Lei Cai, State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, West Bei Cheng Rd, Beijing, China. E-mail: email@example.com
Colletotrichum kahawae is a strongly aggressive pathogen causing coffee berry disease and is specific to Arabica coffee (Coffea arabica) in Africa. In this article, we developed a real-time PCR assay for the species-specific diagnosis of C. kahawae by designing the primers and a TaqMan probe derived from the single nucleotide polymorphism-rich region of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene.
Methods and Results
DNA markers from rDNA internal transcribed spacer, actin, β-tubulin and GAPDH genes of the ex-type culture of C. kahawae and 10 reference strains of Colletotrichum species were analysed for intra- and interspecific variations. The GAPDH gene was selected to develop a species-specific DNA marker. A TaqMan real-time PCR assay for species-specific detection of C. kahawae was developed, and its accuracy was tested against type strains of other phylogenetically closely related species in the C. gloeosporioides species complex, with the detection sensitivity of 80 fg μl−1 of genomic DNA.
This real-time PCR assay is highly specific and sensitive for the diagnosis of C. kahawae and can be applied in qualitative and quantitative tests.
Significance and Impact of the Study
This protocol allows for a rapid and sensitive detection of C. kahawae and will be useful in disease management and pest detection to prevent further spread of this pathogen.
Colletotrichum kahawae Bridge and Waller is an important pathogen causing coffee berry disease (CBD) in Arabica coffee (Coffea arabica) throughout the African continent (Waller et al. 1993). It is the only Colletotrichum species listed as a quarantine pathogen in China (Ministry of Agriculture of the People's Republic of China's 2007), as well as in the many other countries of Asia and Latin America (Flood and Waller 2001). Colletotrichum kahawae was formerly referred to as C. coffeanum (Nutman and Roberts 1960), which actually included four different subgroups (Gibbs 1969), and these groups were latter recognized as C. gloeosporioides, C. acutatum (Hindorf 1970) and C. kahawae (Waller et al. 1993). Thus, for a long time, C. kahawae was regarded as a synonym of C. gloeosporioides. Waller et al. (1993), however, described the causal agent of CBD as C. kahawae, a distinct species of Colletotrichum based on morphological, cultural and biochemical characters, and more recently, this was confirmed based on multi-locus phylogeny (Prihastuti et al. 2009).
Many cryptic species have recently been described in the C. gloeosporioides complex based on multilocus phylogeny in combination with phenotypic characters (Prihastuti et al. 2009; Phoulivong et al. 2010a,b). The ‘gloeosporioides’ complex presently comprises C. asianum, C. fructicola, C. horii, C. jasmini-sambac, C. kahawae, C. gloeosporioides and C. siamense (Waller et al. 1993; Cai et al. 2009; Prihastuti et al. 2009; Yang et al. 2009; Wikee et al. 2011), plus several species that are very closely related to C. kahawae, for example, C. cordylinicola, C. ignotum, C. musae and C. tropicale that were described or epitypified more recently (Phoulivong et al. 2010b; Rojas et al. 2010; Su et al. 2011). Morphological characters of species in the ‘gloeosporioides’ complex often overlap and are ambiguous (Cai et al. 2009; Hyde et al. 2009); thus, differentiating species remains a considerable challenge for taxonomists. To date, we are unaware of any specific, sensitive and rapid molecular tools available to detect important pathogenic Colletotrichum species such as C. kahawae. Even though some molecular detection methods for important Colletotrichum pathogens were established (Sreenivasprasad et al. 1996; Natalia et al. 2002; Nilsson et al. 2008), they failed to detect species among the ‘gloeosporioides’ complex because of their lack of phylogenetic information and detectable molecular sites in rDNA internal transcribed spacer (ITS) regions among species (Cai et al. 2009; Crouch et al. 2009). They are also time-consuming and need special expertise in phylogenetic analysis. Establishing efficient and rapid methods for the identification of Colletotrichum species is therefore essential in practical and applied science.
