Journal of Plant Pathology & Microbiology

The genus Fusarium comprises a high number of well-known fungal pathogens with significant economic impact on plants causing diseases in several agriculturally important crops, including coffee [1], as well as humans and animals [2], Fusarium xylarioides is a typical vascular pathogen known to cause coffee wilt disease (CWD), commonly referred to as tracheomycosis or sometimes carbunculariosis. The sexual (perfect) form is Gibberella xylarioides while F. xylarioides is the conidial (asexual) state of this species [3]. In Uganda and the rest of East and Central Africa, the pathogen causes CWD exclusively on Robusta coffee unlike in Ethiopia where the pathogen infects Arabica coffee and no cross-infection occurs in Robusta coffee with pathogen from Arabica coffee and vice–versa. The disease is endemic to Robusta coffee in Uganda yet this provides over 78% of the coffee exported [4]. The disease advent in Uganda was in 1993 and destroyed 44.5% of Robusta coffee (Coffea canephora) by 2002 [5], leading to economic losses of over $100M. This had a negative impact on the country’s GDP and livelihoods of over 7 million (21%) Ugandans/ 1.7 (42%) million out of the estimated Uganda 42 million agricultural households who depend on coffee growing or employed along the value chain [4]. The disease is still causing severe damage on coffee and of late been reported in Kabale district in Uganda, a situation which could be attributed to effect of climate change.

Detection of CWD pathogen from disease plant tissue has been mainly done microscopically after culturing and has since failed to detect the pathogen from the soil and asymptomatic plants [6], it is expensive, slow, and requires high taxonomic expertise [7], On the other hand, field identification basing on symptoms is confusing with other effects like stem borer, drought and other physiological problems, which produce similar symptoms to CWD [6]. A fast and reliable method for detection and identification of the pathogen is therefore needed. Here, we report the first PCR detection method developed specifically for F. xylarioides that will offer a fast, sensitive and versatile detection and diagnostic procedure with ability of handling bulk specimens over the conventional method [8,9]. The tool will promote timely and correct mitigation measures; epidemiological, agronomic and breeding for resistance research; quarantine check and support seed inspection and will eventually impact on the socio-economic transformation.

The first stage in development of a molecular diagnostic assay is to select the nucleic acid sequences to be used to identify the organism. This can be done by (i) targeting particular genus specific gene but which have useful sequence variation at species level (ii) in silico using sequences in gene banks and (iii) randomly selected DNA fragment amplified with arbitrary primers [10,11]. The principle is that every species has a unique DNA sequence that can be used for its specific identification. In this study, we have used the first approach to develop the specific primer to F. xylarioides.

The work therefore utilized TEF-1α gene sequence to develop the first specific, fast and robust PCR based detection and identification protocol for coffee wilt disease pathogen and disease respectively for improved Robusta coffee plant health and eventually translating into increased coffee production for socio-economic transformation of the people of Uganda. The motivation for using the TEF-1α gene was that it is considered a housekeeping gene with higher sequence polymorphism and are being more extensively used to design species specific markers and probes for the identification, quantification of pathogenic population of Fusarium species [12-17].

Fungal species used in the study

Eleven fungal species including F. xylarioides isolated from infected Robusta coffee tree and ten other different fungal species namely: F. solani, F. oxysporum, F. decemcellulare, F. moniliforme, F. solani, F. lateritium, F. pallidoroseum, Sclerotia spp., Ambrosia spp. and Rhizoctobia solani which were identified culturally and microscopically as described by Booth, et al. [18-20] and available in the culture collection at the National Coffee Research Institute, Kituza, Mukono, Uganda. Apart from Sclerotia spp. and R. solani which originated from infected bean roots, the rest were isolated from Robusta coffee trees. Positive and negative reference gold standards were acquired from CABI Biosciences England and these were F. xylarioides (IMINo: 379925) and F. oxysporum (IMI No: 244509) respectively. Fusarium oxysporum was chosen as a negative standard because it tested positive with all the primers that were discarded for lack of specificity to F. xylarioides.

Reparation of fungal mycelia for DNA extraction

Preparation of cultures for DNA extraction was carried out as described by Olal, et al 2014. [6]. The stored cultures were sub-cultured on fresh Potato dextrose agar (PDA) medium (OXOID, CMO 139, England) and allowed to grow for 4 days under 12-hour fluorescence light and dark cycles at room temperature. Potato dextrose broth (PDB) comprised 200 g washed unpeeled Irish potato, 20 g dextrose/D-glucose powder (BDH, England) and 1L tap water was autoclaved for 20 min at 121OC at 15 Psi, cooled in the laminar flow hood, suspension allowed to settle and 10 mL suspended in 22 individual 9 cm Petri dishes for mycelia production. Four (4 × 4 mm) plugs taken from the periphery of each of the 11 rejuvenated fungal isolates on PDA were suspended in duplicate in PDB plates. The inoculated PDB plates were incubated in the dark without shaking at room temperature (25 ± 2oC) for 4 days to encourage mycelia mat formation while limiting sporulation. The mycelial mats from each fungal species were put in individual tea strainers, rinsed thoroughly with sterile distilled water and blotted dry with sterile absorbent papers ready for DNA extraction.

