Robust Detection and Identification of Multiple Oomycetes and Fungi in Environmental Samples by Using a Novel Cleavable Padlock Probe-Based Ligation Detection Assay

Nucleic acids used in this study.Microorganisms were procured from the culture collection of Plant Research International B.V. (Table 1). Environmental test samples were derived from several Dutch hydroponic horticultural systems. These were designed to provide an examination of detection capabilities in complex environmental samples of practical relevance. To create samples with known plant pathogens, fungal and oomycete spores (10³ to 10⁶ spores, depending on the pathogen and horticultural system) were used to spike recirculation water systems. At least 4 weeks after pathogen inoculation, 250 ml of each recirculation water sample was collected and filtered and the filters were stored at −80°C prior to analysis. All gDNA extractions from cultured microorganisms, as well as filters, were performed with the Puregene gDNA isolation kit (Gentra-Biozym) according to the manufacturer's instructions.

For assay development and optimization, ligation targets were generated by performing PCR preamplification with various amounts of gDNA from cultured microorganisms as the PCR template. Ligation targets for the assay validation experiments were prepared with 10 ng gDNA extracted from the environmental samples as the PCR template. Depending on the experiment (see below), preamplification reactions were performed in the presence or absence of 0.1 aM internal amplification control (IAC) oligonucleotide (5′ TCCGTAGGTGAACCTGCGGCGGATCGTTACAAGGGTCTCCAACTACGTCTAGCGCATAGACCACGTATCGAAGCTAGGTGCATATCAATAAGCGGAGGA 3′). Specific PCR amplification of the internal transcribed spacer (ITS) regions of the fungal and oomycetal rRNA operons was achieved with the ITS1 and ITS4 primers (56). For each PCR, an initial 10-min incubation at 95°C was followed by 40 cycles consisting of 30 s at 95°C, 30 s at 54°C, and 72°C for 60 s. After the final cycle, samples were incubated at 72°C for 10 min and then cooled to 4°C.

PLP design.The PLP target complementary regions were engineered according to previously described design criteria (51) and connected by a 75-nucleotide compound linker sequence. The linker sequence contained a 20-nucleotide PLP-specific sequence (ZipCode) for array hybridization (Table 2), two spacer sequences (S1 and S2), a 15-dT sequence, and a polydeoxyuracil cleavage region. All ZipCodes have equal T_ms to allow standardized array hybridization conditions. The ZipCodes were chosen from the GeneFlex TagArray set (Affymetrix) so as to minimize PLP probe secondary structures and to optimize array hybridization conditions. Potential for secondary structures and ZipCode specificity was examined with Visual OMP 6.0 software (DNA Software Inc.). The prediction parameters were set to match ligation ([monovalent ions] = 0.025 M, [Mg²⁺] = 0.01 M, T = 65°C, [probe] = 20 nM) and array hybridization conditions (1× tetramethylammonium chloride, T = 55°C). When necessary, PLP arm sequences were adjusted to avoid strong secondary structures that might interfere with efficient ligation, as described previously (51).

A thymine-linked desthiobiotin was introduced for specific capture and release with streptavidin-coated magnetic beads (see below). The PLPs listed in Table 2 (for the entire probe sequence, see Fig. S1 in the supplemental material), and all of the other oligonucleotides used in this study, were synthesized by Eurogentec SA.

Ligation.PCR-amplified ITS products (see above) were used as templates for ligation. Prior to ligation, the PCR-amplified ITS products were fragmented by digestion with EcoRI, HindIII, and BamHI (New England BioLabs, Inc.) for 15 min at 37°C. Cycled ligation was performed in a 10-μl reaction mixture containing 20 mM Tris-HCl (pH 7.5), 20 mM KCl, 10 mM MgCl₂, 0.1% Igepal, 0.01 mM rATP, 1 mM dithiothreitol, 20 ng sonicated salmon sperm DNA, and 4 U Pfu DNA ligase (Stratagene). For multiplex detection, the optimized concentration of the individual PLPs was 20 nM. Reaction mixtures were prepared on ice and rapidly transferred into a thermal cycler. Before ligation, samples were denatured at 95°C for 5 min. The samples were subsequently subjected to 20 cycles of 30 s at 95°C and 5 min at 64°C. After the final cycle, the reaction mixture was immediately cooled to 4°C.