ITS has been chosen as the DNA barcode for the kingdom of fungi (Schoch et al. 2012), but it is also widely acknowledged that ITS does not give sufficient resolution in Colletotrichum (Crouch et al. 2009) to differentiate all species. ITS sequence data have been applied to resolve species of the ‘gloeosporioides’ complex (Sreenivasaprasad et al. 1993, 1996; Johnston and Jones 1997); however, resolution of the complex is not satisfactory often adding more confusion (Crouch et al. 2009). Thus, it is now recommended that multilocus phylogeny and a polyphasic approach are utilized in resolving species complexes (Cai et al. 2009). This study aims to identify a single nucleotide polymorphism (SNP)–rich region from ITS and partial genes of actin, β-tubulin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) that could be used in developing a real-time PCR protocol to detect the distinctive species C. kahawae in the ‘gloeosporioides’ complex, specifically and rapidly.
Real-time, or quantitative, polymerase chain reaction (real-time PCR or qPCR) is a powerful molecular tool and has been used in genotyping, quantitative gene expression analysis, drug target validation and pathogen detection (Mackay 2007). It offers a fast, accurate and culture-independent method for the detection of microbial plant pathogens (Schaad et al. 1999; Nicholson et al. 2003).
In this study, ITS, partial actin, partial β-tubulin and partial the GAPDH gene were selected as candidate markers for molecular detection. Intra- and interspecific variations of these genes of the species of C. gloeosporioides complex were analysed to assess feasibility of a DNA fragment as a potential barcode marker. Based on these results, we developed a real-time PCR assay for the species-specific diagnosis of C. kahawae by designing the primers and TaqMan probe derived from selected SNP-rich regions.
Materials and methods
The fungal isolates of C. kahawae were the ex-type cultures deposited in CABI, UK. Other reference taxa employed in this study include ex-type strains of C. asianum, C. cordylinicola, C. fructicola, C. gloeosporioides, C. horii, C. jasmini-sambac, C. musae, C. siamense and C. simmondsii from the culture collection centres worldwide (Table 1). A total of 92 sequences of the four candidate loci, full-length ITS, partial actin, β-tubulin and GAPDH of above strains were retrieved from the GenBank (Table 1).
Comparison of intra- and interspecific divergences
Sequences of each gene fragment of four candidate markers were aligned using ClustalX 1·81 (Thompson et al. 1994) and manually edited to adjust the aligned sequences by BioEdit 7·0 (Hall 1999). The aligned sequences were input into DNAstar 7·1·0 (Lasergene, Madison, WI, USA) to calculate the similarity matrices and then illustrate the intra- and interspecific variations of the candidate marker for each of the 10 investigated species in a visualization analysis tool, TaxonGap 2·4·1 (Slabbinck et al. 2008). Colletotrichum simmondsii was designated as outgroup in the analyses.
Genomic DNA extraction, Colletotrichum kahawae-specific primers and TaqMan probe design
Fungal strains were grown on PDA and incubated at 25°C for 7 days. Genomic DNA was extracted using a Biospin Fungus Genomic DNA Extraction Kit (Bio Flux, Tokyo, Japan) according to the manufacturer's protocol. Quality and quantity of DNA were estimated visually by staining with GelRed on 1% agarose gel electrophoresis. The genomic DNA was diluted to a final concentration of 20–30 ng μl−1 measured by BioSpec-nano (UV–VIS Spectrophotometers, Shimadzu, Japan) and used as the templates for real-time PCR.
The gene sequences from the 10 Colletotrichum species were analysed by software as outlined below. A multiple sequence alignment was made for these sequences using ClustalX 1·81 (Thompson et al. 1994) and manually edited by BioEdit 7·0 (Hall 1999). The alignment was analysed using the program Primer Express ver. 2·0 (Applied Biosystems, Foster City, CA, USA) for the divergent regions that were used in designing primers and probe that could specifically amplify C. kahawae. The primer–probe set of CK-GF (5′-CTGCATCTGGTAGACAAGAAGGT-3′), CK-GR (5′-AGAGCGAGTCAGTAAATGTGACAG-3′) and CK-GP (5′-CCCATGATTTCAA TTCACATCAAGTCAAG-3′) were designed. The optimal melting temperature and secondary structures of primer and probe sequences were predicted by Gene Runner (Hastings Software, Hastings on Hudson, NY, USA).