DNA extraction

To amplify the F. xylarioides TEF-1α gene with Fusarium genus specific TEF-1α primers for sequencing and to test the specificity of the designed primers against the 10 test Fungal species, DNA was extracted from each of the 11 isolates using ZR Plant/Seed DNA MiniPrep (ZYMO RESEARCH, USA) according to the manufacturer’s instructions. The purity and quantity of the DNA extracted was verified using Nanodrop spectrophotometer (Nanodrop 2000C, Thermo Scientific). The purity of the DNA was found to lie between absorbent readings at wavelength (λ) 260/280 ratio ranging from 1.8 - 2.0 which was quite substantial quality.

Amplification of translation elongation factor (TEF-1α) gene

The TEF-1α gene from F. xylarioides isolates from Robusta coffee were amplified with TEF-1αgene specific primer pair, TEF-Fu3f: 5’GGTATCGACAAGCGAACCAT3’, TEF-Fu3r: 5’TAGTAGCGGGGAGTCTCGAA3’ [13]. The PCR was performed in a 40 μl reaction mixture containing 4 μl of 50 ng/μl of genomic DNA, 20 μl of 2x Green GoTag PCR master mix (Promega), 1.2 μl of 10 mM of each primer (forward and reverse), and 1.6 μl ddi sterile water. The amplification conditions were as follows; initial denaturation at 95°C for 8 min, followed by thirty five (35) cycles of 95°C for 15 sec, 53°C for 20 sec, 72°C for 1 min and final extension at 72°C for 5 min (Eppendorf AG 22331 Hamburg, England). Amplification product was loaded in four (4) wells of 10 μl each and separated by electrophoresis (Sub-cell model 96. BIO-RAD) on a 1.4% agarose gel in TAE buffer. The gel was then stained for 15 min in 0.5 μg/ml solution of ethidium bromide (BIO-RAD, USA) and gel photograph was taken using gel documentation system (SyngeneG:Box). The characteristic bands from all the four wells were then excised from the gel, purified with GeneJET PCR Purification Kit (Thermo SCIENTIFIC, Lithuania) according to the manufacturer’s instruction and sequenced directly from Inqaba Lab. LTD, Republic of South Africa.

Primer design and specificity test

Primer-BLAST was used to design and test the specificity of the primers in silico [21]. The primers were selected for syntheses based on similarity in melting temperatures (Tm). The designed primers were synthesized from Ingaba Lab, Republic of South Africa. The specificity of the primers was predetermined according to the Primer-BLAST output and confirmed in-vitro by polymerase chain reaction (PCR). One critical property observed when selecting the primer was that it should only amplify the intended target.

Primer specificity testing in-vitro by PCR

Specificity of the synthesized primers was tested by PCR of the genomic DNA from all the fungal isolates used in the study namely: F. xylarioides, F. solani, F. oxysporum, F. decemcellulare, F. moniliforme, F. lateritium, F. pallidoroseum, Sclerotia spp., Ambrosia spp. and R. solani. The PCR was performed in a 20 μl reaction mixture containing 1 μl of 50 ng/μl of genomic DNA, 10 μl of 2x One tag Hot Start PCR master mix (New England), 0.3 μl of 10 mMof each primer (forward and reverse), and 8.4 μl ddi sterile water. The amplification conditions were as follows: Initial denaturation at 94°C for 4 min, followed by 30 cycles of 94°C for 1 min, 58°C 1 min, and 68°C for 1 min and final extension at 68°C for 5 min (Eppendorf AG 22331 Hamburg, England). Amplification products were separated by electrophoresis (Sub-cell model 96, BIO-RAD) on a 1.4% agarose gel in TAE (Buffer containing Tris base, acetic acid and EDTA) buffer. The gel was then stained for 15 min in 0.5 μg/ml solution of ethidium bromide (BIO-RAD, USA) and gel photograph was taken using gel documentation system (Syngene G: Box).

Sensitivity of the primers

This was done to determine the minimum amount of DNA which can be amplified by the primer. 1 μL containing 50 ng/μL of DNA was serially diluted in a series of seven concentrations ranging from 0.001 ng/mL to 50 ng/μL and a negative control. The same reaction mixture and cycling processes described above for DNA amplification were used for the sensitivity tests.

