Use of a real-time LAMP isothermal assay for detecting 16SrII and XII phytoplasmas in fruit and weeds of the Ethiopian Rift Valley



This article is corrected by:

  1. Errata: Corrigendum: Use of a real-time LAMP isothermal assay for detecting 16SrII and XII phytoplasmas in fruit and weeds of the Ethiopian Rift Valley Volume 60, Issue 6, 1183, Article first published online: 1 November 2011



A study to investigate the association of phytoplasmas with papaya dieback and citrus decline syndromes in Ethiopia was carried out between July 2009 and February 2010, with sampling performed in major papaya- and citrus-growing areas of the Rift Valley. Samples of plants with symptoms were collected from papaya, citrus and suspected phytoplasma weed hosts and crops in and around the papaya and citrus fields studied. Nested polymerase chain reaction (nested-PCR) was used for initial characterization, using primers that amplify regions of the 16S rRNA and secA genes, and results were then confirmed with rapid real-time group-specific LAMP (loop-mediated isothermal amplification) assays. The results identified the occurrence of a phytoplasma belonging to the stolbur (16SrXII-A) group in papaya plants showing dieback symptoms, whilst no phytoplasmas were found associated with citrus decline. These results contradict previous reports that a 16SrII phytoplasma was associated with both papaya dieback and citrus decline in Ethiopia, but correspond with the association of a 16SrXII phytoplasmas with Nivum-Haamir-Dieback of papaya in Israel and papaya dieback in Australia. No 16SrXII phytoplasmas were found in any of the weeds and potential alternative hosts studied, although 16SrII phytoplasmas were consistently found in Parthenium hysterophorus weed plants. These results indicate that a 16SrXII phytoplasma is associated with papaya dieback in Ethiopia, whilst the causal agent of citrus decline is not a phytoplasma and remains unidentified.


Phytoplasma diseases have significant impact on yields of important crops (Lee & Davis, 1992), and the association of phytoplasmas with a wide range of diseases in several economic field crops, vegetables, fruit crops, ornamental plants, and timber and shade trees has been reported worldwide and reviewed by Lee et al. (2000). Many newly emerging diseases and strains have been identified and reported (Hogenhout et al., 2008), and the list of diseases caused by phytoplasmas continues to grow, helped by improved molecular diagnostic techniques such as nested PCR with new primers (Hodgetts et al., 2007, 2008), real-time PCR (Christensen et al., 2004; Hodgetts et al., 2009) and loop-mediated isothermal amplification (LAMP) assays (Tomlinson et al., 2010a) that can detect low titres of phytoplasmas from plant tissues.

No extensive study has been conducted on phytoplasma diseases affecting agricultural crops in Ethiopia, except two reports on the association of phytoplasma phyllody disease with the weed Parthenium hysterophorus (Taye et al., 2004; Janke et al., 2007). Absence of recognition may partly be the result of lack of awareness, along with the technical difficulties of detecting these unculturable organisms, such as the lack of molecular diagnostic capability in many research laboratories in Ethiopia. Papaya dieback syndrome was first noted during the early 1990s because of the level of damage, and an attempt to understand the association of phytoplasmas and viruses with papaya dieback and citrus decline was initiated in 2005. This involved field visits, and papaya and citrus samples with symptoms were collected from farms in the Melkassa area of East Shewa for laboratory analysis. Results indicated the presence of phytoplasma in diseased papaya, and Citrus tristeza virus (CTV) and phytoplasma in citrus trees (Arocha et al., 2007). The phytoplasma detected in 11 out of 12 papaya and six out of six citrus plants was identified as a 16SrII-B ‘Candidatus Phytoplasma aurantifolia’ isolate, related to the causal agent of papaya yellow crinkle and papaya mosaic in Australia. The Australian studies (White et al., 1998) had examined three papaya plants, one suffering from papaya dieback, one from yellow crinkle and one from mosaic, and whilst the yellow crinkle and mosaic plants had 16SrII phytoplasmas, the papaya dieback plant contained a 16SrXII phytoplasma. In Israel, a 16SrXII phytoplasma was found associated with Nivum-Haamir-dieback of papaya (Gera et al., 2005), whilst a 16SrII phytoplasma was associated with bunchy top symptoms of papaya in Cuba (Pérez et al., 2010). In Malaysia, no phytoplasma was found associated with papaya dieback, and the causal agent was identified as Erwinia papayae (Maktar et al., 2008). These confusing reports have prompted the present, more extensive, investigation of phytoplasmas affecting papaya and citrus in Ethiopia.