Real-time PCR assay
Amplification reactions were performed in 25 μl volumes containing 1 μl (20–30 ng) of the template DNA, 12·5 μl of 2 × SsoFast Probes Supermix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), 400 nmol l−1 of each primer, 200 nmol l−1 TaqMan probe (recommended by manufacturers) and ultrapure water. The TaqMan probe (CK-GP) was labelled at the 5′ end with fluorescent reporter dye 6-carboxy-fluorescein (FAM) and at the 3′ end with quencher dye 6-carboxy-tetramethylrhodamine (TAMRA) (Invitrogen Bio Inc., Carlsbad, CA, USA). As a negative control, ultrapure water was included in each PCR run. The Bio-Rad IQ5 Real-time PCR system (Bio-Rad Laboratories, Inc.) was used for amplification and fluorescence measurement at each temperature step and cycle during the reaction. The thermal cycling conditions for C. kahawae DNA template amplification include an initial denaturation at 95°C for 2 min, followed by 40 cycles of 95°C for 5 s and 60°C for 20 s. Data acquisition and analysis were conducted using Bio-Rad software ver. 2·1 (Bio-Rad Laboratories, Inc.), and the threshold cycle (Ct value) was calculated to indicate significant fluorescence signals rising above background during the exponential amplification cycles of the PCR amplification process.
To evaluate the specificity of the real-time PCR primers, the melting curves were acquired after each run to check primer–dimers. Optimal conditions for establishing the melting curve, which was performed without the probe, were slightly different to TaqMan probe PCR above, with the 12·5 μl of 2 × SsoFast EvaGreen Supermix (Bio-Rad Laboratories, Inc.) replacing the SsoFast Probes Supermix. The thermal cycling conditions of an initial denaturation at 98°C for 2 min followed by 40 cycles at 98°C for 5 s, 57°C for 10 s and followed by melting from 65 to 95°C with 0·5°C increments for 10 s (Bio-Rad Laboratories, Inc.).
Specificity and sensitivity of Colletotrichum kahawae real-time PCR assay
A 5-point standard curve was obtained by plotting the Ct value vs the logarithm of the concentration of each 10-fold dilution series of fungal genomic DNA from 0·02 to 200 ng in a 25-μl PCR mixture. A linear relationship between the Ct value and the log input template DNA amount was calculated. The real-time PCR efficiency and the correlation coefficient (R2) were also calculated using Bio-Rad software ver. 2·1 (Bio-Rad Laboratories, Inc.) from this standard curve.
For the specificity test, 10 species in the C. gloeosporioides complex, that is, C. asianum, C. cordylinicola, C. fragariae, C. fructicola, C. gloeosporioides, C. horii, C. jasmini-sambac, C. musae and C. siamense were selected as reference strains. The genomic DNA of above species was diluted to a final concentration of 20–30 ng μl−1 and used as the templates of real-time PCR assay (Table 2).
Table 2. List of Colletotrichum species and isolates used to evaluate primer and probe specificity for C. kahawae in the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene and real-time PCR assays
To test the sensitivity of real-time PCR assay for C. kahawae, a dilution series of genomic DNA of type strain of C. kahawae (IMI 319418) at concentrations from 8 fg μl−1 to 8 ng μl−1 were amplified by real-time PCR assay (Table 3).
Table 3. Sensitivity of detection of Colletotrichum kahawae (IMI319418) by real-time PCR assessed using serial dilutions of pure genomic DNA
The standard curve, specificity and sensitivity tests of the real-time PCR assay were made with three replicates. The negative controls (NTC) containing ultrapure water instead of DNA were included in each run.
