RNAi technology targeting Pb GP 43 and Pb P 27 in Paracoccidioides brasiliensis

Efficient technologies for gene silencing would be important to carry out functional analysis with P. brasiliensis genes, as well as for a better understanding of the biology and pathogenesis of this pathogenic fungus. Due to the fact that homologous recombination is unusual in P. brasiliensis, the development of knockout isolates is currently non-feasible. The goal of this work was to assess RNA interference (RNAi) technology as an alternative tool for gene silencing previously employed successfully in H. capsulatum. For this purpose, we built different inverted repeat transgenic hairpin constructs to down-regulate the PbGP43 and PbP27 genes known to codify for two fungal immunogenic proteins that elicit a strong immune response during experimental paracoccidioidomycosis. Using the RNAi strategy, a reduction in the mRNA levels of the PbGP43 and PbP27 genes was observed during the first 20 days after selection; however, in the transformed yeast cells, the gene silencing status proved non-stable through the assay. We demonstrated that electrotransformation was suitable to transform P. brasiliensis yeast cells and integrate the hairpin constructions; nonetheless, gene silencing was not stable along the experimental time. A detailed analysis of the underlying molecular RNAi machinery may provide further insights into the intracellular mechanism that governs this reverse genetic tool.


INTRODUCTION
RNA interference (RNAi), a natural mechanism conserved all along evolution, has been implicated in gene silencing in eukaryotic systems [1].In addition, RNAi participates in the regulation of genetic expression mediated by certain classes of small endogenous RNAs such as micro RNA (miRNA), which acts by using double stranded RNA (dsRNA) homologous to the target sequence [2].
Due to the fact that generation of knockout isolates is time consuming and requires sequential positive and negative selection steps, in order to enrich the desired recombination event, RNAi technology has rapidly become one of the key methods in functional genomics studies, and is used to block gene expression and create potential phenotypes capable of yielding clues concerning the function of these genes [2].This technology has been successfully used to obtain gene disruption in dimorphic fungi, e.g.Histoplasma capsulatum [3] and Blastomyces dermatitidis [4].
Paracoccidioides brasiliensis, a thermally dimorphic fungus, is the etiological agent of paracoccidioidomycosis (PCM), an important systemic, endemic mycosis in Central and South America [5].In this fungus, gene disruption methods are even more laborious and currently seem unfeasible.This could be due to the presence of dominant illegitimate recombinant events (by non-homologous end-joining) overriding homologous recombination [6].In P. brasiliensis, RNAi could be employed as an alternative method to transcriptional gene silencing mediated by sequence-specific mRNA depletion.The aim of this work was to determine if RNAi strategy was a proficient tool to down regulate both PbGP43 (a 43 kDa gly-coprotein) [7] and PbP27 (a 27-kDa protein) [8] gene expression, two P. brasiliensis immunogenic antigens with as yet unknown biological functions.
In this work, by means of bioinformatics analysis, we demonstrated the presence of genes involved in the RNAi route in the Paracoccidioides spp.genome.Our results indicated that RNAi strategy appears to be an effective method to integrate the RNAi hairpin constructions in the genome of this fungus; however, the gene silencing status was not stable along the time in the isolates evaluated.

Strains and Culture Conditions
P. brasiliensis Pb339, a strain that produces high quantities of extracellular antigens, especially gp43 [9][10][11], and p27 [8,12] during its parasitic phase was used in this study.Yeast cell cultures and growth curves were performed in BHI media supplemented with 1% glucose (Beckton Dickinson and Company, Sparks, MD) at 37˚C with aeration in a mechanical shaker and were routinely collected during the early exponential phase (72 -96 h).Escherichia coli DH5 grown at 36˚C in Luria Bertani (LB) culture medium supplemented with appropriate antibiotics, was used for cloning and plasmids propagation assays [13].

Molecular Cloning and Silencing Cassette Construction
The RNAi plasmid pCR99 (Figure 1  E2RNAi and PbP27RNAi cassettes were designed, amplified and constructed using the same strategy employed to PbGP43E1RNAi cassette (Figure 1(b)).The primers are described in supplementary Table S1, the sizes of the target sequences were 788-pb and 500-pb respectively.All PCR products were amplified using the Platinum high-fidelity TaqDNA polymerase (Invitrogen, Carlsbad, CA, USA).

P. brasiliensis Transformations and Screening
The Pb339 strain was electrotransformed with PmeI-linearized plasmid according to the protocol previously described [16].Briefly, P. brasiliensis yeast cells were grown in BHI batch cultures to their exponenttial growth phase with shaking at 36˚C, washed once with 10% mannitol, sterilized by filtration as electroporation solution.
Yeast cells were electrotransformed with 2 g of PmeI-digested pCR99 constructions (PbGP43E1RNAi, PbGP43E2 RNAi and PbP27 RNAi) and an empty vector (PbEV) as a control, in a Gene Pulser Electroporator (Bio-Rad, Hercules, CA), using the following conditions: capacitance of 25 μF, resistance of 600 Ω and set voltage of 0.75 kV [16,17].Following transformation, cells were spread onto selective BHI media supplemented with 100 μg/ml of hygromycin B (Sigma, Aldrich, MO, USA).Selection plates were monitored for colony forming ability at 37˚C for 15 to 20 days.The phenotypic stability of P. brasiliensis transformants yeast cells was determined by analyzing the stability of hygromycin B resistance [18].Sixty individual colonies were selected and subcultured in selective medium, (solid BHI containing 150 μgml hygromycin B) each 5 days at 37˚C for three consecutive times and then subcultured in liquid BHI containing 150 μgml hygromycin B three times, RNA extraction was then done and used in the RT-qPCR assay.

