Molecular determination of Cladosporium sp. isolated from diseased A. angustifolia

100 mg of a mixture of hyphal cells and conidiophores were taken from PDA containing the filamentous fungus. The cells were disrupted with acid-washed glass beads (G-9268, 425-600 μm 0.5 mm diameter, Sigma-Aldrich Co., USA.) in the presence of 300 µL of DNAzol reagent (Thermofisher ^TM, USA). The following steps were carried out in accordance with the manufacturer’s recommendations. The isolated DNA was washed twice with a mix of phenol:chloroform:isoamyl alcohol (25:24:1, v/v: Sigma-Aldrich Co., USA). PCR reactions were carried out using 1 µg of DNA dissolved in a final volume of 50 µL. The identity of the microorganism was determined by the amplification of internal transcribed spacers (ITS) using the primers ITS1 (TCCGTAGGTGAACCTGCGG) and ITS4 (TCCTCCGCTTATTGATATGC) and performing the PCR conditions described by ^{White et al. (1990)}. The amplicon was cloned into TOPO TA cloning vector (Thermofisher ^TM, USA) and sequenced in triplicate in an ABI PRISM 3700 instrument sequencer (ABI, Foster City, CA). The allergenic protein of C. herbarum (heat shock protein 70, accession S83210) was amplified by reverser transcription from total RNA, using TRIZOL (Thermofisher ^TM, USA) and SuperScritp II enzyme (Thermofisher ^TM, USA). PCR assays were carried out with Platinum® DNA polymerase high fidelity (Thermofisher ^TM, USA) and the primers GAGATCCTTCTTCTCGACGTCG and CCTTCTAATCGTTAACGCCATG. The PCR protocol to amplify this cDNA was 94 °C for 5 min of initial denaturation, followed by 35 cycles at 94 °C for 30 s, 57 °C for 30 s, 72 °C for 1 min and a final extension of 72 °C for 4 min. The amplicon was visualized in an agarose 1 % gel electrophoresis after staining with ethidium bromide and it was simultaneously cloned and sequenced as previously described. The sequences were compared with those deposited in the nucleotide data base of the NCBI (by nucleotide BLAST) and MycoBank.

Pathogenicity tests of Cladosporium sp. on A. angustifolia stalks

Healthy young stalks of A. angustifolia were excised from the plant and were immersed in water to avoid dehydration along the experimental tests and finally maintained at 25 °C in normal photoperiod (12 hours light). Conidiophores from the fungus were directly harvested with a sterile needle and were immediately inoculated in the stalks by mechanic penetration in order to examine Koch’s postulates. The process was carried in a sterile cabinet. The appearance of the spots was observed during 47 days.

Isolation of ent-kaurenoids and its in vitro evaluation on Cladosporium sp.

500 g of dried roots from Zaluzania montagnifolia were extracted for 10 days at room temperature with 2 L of absolute ethanol (J.T. Baker, USA). The crude ethanolic extract was filtered with Whatman filter paper grade 1 and subsequently reduced until dryness in a rotary evaporator (Buchi R200). The resulting gum was recovered and then resuspended in the same solvent in order to obtain stock solutions of 100 mg mL^-1. These solutions were directly used for broth microdilution and agar dilution methods or were fractioned by HPLC for the obtainment of kaurenoic acid (ent-16-kauren-19-oic acid) and grandiflorenic acid (ent-kaur-9(11),16-dien-19-oic acid). Semipreparative purification was carried out in a Hewlett Packard 1050 system coupled to a HP G1306A diode array detector equipped with a Varian C₁₈ column (250×4.5 mm I.D.; 5 μm particle size). The metabolites were collected during continuous runs (100 μL injection volume) using an isocratic mobile phase consisting of 70 % acetonitrile (J.T. Baker, USA) with 0.05 % acetic acid (v/v) at 1 mL min⁻¹ flow rate as previously described by ^{Villa-Ruano et al. (2009)}. The retention time was compared with that of authentic standards available in our laboratory. The diterpenoids were resuspended in methanol (J.T. Baker, USA) for further antifungal assessment. Agar diffusion method was carried out by dissolving distinct concentrations (dose-response curves of 10-300 µg mL^-1) of ethanolic extracts and/or pure diterpenoids in 10 mL of PDA medium (^{Valgas et al., 2007}). The phytopathogenic fungus was inoculated in the medium by massive streaking and these experiments were performed in quintuplicate. The plates were visualized in a white light transilluminator. The obtainment of MIC values was achieved by the broth microdilution method for filamentous fungi proposed by ^{Pfaller et al. (2000)} using resazurin (Sigma-Aldrich Co., USA) as an indicator of cell viability. The dose-response curves (10-300 µg mL^-1) to obtain MIC values were performed in quintuplicate. Colorimetric reactions were measured at 545 nm. The MIC values were additionally validated by ANOVA-Tukey Test (p<0.01) using the software GraphPad 6.05.