Papaya (Carica papaya) and orange (Citrus cinensis) are amongst the most common fruits grown in Ethiopia and the area devoted to their cultivation has increased considerably in the last six decades (Seifu, 2003). Papaya dieback has been observed distributed over large areas, extending several hundred kilometres from northeast to south along the Great East African Rift Valley crossing Ethiopia, which is the major papaya- and citrus-growing belt in the country. Papaya dieback causes total plant failure at the late stage of infection and some farmers are abandoning papaya fields and replacing it with other less profitable crops, such as sugarcane. The first characteristic symptom of dieback in papaya is a bright yellowing of the upper youngest leaves, followed by mosaic, crinkling and leaf-tip necrosis. At the later stages, symptoms are accompanied by drying of the upper leaves, which progresses downwards, ending with death of the whole plant. Symptoms of citrus decline include leaf interveinal chlorosis, mosaic or mottling, and a reduction in the size and curling of leaves. In severe cases, premature fruit fall, twig and branch dieback and finally death of trees occurs.

In a preliminary study carried out by Arocha et al. (2007), only a few farms in one location were studied, and representative locations were not systematically covered. In addition, suspected alternative hosts for papaya dieback disease and other phytoplasmas were not studied. This paper presents the results of a comprehensive study aimed at assessing the distribution and importance of papaya dieback and other phytoplasmas in papaya, citrus and suspected alternative hosts in fruit production systems of the Great Rift Valley in Ethiopia. It also details the development of a real-time LAMP isothermal assay that is much quicker than PCR and less prone to inhibition from DNA preparations.

Materials and methods

Phytoplasma samples and DNA extraction

Large- and small-scale papaya and citrus farms were visited for sampling along the Great Rift Valley in Ethiopia (Fig. 1). Grass and non-grass weeds and crop plants showing symptoms such as yellowing, stunting, phyllody, virescence, mosaics and leaf curling were collected in and around papaya and citrus fields for phytoplasma testing. Sample collection was restricted to plants showing typical phytoplasma-like symptoms, since the aim of the study was to type the phytoplasmas present in papaya and citrus, and in putative alternative weed hosts growing in papaya and citrus fields. Parts of the collected leaf samples to be used for DNA extraction were dried using calcium chloride in vials, and the remainder of each sample was pressed using a herbarium press for later use in the identification of hosts to species level and as future reference material. The identities and sources of the samples are summarized in Table 1. Data collected in the field included altitude, geographical position using GPS, crop and crop variables, weed species, collection site and symptom description.

Figure 1.

 Map of Ethiopia showing phytoplasma survey areas (within dotted lines).

Table 1.   Survey locations, hosts and phytoplasmas identified in plants growing in the Rift Valley in Ethiopia, 2009
Zone/LocationAltitude (m a.s.l.)HostSymptomsNumber of samples in categoryaPositive samples with primersbResults with LAMPsecA sequence16Sr groupc
  1. aNumber of samples tested from this site with these symptoms and these PCR/LAMP results.

  2. bRefers to pairs of primers used for nested- PCR. SecA = SecAfor1/SecArev3 & SecAfor5/SecArev2; 16S rRNA = P1/P7 & R16F2n/R16R2.

  3. c16Sr subgroups are given if the PCR products were sequenced. If the results are based on the Lamp assay, just the 16Sr group is given.

