate may be directly toxic to host plants, presumably because of its acidity, the secretion of oxalate has been suggested to weaken the plant, thereby facilitating invasion (Noyes and Hancock, 1981).Finally, chelation of cell wall Ca 2+ by the oxalate anion has been proposed both to compromise the function of Ca 2+ -dependent defense responses and B to weaken the plant cell wall (Bateman and Beer, 1965).Additionally, maceration of onion (Allium cepa Linnaeus) host tissue by S. cepivorum Berk correlates with fungal oxalic acid secretion and EP activity (Stone and Armentrout, 1985).On the other hand, RT-PCR analysis for ScOah transcript described in this study (Fig. 4B), demonstrate that this gene is more strongly expressed during pathogenesis.

Figure 2 .
Figure 2. Secondary PCR of subtracted samples and screening for inserts.(A) PCR using nested 1 and 2R primers.Lane M, 1 kb DNA marker; lane 1, SSH from tester and driver; lane 2, SSH from tester and driver human placental total RNA.

Figure 3 .
Figure 3. Differential screening of clones from subtracted library generated using SSH by Southern blot analysis with driver; (A) and tester (B) cDNA as probes.Each number indicates HR clones.One hundred nano-grams of cDNA were added as positive controls (C+): A, driver; B, Tester.

Figure 4 .
Figure 4. Homology between HR62 clone a putative ScOah from S. cepivorum Berk and BfOah from Botryotinia fuckeliana Whetz and its expression level in driver and tester DNA.(A) Alignment between putative OAH proteins from S. cepivorum Berk and B. fuckeliana Whetz using the blastx algorithm from NCBI, +, indicates similar amino acids.(B) RT-PCR detection of ScOah (175 bp) transcript by amplification with specific primers.Lane M, 1 kb DNA marker; lane 1, ScOah amplified from cDNA driver; lane 2, ScOah amplified from cDNA tester.ScGpdh (369 bp) was used as housekeeping gene.

Table 1 .
Functional classification of gene products activated during .S cepivorum Berk pathogenesis toward garlic.
Consequently, we speculate that HR62 (ScOah) gene plays an important role during garlic (A.sativum Linnaeus) white rot caused by S. cepivorum Berk, codifying the OAH enzyme to synthesize oxalic acid avoiding the oxidative burst of garlic defense response.Only clone HR15 (Accession number DR774655) with homology (Score, 42.00; E-value, 0.008) to a hypothetical protein of Streptococcus suis (Guevara-Olvera et al., 2006) could represent a novel gene involved in S. cepivorum Berk pathogenesis toward garlic.Seven among 14 clones exhibited P values between 1.80-8.80units(Table1),suggesting that all of them are putative genes, these values could change obtaining 5´and 3´ cDNA by rapid amplification of cDNA ends (RACE).In summary, molecular cloning of 14 gene fragments identified in this work raises the possibility to use them as genetic targets for garlic white rot control which could give great benefit for agriculture.