As a result of the incubation of the necrotic tissue in PDA medium, it was isolated a brown-green filamentous fungi (between others isolates) with small terminal conidia (Figure 1 A-B). This fungus was observed as the most dominant species in all the incubations. Macroscopy and microscopy of the microorganism suggested its similarity with Cladosporium species (^{Bensch et al., 2012}). PCR reactions with genomic DNA and cDNAs from reverse transcription produced fragments of ~500 and ~800 bp (Figure 1 C). The analyses of the ITS sequence with NCBI and MycoBank data bases demonstrated its homology (94 %) with the 18S ribosomal RNA gene from the isolate “olrim85” (ID: AY354234.1) of C. herbarum (Figure 2A). Whereas, the coding sequence of the allergenic protein had 99 % homology with the sequence ID:X81860.1 from C. herbarum deposited in the NCBI (Figure 2B). These molecular data strongly suggested the identity of the fungus as C. herbarum.

Figure 1. Macroscopy and optical microscopy of C. herbarum isolated from diseased Agave stalks. A, C. herbarum grown in PDA medium. B, hyphal organization and conodiophores of C. herbarum, the image was recorded at 40X. C, Molecular identification of C. herbarum by the amplification of a partial cDNA from the allergenic HSP70 (~800 bp) and the ITS from 18S ribosomal RNA gene (~500bp). The approximated molecular weights are shown. 

Figure 2. Results of the BLAST alignment for the original DNA sequences (query) from C. herbarum and its homology with those deposited in the nucleotide data base from the NCBI (subject). 

The inoculation of the fungus in asymptomatic Agave stalks exhibited necrotic symptoms since day 4 (Figure 3B). Despite the penetration with the sterile needle produced a damage in the stalks (Figure 3A), the inoculation of the fungus caused an evident systemic response and a darkening of the vasculature (Figure 3B). The systemic response was also perceived by emerging yellow spots around the necrotic cells which were visible since day 15 (Figure 3B). Interestingly, the lesions generated by the sole mechanical damage were almost imperceptible at day 47 (Figure 3A). Contrarily, those infected with the fungus were clearly aggravated at the same period of time (Figure 3B). These data strongly suggested the susceptibility of those Agave stalks to the C. herbarum infection. Despite this evidence supports the fact that C. herbarum is associated to the generation of necrotic spots, the involvement of another opportunistic microorganisms in the infection of open-air crops, cannot be discarded.

Figure 3. Effect of mechanical damage (A) and mechanical damage with inoculation by C. herbarum (B) in healthy young stalks from A. angustifolia. 

The in vitro antifungal activity of the ethanolic extracts from the roots of Z. montagnifolia against C. herbarum showed a clear inhibitory effect (Figure 4). Agar dilution method using pure ent-kaurenoic and grandiflorenic acids (usually the main constituents of the root extracts of the plant) exhibited their involvement in the observed antifungal effect (Figure 4). As shown in Figure 4, the fungus perfectly grew in sole PDA medium (0 µg mL^-1), whereas an evident growth inhibition was observed as the concentration of each component assessed was increased (20-120 µg mL^-1). These semi-quantitative experiments showed that the ethanolic extract containing both diterpenes was able to inhibit the normal grow of the fungus at less than 0.5 mg mL^-1. The assays performed with the broth microdilution method allowed to the obtainment of MIC values. These values reflected the concentration of the antimicrobial agent in which there are no signs of growth. The values were 150.4±3.5, 100.3±2.1, and 80.7±1.4 µg mL^-1, for ethanolic extract, kaurenoic acid and grandiflorenic acid, respectively (Figure 5). Significant differences (p<0.01) were observed among those treatments (Figure 5). Thus, the use of the ethanolic extract should be considered as a less expensive alternative for further in vivo trials in order to determine its possible use in the biological control of C. herbarum. The antimicrobial activity of these assayed diterpenes has been already proposed and confirmed in diverse bacterial and fungal species (^{Villa-Ruano et al., 2016}). This work contributes to the investigation of novel antimicrobial activities from plant extracts containing ent-kaurenoids. Considering that Z. montagnifolia can be found in the vegetation of those regions, its viability as a source for further antifungal in vivo test should be contemplated.

Figure 4. Effect of different concentrations of the ethanolic extract (EE) of Z. montagnifolia, ent-kaurenoic acid (KA) and grandiflorenic acid (GF) in the in vitro growth of C. herbarum. 

Figure 5. Comparison of MIC values of the ethanolic extract from Z. montagnifolia (EE), ent-kaurenoic acid (KA) and grandiflorenic acid (GF) in the in vitro growth of Cladosporium herbarum. Bars indicate the standard deviation of five repetitions (n=5) and diverse letters indicate significant differences between treatments by ANOVA-Tukey test (p<0.01).