Probe capture, elution, and cleavage.After ligation, 30 μl distilled water was added to each reaction mixture and the samples were denatured at 95°C for 10 min, followed by rapid transfer onto ice. Capturing of the desthiobiotin moiety of the PLPs was performed in 80 μl of a capturing mixture containing 1 M NaCl, 5 mM Tris-HCl (pH 7.5), 0.5 mM EDTA, 0.1 M NaOH, and 20 μg magnetic MyOne streptavidin C1 Dynabeads (Dynal Biotech ASA) with rotation at 4°C for 1 h. Subsequently, samples were centrifuged at 2,000 × g for 10 s, the Dynabeads were collected and separated from the sample via application of a magnetic field, and the Dynabeads were washed with 100 μl of a washing solution containing 100 mM Tris-HCl (pH 7.5) and 150 mM NaCl. Consequently, gDNA and preamplicons were washed away. The Dynabeads were resuspended in 10 μl distilled water and incubated at 95°C for 10 min, allowing elution of the PLPs from the Dynabeads. Samples were then rapidly transferred onto ice, and the empty magnetic streptavidin beads were removed via application of a magnetic field, leaving the washed PLPs in the solution. Next, 10 μl of a cleavage mixture (10 U uracil-N-glycosylase [UNG; Applied Biosystems], 10 U endonuclease IV [New England BioLabs], 2× NEBuffer 3 [New England BioLabs], 2× bovine serum albumin) was added to each reaction mixture and the samples were incubated at 37°C for 1.5 h. Subsequently, 2 μl of 1.1 M NaOH was added and samples were incubated at 95°C for 10 min. Finally, 8 μl of 0.5 M Tris buffer (pH 6.8) was added to neutralize the solution.

MB assay.In order to determine the capture-and-release efficiency of the PLPs, a molecular beacon (MB) complementary to the PLP was designed with Visual OMP 6.0 software (DNA Software Inc.). Samples containing a dilution series of PLPs were captured and eluted as described above. Next, the fluorescence of 50 nM MB in each sample was monitored in a solution containing 1× PCR buffer with MgCl₂ (Roche Diagnostics) as a function of temperature. The measurements were carried out with an ABI Prism 7700 sequence detector (Applied Biosystems). After 15 min of equilibration at 80°C, the temperature of the samples was reduced from 80°C to 15°C in steps of 1°C lasting for 5 min each. Fluorescence was measured for the last 30 s of each step. The MB melting curves of three parallel samples were measured, and the results were averaged. At 30°C, the melting curves of all of the samples had reached the plateau phase, and the corresponding fluorescence levels were used for target input determination.

Microarray preparation and hybridization.The cZipCode oligonucleotides carrying a C₁₂ linker and a 5′ NH₂ group were synthesized and spotted onto Nexterion MPX-E16 epoxy-coated slides by Isogen B.V. according to the manufacturer's instructions (Schott Nexterion). Before hybridization, the arrays were washed and blocked according to the manufacturer's instructions. The cleaved ligation samples were heated for 10 min at 99°C and then cooled rapidly on ice. The array hybridization mixtures were made up of 15 μl of sample and 1 μl of 0.1 μM Cy5-labeled corner oligonucleotide in 1× tetramethylammonium chloride in a final volume of 48 μl. Sixteen-well silicon structures (Schott Nexterion) were attached to the arrays to create 16 separate subarray chambers. After adding 40 μl of the array hybridization mixture to each well, the chambers were sealed and the arrays were hybridized at 55°C overnight in high humidity. The array was then washed once at 55°C for 5 min in preheated 1× SSC (0.15 M NaCl plus 0.015 M sodium citrate)-0.1% sodium dodecyl sulfate and twice for an additional 1 min at room temperature in 0.1× SSC-0.1% sodium dodecyl sulfate and in TNT (0.1 M Tris-HCl [pH 7.6], 0.15 M NaCl, 0.05% Tween 20), respectively. Next, the array was incubated in blocking solution (0.1 M Tris-HCl [pH 7.5], 0.15 M NaCl, 0.5% blocking reagent [Perkin-Elmer]) for 10 min at high humidity at room temperature and then washed for 1 min in TNT. After the addition of 20 μl of staining solution (15 μg/ml streptavidin R-phycoerythrin [PE; Qiagen] in 20 μl of blocking solution) to each well, the array was incubated in the dark at room temperature for 15 min. The silicon structures were then removed, and the slides were washed three times for 1 min in TNT and twice for 1 min in 0.1× SSC, respectively. Finally, the slides were dried by spinning at 250 × g for 3 min.