Estimation of the candidate genetic markers
To select an appropriate or ideal DNA marker, four candidate gene markers of ITS, actin gene, β-tubulin gene and the GAPDH gene were chosen and intra- and interspecific variations of them were evaluated. It is clear that in GAPDH sequence (Fig. 1), the maximum intraspecific variations are smaller than the minimum interspecific variations for all the tested species (Fig. 1). In contrast, the intraspecific variations of an individual species exceeded the interspecific variations of some others in the actin and β-tubulin (TUB2) genes, and the similarity rates of the actin and β-tubulin (TUB2) genes among species in C. gloeosporioides complex were more than 95 and 98% (Fig. 1). There was almost no intra- and interspecific variation in ITS regions of these species of C. gloeosporioides complex (Fig. 1). Therefore, the GAPDH gene was selected as candidate marker for subsequence protocol development.
Design of Colletotrichum kahawae-specific primers and TaqMan probe
The GAPDH gene sequences derived from the ex-type strains were obtained from GenBank and aligned (Table 1). The alignment revealed several single-nucleotide variable regions. We selected three species-specific locations for forward primer (CK-GF), reverse primer (CK-GR) and TaqMan probe (CK-GP), which were conserved among C. kahawae and variable in other related species (Fig. 2). The primer–probe set specific for C. kahawae yield a 115-bp amplicon of the GAPDH gene (Fig. 2). The Primer-Blast analysis of the designed primers against GenBank database showed no homology with sequences of other organisms. The parameters such as percentage of GC content, absence of self-complimentarity in oligonucleotides and complimentarity between the primers and probe were analysed, and no such structures were found.
Specificity and sensitivity of real-time PCR assay for Colletotrichum kahawae
The melting curve of C. kahawae showed a single dissociation peak at 82·74°C (data not shown), indicating the high specificity of the PCR primers. The standard curve was linear over the tested range, and no signal was detected from the no template control (NTC). The linear correlation coefficient of the standard curve was R2 = 0·992 (slope = −3·324), with the PCR efficiency (E) of 99·8% based on the formula: y = −3·324 x + 27·64, in which the ‘x’ and ‘y’ represent the log amount of template DNA concentrations and their corresponding threshold cycle values (Ct), respectively.
The ex-type strains of closely related Colletotrichum species including different species in C. gloeosporioides complex were selected for real-time PCR to test the specificity of this assay (Table 2). The positive amplification in 2 C. kahawae ex-type isolates showed that primers and probe were highly specific to the target species, and no fluorescence signal exceeding threshold baseline was observed in all the ‘non-kahawae’ samples (Table 2). The detection limit of the real-time PCR assay was 80 fg μl−1 of genomic DNA in PCR mixture, with a Ct value of 38·17 ± 0·46 (Table 3).
Colletotrichum kahawae is the only Colletotrichum species listed under quarantine regulations in many countries (Flood and Waller 2001; Ministry of Agriculture of the People's Republic of China's 2007). Rapid and accurate detection of C. kahawae is essential for quarantine purposes, pathogen management and control (Waller et al. 1993). There was no effective molecular diagnosis protocol that provides a specific and sensitive probe for this severe plant pathogen. It is necessary to identify a SNP–rich region to enable rapid species-specific primer and probe designs. It is also now possible to design a species-specific primer for C. kahawae, as the taxonomy in C. gloeosporioides complex has recently undergone significant progress. The examples include three new species (C. asianum, C. siamense and C. fructicola) associated with coffee berry disease (Prihastuti et al. 2009) and a new species C. cordylinicola causing disease of Cordyline fruticosa (Phoulivong et al. 2010b). These species were established for well-defined phylogenetic lineages, and because of this, it is now possible to develop a molecular detection tool for C. kahawae.