The diagnostic potential of the primers

This was done to find the robustness of the primers to be used in any pathogen strain across the country. The primers were tested on DNA isolated from the pathogen originating from different regions in the country.

Phylogenetic analysis

BLAST [22] was used to search for similar sequences comparing Fusarium xylarioides sequence used in primer design with those in the Gen Bank database and local sequence databases, the cut off for score were 952 and E-value 0.0). Multiple sequence alignment of the identified sequences from Blast analysis was performed using Clastal X/W version 2.0 [23]. Phylogenetic analysis of the aligned sequences was performed by Maximum likelihood (ML) using the MEGA 6 phylogenetic package [24].

in silico primer specificity test

in silico test was a pre-specificity test and showed that the primer matched only F. xylarioides TEF-1α gene sequences available in the gene banks. This is a positive indication that the primer is not likely to amplify TEF-1α gene of any organism whose TEF-1α gene sequence is currently available in the gene banks, yet F. xylarioides is one of the least organism that has been studied at molecular level [25]. It is also possible that the primers can detect F. xylarioides race pathogenic to Arabica since 25% of the sequences matched belong to the race pathogenic to Arabica coffee. The result also indicates possibility of using the primer regionally since the pathogen sequences from all the country where currently CWD exist namely Tanzania, Uganda, the DRC and Ethiopia were matched (Table 1).

Table 1: Insilico specificity test of the primer OSTF15.

In vitro specificity test using PCR

The PCR reaction was a confirmatory test for the specificity of the primer. The primer was able to amplify only DNA from F. xylarioides and not from any test species The specific primer to F. xylarioides is therefore OSTF15 F (5’ GACCTGGCGGGGTATTTCTC) and OSTF15R (5’AATGGGAGAGGGCAGAAACG ) (Figure 1).

Figure 1: In vitro specificity test of primer SOTF15(284bp) by PCR. Lane 1,100bp DNA Ladder; Lane 2, Negative control (H20); Lane 3, F. xylarioides (Used in primer development;Lane 4, F. xylarioides(Gold standard);Lane 5,F.oxysporum (Gold standard negative control);Lane6,F. solani; Lane7,F. decemcellulare and;Lane 8, F.lateritium.

Diagnostic potential of primer OSTF15

This was done to test the robustness of the primer to detect isolates from different parts of the country. The result showed that the primer was able to amplify DNA from all the F. xylarioides isolates from different part of the country tested. This means there is no variation in the pathogen across the country at TEF-1αgene level (Figure 2).

Figure 2: Diagnostic potential of primer OSTF15. Lane 1, 100bp DNA ladder; Lane 2,Central (Kituza-Mukono; Lane 3,Western (Bundibidgyo); Lane 4, South western (Rukungiri); Lane 5, West Nile (Arua); Lane 6,Kalangala; Lane 7,Eastern (Iganga) and; Lane 8,Kabale. The primers were able to amplify the DNA from all isolates obtained from different part of the country (Uganda).

In vitro sensitivity test

This was done to find the minutest amount of DNA which can be amplified by the primer. The result showed that the minimum amount of DNA amplifiable by the primer was as minute as 0.78 ng/μL (Figure 3) and this is vital for the detection of the minute amount of the pathogen DNA available in infected asymptomatic plants and soil (Figure 3).

Figure 3: Sensitivity of the OSTF15.Lane 1, 100 bp DNA Ladder; Lane 2, Negative control (H₂0); Lane 3, 50 ng/μL; Lane 4, 12.5 ng/μL, Lane 5; 3.124 ng/μL; Lane 6, 0.78 ng/μL, Lane7, 0.2 ng/μL; Lane8, 0.05 ng/μL. The sensitivity stopped at DNA concentration of 0.78 ng/μL. Serial dilution of 50 ng/μL by 4 folds was used progressively in ascending well numbers. Bands were separated on 1.4% agarose gel.

Phylogenetic analysis

The sequences use to design the specific primer was blasted and the selected species from the blast product was used to generate the tree. It can be clearly seen that F. xylarioides was grouped separately from other species and hence the specific primers may not amplify DNA from any of those related species used in the analysis (Figure 4).

Figure 4: Phylogenetic relationship of F. xylarioided and other related specie generated by the blast output. The sequences were aligned using Clastal X and tree created using Mega 6 software package. The Numbers below branches represent bootstrap values based on 1000 replicates.