Analysis of microarray data.Microarrays were analyzed with a confocal ScanArray 4000 laser scanning system (Packard GSI Lumonics) containing a GreNe 543-nm laser for PE and a HeNe 633-nm laser for Cy5 fluorescence measurement. Laser power was fixed at 70% for both lasers, while the photomultiplier tube power ranged from 50 to 70%, depending on the signal intensity. Fluorescence intensities were quantified by using QuantArray1 (Packard GSI Lumonics), and the mean signal minus the mean local background (mean PE-B) was used. The absolute signal intensity was defined as the mean PE-B minus the assay background. The assay background for each subarray was defined as the mean PE-B + 2 standard deviations (SDs) of the fluorescence of the unused cZipCodes. The cZipCode oligonucleotides were spotted in threefold triplicates (nine parallels) or twofold quadruplicates (eight parallels), depending on the array batch used. After exclusion of the outliers (mean PE-B ± 2 SDs), signals were averaged for the probes, and SDs were calculated.

Real-time PCR.Amplification of targets in the biological samples was monitored in real time with the 7500 Real-Time PCR system (Applied Biosystems). TaqMan PCR was performed with 1× TaqMan universal PCR Master Mix (Applied Biosystems) containing AmpliTaq Gold DNA polymerase, deoxynucleoside triphosphates with dUTP, passive reference 1, optimized buffer components, 0.12 μl UNG (Applied Biosystems), 10 ng DNA extract, 300 nM of each primer, and 100 nM of the corresponding TaqMan probe. SYBR green PCR was performed with 1× SYBR green PCR Master Mix (Applied Biosystems) containing SYBR green I dye, AmpliTaq Gold DNA polymerase, deoxynucleoside triphosphates with dUTP, passive reference 1, optimized buffer components, 0.12 μl UNG (Applied Biosystems), 10 ng DNA extract, and 300 nM of each primer. The reaction mixtures were subjected to initial incubation at 50°C for 2 min, followed by 10 min denaturation at 95°C and 40 cycles of 15 s at 95°C and 1 min at 60°C. Three different TaqMan probes targeting Phytophthora spp. (28), P. aphanidermatum (forward primer, 5′-CGTGAACCGTTGAAATCATG-3′; reverse primer, 5′-TACATCGGCAGACTACAATT-3′; TaqMan probe, 5′-FAM-TTCGTTCAGCCCTCCC-BHQ-3′ [locked nucleic acids are in bold]), and V. dahliae (forward primer, 5′-CCGGTCCATCAGTCTCTCTG-3′; reverse primer, 5′-ACTCCGATGCGAGCTGTAAC-3′; TaqMan probe 5′-FAM-CGGGTCCGCCACTGC-BHQ-3′ [locked nucleic acids are in bold]) and a SYBR green assay-based primer pair targeting F. oxysporum spp. (35) were used for comparative assay validation.

Probe design and optimization.The designed probing system was experimentally tested with nine PLPs engineered to detect several economically important plant pathogens at different taxonomic levels (Tables 1 and 2). In contrast to previously designed methods, the PLPs used in this assay contain a desthiobiotin moiety between the spacer sequences for reversible PLP capture and a polydeoxyuracil nucleotide sequence for probe cleavage.

This novel PLP-based LD system was optimized and validated in several steps. In order to determine the capture-and-release efficiency of desthiobiotin-labeled probes, a dilution series of a desthiobiotin-labeled probe containing an internal MB hybridization site was captured, washed, and released. Next, MB melting curves were determined with these samples to visualize the amount of remaining probe after the capture-and-release procedure. As a control, MB melting curves were also determined with a dilution series of desthiobiotin-labeled probes that had not undergone the capture-and-release procedure. These experiments showed no significant differences in fluorescent signal between the reversibly captured PLPs and the control treatments at the PLP mixture concentration used (P > 0.05) (Fig. 2).