An appropriate DNA barcode marker for species identification is judged by two important criteria, one is the suitable intra- and interspecific variations (Hebert et al. 2004) and the other is high success rate of PCR amplification and sequencing (Hollingsworth et al. 2009). In this study, we confirmed that the partial GAPDH is a good DNA marker of the C. gloeosporioides complex, against ITS, actin and β-tubulin gene. First, ITS has been routinely employed to explore the phylogenetic relationships among species of different fungal groups and has also been proposed as the universal DNA barcode for the kingdom of fungi (Schoch et al. 2012). In our study, however, we found that ITS has a high similarity of more than 98% among all tested species in the C. gloeosporioides complex (Fig. 1) and overlapping occurred between the intra- and interspecific variations of these species (Fig. 1). Second, there is little divergence in the actin and β-tubulin gene sequences among closely related species, which thus fail to provide sufficient single-nucleotide polymorphisms (SNPs) for design of species-specific primers and probes, and thus fail to effectively detect the specific species from the complex. The partial GAPDH gene, however, is an excellent candidate as a DNA marker for the C. gloeosporioides species complex tested. The reasons are (i) the gene possesses the appropriate intra- and interspecific variations (Fig. 1); (ii) the PCR amplification and sequencing of the GAPDH gene are relatively easy for Colletotrichum species.
A Blast search in NCBI database showed that the targeted GAPDH sequence of C. kahawae is highly similar to that of the Colletotrichum crassipes assigned strain CORCS3 isolated from the orchid Pleione bulbocodioides (acc. HM585379, Yang et al. 2011). No type specimen is available for Colletotrichum crassipes originally reported from diseased grape fruit of Vitis vinifera, and the literature shows considerable morphological confusion in regard of this epithet (von Arx 1970; Sutton 1980). We consider it plausible that the assignment of the strain CORCS3 as Colletotrichum crassipes represents a case of misidentification.
Real-time PCR for detection of CBD
A real-time TaqMan PCR assay was developed in this study to provide a specific, sensitive and robust technique to quantify biomass of C. kahawae. Assessment against strains in the C. gloeosporioides complex showed the TaqMan probe and primer are highly specific to C. kahawae (Table 2). The probe is highly sensitive, when compared to previously reported assays for other species (Kuan et al. 2011; Yadav et al. 2011). We are unaware of any previously molecular probes developed for C. kahawae-specific primers or TaqMan probe, and none of the currently used techniques to quantify C. kahawae-specific biomass reaches this level of sensitivity. The early and accurate diagnosis of diseased plants caused by C. kahawae would facilitate pathogen identification and lead to more effective control of pre- and postharvest diseases.
Molecular diagnosis of cryptic species
With the application of the phylogenetic species recognition criteria (Taylor et al. 2000; Cai et al. 2011), there has been a rapid increase in the number of cryptic species of plant pathogenic fungi discovered in recent years. Many well-known, traditionally morphologically defined important plant pathogenic fungi have recently been revealed to be species complex containing many phylogenetically and physiologically divergent species (Damm et al. 2009; Hyde et al. 2009; Kvas et al. 2009; Shivas and Tan 2009; Wulandari et al. 2009; Zhang et al. 2009; Summerell et al. 2010). For these recently described cryptic species, it has been critical for mycologists and plant pathologists to determine the host range, the severity of diseases they cause and their biosecurity significance. With rapidly expanding global trade, it has also become imperative that we develop effective and reliable protocols to detect these previously unrecognized pathogens.
The authors thank Dr Yongjie Zhang, from Shanxi University, for the technical help of the primer and probe designs and Dr Zhaoqing Zeng, from Institute of Microbiology, CAS, for the technical assistance of the analysis of intra- and interspecific variations. This study was supported by the National Natural Science Foundation of China (No. 31070020), the Knowledge Innovation Program of the CAS (KSCX2-EW-J-6/KSCX2-YW-Z-1026) and the China Postdoctoral Science Foundation (No. 2011M500421). K.D. Hyde thanks King Saud University for a visiting Professorship and National Plan of Science and Technology, King Abdulaziz City of Science and Technology, Riyadh, Saudi Arabia (10-Bio-965-02).