  4. NT, not tested.

Arsi/Awash Bishola1537Carica papayaBright yellowing causing dieback3+NT+XII-A16SrXII-A
Arsi/Awash Bishola1537C. papayaYellowing/dieback1
Arsi/Tibila1336C. papayaYellowing (dieback tree)2
Arsi/Tibila1336C. papayaYellowing1
Arsi/Tibila1336C. papayaYellowing1+16SrXII
East Shewa/Degaga1287C. papayaYellowing of younger leaves2+NT+XII-A16SrXII-A
East Shewa/Degaga1287C. papayaYellowing leading to dieback4+NT+XII-A16SrXII-A
East Shewa/Degaga1287C. papayaLeaf distortion/malformation1+16SrXII
East Shewa/Degaga1287C. papayaLeaf distortion/malformation4+NT+XII-A16SrXII-A
East Shewa/Degaga1287C. papayaDieback/yellowing2+NT+XII-A16SrXII-A
East Shewa/MARC1545C. papayaYellowing leading to dieback1+NT+XII-A16SrXII-A
East Shewa/MARC1545C. papayaYellowing leading to dieback4
East Shewa/MARC1545C. papayaYellowing/pale green1
East Shewa/MARC1545C. papayaYellowing/stunted growth2
East Shewa/MARC1545C. papayaDieback/bright yellowing1+16SrXII
East Shewa/Ethioflora1630C. papayaBright yellowing of leaf tips2
East Shewa/Ethioflora1630C. papayaLeaf yellowing/narrow leaves1
East Shewa/Ethioflora1630C. papayaYellowing leading to dieback1
East Shewa/ZFP1650C. papayaYellowing of whole leaf1
East Shewa/ZFP1650C. papayaMosaic/small leaves1
East Shewa/ZFP1650C. papayaPale yellowing of leaves1
East Shewa/Meki1695C. papayaLeaf yellowing/stunting1
East Shewa/Meki1695C. papayaLeaf yellowing/stunting1+NT+XII-A16SrXII-A
West Shewa/Guder2020C. papayaYellowing of leaf tips1
West Shewa/Guder2020C. papayaMottling/mosaic2
East Hararghe/Babile1640C. papayaLeaf pale yellowing1+NT+XII-A16SrXII-A
Dire Dawa/Tony1150C. papayaPale yellowing/small leaves1
West Hararghe/Chiro1930C. papayaYellowing/dieback1
West Hararghe/Chiro1930C. papayaYellowing at leaf margins1
Arsi/Awash Bishola1537Citrus sp.Yellowing/mosaic/mottling1
Arsi/Tibila1336Citrus reticulataInterveinal mosaic1
Arsi/Tibila1336Citrus sinensisStunting/small & pale green fruit1+BacillusBacillus
Arsi/Tibila1336C. reticulataYellowing of specific branch1
East Shewa/Nura Era-11118C. sinensisInterveinal oranging1
East Shewa/Nura Era-21118C. sinensisTwig dieback/yellowing1
East Shewa/Nura Era-21118C. sinensisBranch drying/twig dieback1+BacillusBacillus
East Shewa/MARC1545C. reticulataBranch dying/interveinal mosaic1
East Shewa/MARC1545C. reticulataInterveinal mosaic/yellowing1
East Shewa/MARC1545C. sinensisPale yellowing1
West Shewa/Guder2020C. sinensisPoor stand/pale yellowing1
West Shewa/Guder2020C. sinensisPoor stand/twig dieback1
West Shewa/PPRC?Citrus sp.Distorted leaves1
Shinile/Errer Gota1175Citrus aurantiumPale yellow leaf colour1
Shinile/Errer Gota1175C. sinensisInterveinal mosaic/yellowing1
Arsi/Awash Bishola1537Coffea arabicaInterveinal mosaic/mottling1NTNT
East Shewa/Nura Era-11118Parthenium hysterophorusPhyllody/yellowing/stunting1++II-B16SrII-B
East Shewa/Nura Era-11118Solanum nigrumLeaf curling/mild mosaic1NTNT
East Shewa/Degaga1287P. hysterophorusYellowing1++II-B16SrII-B
East Shewa/Degaga1287Solanaceous sp.Mottling/mosaic1NTNT
East Shewa/Degaga1287P. hysterophorusYellowing/mottling1NTNT
East Shewa/Degaga1287Commelina sp.Mosaic3NTNT
Arsi/Tibila1336Digitaria sp.Yellowing1NTNT
Arsi/Tibila1336Azadirachta indicaChlorophyll clearing2NTNT
Arsi/Tibila1336Solanaceous sp.Mosaic/mottling2NTNT
Arsi/Tibila1336Passiflora edulisInterveinal yellowing2NTNT
Arsi/Tibila1336P. hysterophorusPhyllody and yellowing1+NT+II-D16SrII-D
East Shewa/Abadiska1194P. hysterophorusYellowing1NTNT
East Shewa/Abadiska1194Galinsoga sp.Yellowing1NTNT
Shinile/Errer Gota1175Setaria sp.Pale yellowing/mosaic1NTNT
East Shewa/MARC1545Datura sp.Mosaic/yellowing/stunting1NTNT
East Shewa/MARC1545P. hysterophorusYellowing/phyllody1++II-B16SrII-B
East Shewa/MARC1545Ricinus communisLeaf yellowing/mosaic1NTNT
East Shewa/Ethioflora1630Digitaria sp.Yellowing/mosaic1NTNT
East Shewa/Ethioflora1630Brassica sp.Yellowing1NTNT
East Shewa/Ethioflora1630R. communisInterveinal mosaic1NTNT
East Shewa/Ethioflora1630Brassica sp.Interveinal mosaic/yellowing1NTNT
East Shewa/Ethioflora1630Pennisetum purpureumYellow stripe1NTNT
East Shewa/ZFP1650R. communisInterveinal mosaic/yellowing1NTNT
Arsi/Awash Bishola1537Datura sp.Leaf tip yellowing2NTNT