This study represents the first application of PCR based detection and diagnostic test for F. xylarioides using primers amplifying 284bp region of the translation elongation factor gene (TEF-1α). This F. xylarioides-specific amplification product differentiated F. xylarioides from all other Fungal species used in the study shown above. The primer was also able to amplify DNA from F. xylarioides isolates from different part of the country meaning it can be used robustly on all the strain within the country. The F. xylarioides-specific primers will thus allow rapid and simple diagnoses of pathogen in diverse habitats. The output of the primer-BLAST showed that the primer sequence matched only F. xylarioides (100%), 75% from the race pathogenic to Robusta coffee and 25% to Arabica coffee, a clear evidence of the race differences as had been reported by other workers that there exist two host-race specificity in F. xylarioides one for Coffea canephora and one for Coffeaarabica [26,27]. The primers were able to amplify DNA from all isolates tested from different region (Figure 2) meaning the primers can be used with high reliability to detect the pathogen from any part of the country; this further emphasized uniformity in the CWD pathogen as suggested by Lewis, et al. [28]. The primers were also found to be quite sensitive amplifying DNA concentrations as low as 0.78 ng/μL of DNA. This fast, specific and sensitive contemporary method for detection and identification of F. xylarioides will reinforce the conventional method that has been in use before. Since the primer sequence matched also the race pathogenic to Arabica coffee, it is possible that the primers can be used to detect the pathogen from Arabica coffee as well.

Translation elongation factor 1α (TEF-1α) gene was used to design the primers because it is considered a housekeeping gene with higher sequence polymorphism and are being more extensively used to design species specific markers and probes for the identification, quantification of pathogenic population of Fusarium species [12-16].

Evolutionary history of an organism is recorded in its genes and the development of sequence analysis techniques has allowed reconstruct biological evolution by comparing the structure of genes or geneproducts in different species [29]. In this study, phylogenetic analyses were performed using BLAST results of the nuclear (translation elongation factor 1-alpha, EF-1α) sequence used to design the primers (Figure 4). The analysis categorized the organisms into three clusters. The upper cluster consisted of mainly fungus that infects leaves, the middle cluster comprised mainly F. oxysporum and the lower cluster comprised F. xylarioides strains in the lower sub-cluster and those of F. odum, F. indica and F. phyllophilum in the upper sub-cluster. The two sub-clusters in the lower cluster are very close relatives. All the species in the upper sub-cluster belong to the G. fujikuroi species complex (GFC) [30]. The GFC is a monophyletic and diverse group of several phylogenetic species divided into three sub-clades, often referred to as African, American and Asian, based on the putative geographic origin of most of the species within them (Geiser, 2005)[31]. Some authors had grouped F. xylarioides as belonging to the African clades of the G. fujikuroi species complex (Rutheford, 2006) while others have refuted it suggesting that recent outbreaks of vascular wilt in Uganda may have been caused by a new pathogen [28] Clades are characterized by shared possession of uniquely-derived evolutionary novelties or “synapomorphies” (literally “together, derived shape”). The so-called “African Clade” is the largest of the three clades with 23 phylogenetic lineages, of which four represent biological species (Capable of undergoing sexual cycle) namely: F. verticillioides, F. thapsinum, F. nygamai and F. xylarioides [32]. The agriculturally important pathogenic ones include F. verticillioides, F. denticulatum, F. thapsinum, F. nygamai, F. lactis, F. phyllophilum, F. udumand, and F. xylarioides (O’Donnell, 1998) [30]. The close relationship between F. phyllophilum, F. udum and F. xylarioides was demonstrated by Lima,et al. [33]. One of the attribute of GCF membership is the presence of chlamydospore [29] of which F. xylarioides does not have and this could be one of the reason why it is put in a different lower sub-cluster. The analysis shows that the members in the two sub-clusters are very close relatives supported at a high bootstrap value of 846 / 84.6% (Figure 4), but since they have not been clustered together, there may be no false positive reaction.

The salient outcome of this study is the presentation of the first report on the application of PCR for sensitive and specific detection of F. xylarioides, the cause of CWD. This will be a great opening for research areas that were not possible before, including alternative host, effect of agronomic studies on the pathogen load, seed system certification and quarantine check on materials. This study contributes to impact on production and improvement in livelihoods and socioeconomic development in people involved in coffee production value chain. The primers can be robustly used since in silico test indicated the ability to detect the Arabica coffee pathogen strain as well. These primers can only detect the presence of the pathogen so there is need to develop quantitative PCR (qPCR) as well to quantify the pathogen load in soil and plant tissues.

The research was sponsored by the National Agricultural Research Organization (NARO) under Agricultural Training and advisory services (ATAAS) project funded by the World Bank of grant number 4769UG. Appreciation extended to COVAB, Makerere University and NARL for availing laboratory facilities for the research.