• Open in new tab
• Download powerpoint

FIG. 2.

Capture-and-release efficiency of the LD system. Dilution series of PLPs were captured and released as described in Materials and Methods (n = 3 for each treatment), and then MB melting curves were constructed. □ and solid line, uncaptured control samples (r² = 0.995); Δ and dashed line, captured and released samples (r² = 0.993); *, applied PLP concentration in the developed LD assay (2.0 pmol PLP). For both treatments at all PLP concentrations, P > 0.05 (t test).

To test the cleavability of the probes, unligated PLPs were subjected to UNG and endonuclease IV treatment and subsequently hybridized on the array. Hybridization of uncleaved probes could be visualized by labeling the desthiobiotin moiety with a fluorescent label. The cleavage efficiency was highly sensitive to the position of the desthiobiotin moiety with respect to the polydeoxyuracil cleavage site. When placed in close proximity, the desthiobiotin moiety was found to interfere with the removal of uracil bases by UNG and endonuclease IV (Fig. 3). Therefore, oligo(dT) linkers of various lengths were introduced between the two modifications and tested for their effects on the efficiency of PLP cleavage. As shown in Fig. 3, a 15-dT linker sequence was required to provide quantitative (100%) cleavage of all of the probes. This design was adopted for all of the PLPs in subsequent experiments.

• Open in new tab
• Download powerpoint

FIG. 3.

Thymine (T) linker length-dependent probe cleavage. The percentage of cleaved probe was calculated relative to uncleaved control probe hybridizations. *, uncleaved control sample.

Assay optimization and characterization.To test the performance of the LD PLPs, the mixture of nine probes was ligated to various individual (Fig. 4A to C) and mixed (Fig. 4D to E) PCR amplicon targets. Without exception, targets were specifically detected and no signal was observed in the absence of target DNA (Fig. 4H).

• Open in new tab
• Download powerpoint

FIG. 4.

Detection of genomic DNAs corresponding to individual pathogen samples (A to C), mixed-pathogen samples (D and E), targets with closely related nontarget sequences (F and G), and a no-target control (H). The targets analyzed were as follows: (A) V. dahliae, 100 pg; (B) M. roridum, 100 pg; (C) F. oxysporum, 100 pg; (D) P. nicotianae, 100 pg, and V. dahliae, 100 pg; (E) V. dahliae, 100 pg, M. roridum, 100 pg, F. oxysporum, 100 pg, P. nicotianae, 100 pg, P. aphanidermatum, 100 pg, and R. solani AG 2-2, 100 pg; (F) P. undulatum, 10 ng gDNA in the preamplification reaction mixture (closest nontarget of Phytophthora species PLP; one nucleotide difference compared to perfect ligation sequence), and V. alboatrum, 10 ng gDNA in the preamplification reaction mixture (closest nontarget of the V. dahliae PLP; two nucleotide differences); (G) P. cactorum, 10 ng gDNA in the preamplification reaction mixture (closest nontarget of P. nicotianae PLP; seven nucleotide differences); (H) no-target control. Green fluorescent spots represent PE-labeled PLPs. Red fluorescent spots represent Cy5-labeled hybridization control oligonucleotides. Similar data were obtained with all of the other PLPs in the presence of the corresponding targets. For the array spotting pattern, see Fig. S2 in the supplemental material.

To ensure specificity for the targeted pathogens, three nontarget organisms containing similar ligation regions (1, 2, and 7 mismatches compared to the perfect ligation sequences) for the PLP mixture were tested. No detectable fluorescent signals were observed in the presence of P. undulatum (with one nucleotide difference, the most similar nontarget of the Phytophthora species PLPs) and V. alboatrum (with two nucleotide differences, the most similar nontarget of the V. dahliae PLPs), even when an initial amount of 10 ng nontarget gDNA was added to the preamplification PCR mixture (Fig. 4F). LD in the presence of P. cactorum (with seven nucleotide differences, the most similar nontarget of the P. nicotianae PLPs) resulted in fluorescent signals for the Phytophthora species PLP, while the P. nicotianae PLP signal was completely absent (Fig. 4G). Therefore, the asymmetric PLPs could discriminate targets with a single nucleotide difference.