Molecular Detection of the Hygromycin Resistance Gene (HPH)
Genomic DNAs from PbWt, PbEV, PbGP43E1RNAi, PbGP43E2 RNAi and PbP27 RNAi transformants yeast cells were isolated using the glass beads protocol described by Van Burik (1998) [19].In order to confirm the presence of the hygromycin B resistance cassette, PCR analysis was carried out to detect an hph 1000-bp amplification product using primers hphF (5'-AACTCACC-GCGACGTCTGTCGA-3') and hphR (5'-CTACACA-GCCATCGGTCCAGA-3').PCR amplification included 30 cycles of 1 min at 94˚C, for denaturation, 1 min at 68˚C for annealing, and 1.5 min at 72˚C for extension.The reaction products were analyzed in 1% agarose gel and visualized with ethidium bromide under UV light.

RNA Extraction, cDNA Synthesis and Real-Time RT-qPCR Analysis
P. brasiliensis yeast cells were grown in liquid BHI supplemented with glucose 1% and hygromycin B 150 g/ml at 37˚C and harvested after 5 days of growth.RNA was obtained using the TRIzol ® reagent according to the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA).Total RNA was treated with DNaseI (Invitrogen, Carlsbad, CA, USA) and tested for chromosomal DNA contamination using conventional PCR for the βtubulin gene [20].cDNA was synthesized using 1 µg of total RNA and Superscript III reverse transcriptase according to the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA).Real-time PCR (RT-qPCR) was performed using Maxima ® SYBR Green/Fluorescein qPCR Master Mix, according to the manufacturer's instructions (Fermentas Maryland, USA).The CFX96 Real-Time PCR Detection System (Bio-Rad, Headquarters Hercules, California, USA) was used to evaluate PbGP43 or PbP27 gene expression; -tubulin was selected as house keeping gene [20].Melting curve analysis was done after the amplification phase to eliminate the possibility of nonspecific amplification or primer dimer formation.Folding changes in mRNA expression were calculated using the 2 ∆∆CT formula, where ∆∆CT is the difference between target and -tubulin genes [21].Each experiment was carried out in triplicate and the expression level was measured three times.

In P. brasiliensis Electrotransformed Yeast Cells, PbGP43 and PbP27 Gene Silencing Achieved by Using the RNAi System, Was Effective but Not Stable during Time of the Experiments
Three different RNAi constructions were designed from exon 1 and exon 2 from PbGP43 (PbGP43E1 RNAi, PbGP43E2 RNAi) and one from PbP27 (PbP27 RNAi), and transformed individually in PbWt yeast cells.Fourteen transformants with reduced gene expression levels were obtained by electrotransformation with the constructions PbGP43E1-RNAi (n: 6), PbGP43E2-RNAi (n: 5) and PbP27-RNAi (n: 3) and selected after phenotypic and mitotic stability tests (Table 1).The presence of the RNAi cassette in the knock down strains and the yeast cells transformed with PbEV was demonstrated by hph gene amplification.This gene was no observed in PbWt yeast cells (Figure 2).In all yeast transformants cells, mitotic stability remained stable in vitro cultures with continuous subcultures in selective medium containing hygromycin B; in parallel, all transformants were cultured several times in non-selective medium; after this, all of them remained  Number of transformants obtained using 2 g of PmeI-linearized plasmid DNA;  Mitotic stability of putative transformants was analyzed after 5 subcultures on non-selective medium followed by plating on hygromycin B (150 g/ml).their capacity to grow in the hygromycin B-containing medium (Table 1).
After 20 days of growth in culture media, a decrease in gene expression level ranging from 64% to 84% was observed in PbGP43E1 RNAi isolates (Figure 3(a)), and from 42% to 71% in the PbGP43E2 RNAi isolates (Figure 3(b)).In an equal manner, in the yeast cells transformants that had been obtained using the PbP27RNAi construction, a decrease in PbP27 gene expression level ranging from 39% to 79%, was observed in comparison with the expression levels in PbWt and PbEV (Figure 3(c)).However, after 45 and 60 days of growth in the selective culture media, an increase in PbGP43 gene expression level in both PbGP43E1 RNAi and in PbGP43E2 RNAi transformants was observed.Similarly, an increase of PbP27 gene expression was observed in PbP27RNAi yeast cells, indicating that in P. brasiliensis electrotransformed yeast cells silencing had not been stable during the course of time.

DISCUSSION
Efficient technologies to achieve gene silencing could help to undertake functional analysis of P. brasiliensis genes and increase the understanding of the biology and virulence attributes of this fungus, overcoming many of the difficulties associated with traditional gene disruption.RNAi does not depend on the homologous recombination machinery, making this strategy an attractive alternative to gene silencing [17].
In this study, we found that in Paracoccidioides spp, there were homologous proteins involved in the RNA silencing process suggesting the presence of the components required for RNAi gene silencing thus supporting the hypothesis that in this fungus such event should occur under natural conditions.
Our results show that the electroporation is an efficient tool to introduce the linearized construction of RNAi in P. brasiliensis yeast cells genome, as confirmed by the presence of the hph gene into the transformed cells.However, the reduced mitotic stability observed in the transformant isolates would indicate that the fungus tends to loose the hph fragment during continuous fungal growth under non-selective pressure.As previously reported, the efficiency of silencing depends upon several factors including the length and structure of the hairpin constructions.Rappleye et al., (2004) [17] have shown, that a greater silencing effect was observed with a shorter loop and a longer target gene sequence; these conditions were used in our experiments but our results were different.
Currently, a gene silencing strategy based on antisense-RNA (asRNA) has been efficiently employed in P. brasiliensis to induce the knockdown of PbCDC42, PbHAD32, PbAOX, and PbHSP90 [23][24][25][26].Additionally, based in our previous results using the antisense RNA strategy targeting PbGP43 and PbP27 (unpublished), and  the results obtained in this work using RNA interference strategy, we conclude that the non-stability in gene silencing is due to strategy and transformation methodology employed to down regulate specific gene.However, it will be important to explore an alternative gene silencing strategy such as RNAi, in order to diminish the time required for obtaining transformants with the desired phenotype, and achieve higher transformation efficiency.
For the above reasons, and taking into account that RNAi machinery is present in the Paracoccidioidesspp genomes, and that the strategy has been employed efficiently in H. capsulatum, we conclude that RNAi is an easier and faster tool than asRNA for gene silencing and has all potentialities for the study of many functional genes in P. brasiliensis.However further studies need to be conducted for the synthesis of new and more stable constructions based on RNAi technology.Presently, use of the transformation system mediated by A. tumefaciens, which could show best results in P. brasiliensis gene silencing [23][24][25][26].This is the first report using electrotransformation of hairpin construction based on RNAi technology targeting two important P. brasiliensis antigens (gp43 and p27).A detailed analysis of the underlying molecular RNAi machinery may provide further insight into the intracellular mechanism that governs this reverse genetic tool.Present results provide an initial framework for further studies of virulence factors in fungi such as P. brasiliensis, in which the genetic manipulation represents a challenge.

Figure 1 .
Figure 1.Hairpin RNAi constructions aimed at triggering gene silencing of PbGP43 and PbP27 in P. brasiliensis.A. Map of the telomeric plasmid pCR99 used to trigger RNAi in P brasiliensis.H. capsulatum telomeres (TEL) were described previously 16.The CBP1 promoter (889 bp) was used to initiate transcription of RNAi targets and the 733 bp intergenic region down-stream of the CatB gene was used for transcriptional termination signals (T).This plasmid was used to clone individually the inverted copies of target genes for PbGP43 exon 1 (PbGP43E1), exon 2 (PbGP43E2) and PbP27, separated by a lacZ fragment to produce the RNA hairpin.Kan: kanamycin resistance; hph: hygromycin resistance cassette.B. RNA hairpin constructs targeting PbGP43E1 (457 bp), PbGP43E2 (788 bp) and PbP27 (500 pb).

Figure 2 .
Figure 2. Molecular analysis of the integration of the hph resistance cassette into P. brasiliensis putative transformants.Wild type host strain, PbWt harboring the empty vector (PbEV).Three putative transformants from the RNAi hairpin constructions selected randomly were subjected to PCR using the phosphotransferase gene (hph) specific primers, hph-F and hph-R, in order to amplify a 1000 bp internal fragment of the hph gene MW: DNA molecular size marker.

Figure 3 .
Figure 3. Gene expression levels of PbGP43 and PbP27 by RT-qPCR.(a) PbGP43 gene expression in the PbWt, PbEV and the PbGP43E1 transformant yeast cells, selected after phenotypic and mitotic stability.(b) PbGP43E2 RNAi gene expression in the PbWt, PbEV and PbGP43E2 RNAi transformants and (c) PbP27 gene expression in PbWt, PbEV and PbP27 RNAi transformants.All evaluations were performed in the yeast cells after subculture for 20, 40 and 60 days.Low expression was observed during the 20-day evaluation but not in those corresponding to the 45 and 60 days evaluations.Gene expression levels obtained by RT-qPCR were normalized to the level of expression of the internal control gene TUB2 20.

Table 1 .
Genetic transformation of P. brasiliensis by electroporation using hairpin RNAi constructions.

Table 2 .
[14]ence of related RNA silencing genes in Paracoccidioides SP.Accession numbers of homologous genes founded in P. brasiliensis (Pb18 and Pb03) and P. lutzii (Pb01) genomes, using BLASTX with predicted protein involved in the RNA machinery in N. crassa[14].