Applying the LD assay to a 10-fold dilution series of a multiplex target gDNA mixture containing all targets in equal ratios resulted in a detection sensitivity of 1 pg gDNA for each target (data not shown). In biological samples, however, targets are often present in highly unbalanced concentrations. To test the maximum dynamic range of the LD assay, a dilution series of target gDNAs was used to spike a background of 10 ng of nontarget ITS1-ITS4 gDNA before the preamplification reaction (Table 3). As shown in Table 3, the maximum dynamic range of the LD assay was 1:10⁴ and the presence of high concentrations of competing ITS1-ITS4 gDNA in the preamplification reaction mixture did not influence the final target detection sensitivity of 1 pg gDNA.

IAC.It is often observed that DNA extracted from environmental samples contains PCR-inhibiting compounds (55), which may lead to false negatives. In order to monitor PCR inhibition during the preamplification reaction, an IAC was developed. This IAC consisted of a single-stranded oligonucleotide containing a random nonsense sequence flanked by ITS1-ITS4 primer regions, which was added to each preamplification reaction mixture. An IAC PLP containing ligation arms targeting the reverse complement of the IAC oligonucleotide was added to each ligation mixture. In the case of PCR inhibition, the complementary strand of the IAC oligonucleotide will not be generated. Therefore, the target of the IAC PLP will not be present during the ligation reaction and, consequently, no IAC PLP signal will be observed on the array.

IAC performance was tested under several different reaction conditions. A preamplification reaction of the IAC oligonucleotide resulted in the detection of an IAC PLP signal (see Fig. S3a in the supplemental material). In the absence of a polymerase in the preamplification reaction mixture, no IAC PLP signal was observed (see Fig. S3b in the supplemental material), indicating that the IAC PLP could not be circularized in the absence of the complementary strand of the IAC oligonucleotide. The presence of 10 ng competing V. dahliae target gDNA in the preamplification reaction mixture did not influence IAC PLP detection (see Fig. S3c in the supplemental material). Additionally, the detection sensitivity of 1 pg target gDNA in a 1:10⁴ ratio background DNA remained unaltered (see Fig. S3d in the supplemental material). To test the influence of PCR-inhibiting compounds on IAC PLP performance, 50 pg F. oxysporum gDNA was used to spike 10 ng of a DNA extract previously observed to fully inhibit PCR amplification. In the presence of the inhibiting DNA extract, no F. oxysporum and IAC PLP signals were detected on the array (see Fig. S3e in the supplemental material). Diluting the inhibiting DNA extract by a factor of 50 after spiking restored the detection of the IAC and F. oxysporum PLPs (see Fig. S3f in the supplemental material). Thus, in the presence of spiked target DNA in a background of nontarget DNA, the target is clearly detected when the IAC shows amplification as well. These results indicate that the IAC is a suitable tool to reduce false negatives caused by the presence of PCR-inhibiting compounds.

Assay validation.The robustness of the LD assay was demonstrated with biological samples recovered from recirculation water from several Dutch horticultural systems which had been inoculated previously with various plant pathogens. At least 4 weeks after pathogen inoculation, recirculation water samples were collected and filtered, DNA was extracted from the filters, and the resulting samples were subjected to the developed LD procedure. The data generated with the LD assay were compared with data obtained by TaqMan PCR detection for a selected range of targets for which TaqMan probes were available (Phytophthora spp., P. aphanidermatum, and V. dahliae) and data obtained by SYBR green-based PCR detection for F. oxysporum spp. (35).

As shown in Table 4, the detected PLP signals corresponded to samples spiked with pathogens, while the nonspiked control samples were negative. Moreover, all of the observed LD assay results were supported by the real-time PCR data. Positive real-time PCR data always corresponded to positive LD assay signals, and negative real-time PCR data always corresponded to negative LD assay signals. Thus, the newly developed multiplex LD procedure and real-time PCR showed complete congruence when applied to environmental samples typical of a practical application scenario.