For DNA extractions, the Doyle & Doyle (1990) DNA extraction procedure using cetyl trimethyl ammonium bromide (CTAB) buffer was used with the modification that samples were ground using a Fast PrepTM FP 120 (QBiogene), run at 65 r.p.m. for 45 s without and with CTAB buffer in the first and second grindings, respectively.

PCR analysis of the 16S rDNA and secA genes

Nested PCR amplifications of phytoplasma 16S rRNA and secA genes were performed in 25-μL reaction volumes containing 12·5 μL 2 × Mangomix (BioLine), 1 μL each primer (10 μm), 10·5 μL ddH2O and 1 μL template DNA. PCR was run on an MJ Research PTc-200 thermocycler with reaction conditions of 94°C for 2 min followed by 35 cycles of 94°C for 30 s, 53°C for 60 s and 72°C for 60 s, and a final extension step of 72°C for 10 min. The same reaction conditions were followed in the first- and second-round PCRs.

Two sets of primer pairs were used in this study. For secA gene amplification, the primer pair SecAfor1 (5′-GARATGAAAACTGGRGAAGG-3′) and SecArev3 (5′-GTTTTRGCAGTTCCTGTCATNCC-3′) (Hodgetts et al., 2008) was used in the first round, and SecAfor5 [(5′-GARATGAAAACTGGRGAAGG-3′) + (5′-ASTCGTGAAGCTGAAGG-3′) + (5′-AGCTAAAAGAGAATTTGAGG-3′) + (5′-CTGATAGAGAAGCTAATGG-3′)] and SecArev2 (5′-CCNTCRCTAAATTGNCGTCC-3′) in the second round. One microlitre of the first-round PCR product was diluted in 40 μL ddH2O, and 1 μL of the diluted DNA was used as the template in the second-round PCR. For the 16S rDNA, the phytoplasma-universal primer pair P1 (Deng & Hiruki, 1991) and P7 (Smart et al., 1996) was used in the first-round PCR, and R16F2n/R16R2 (Gundersen & Lee, 1996) in the second round, under the same reaction conditions described above. All PCR reactions included sweet potato little leaf DNA from the University of Nottingham phytoplasma collection as a positive control and healthy Madagascar periwinkle (Catharanthus roseus) plant DNA as the negative control.

All PCR products were separated by 1% agarose gel electrophoresis in 1 × TBE (90 mm Tris-Borate-EDTA) buffer stained with ethidium bromide and visualized using UV light transillumination. DNA purification of PCR amplicons was performed using the GenEluteTM PCR clean-up kit (Sigma) following the manufacturer’s instructions prior to sequencing at the University of Nottingham, School of Biosciences sequencing centre. Sequences were processed on single strands using the Beckman Quickstart kit technology and wellRed Dye Chemistry (infrared dyes) with a CEQ 8000 Genetic analysis system (Beckman Coulter). blast analysis of sequenced data to obtain amplified phytoplasma phylogenetic group and taxonomic information was performed using the National Centre for Biotechnology Information (NCBI) database (Zhang et al., 2000). Sequence alignments were performed using clustalw (Thompson et al., 1994). Phylogenetic and molecular evolutionary analyses were performed with mega version 3·1 software (Kumar et al., 2004) using the neighbour-joining method with default values, and 1000 replications for bootstrap analysis.

LAMP analyses

Real-time LAMP assays were performed as described in Tomlinson et al. (2010a) except that amplifications were carried out on a Genie I instrument using Isothermal MasterMix (OptiGene) at 1 × concentration using the primer concentrations as described previously. Reactions were held at 63°C for 30 min and real-time LAMP results were analysed in terms of Tp values (the time taken to generate a positive result) with annealing temperature analysis used to validate the authenticity of the LAMP products. Two LAMP assays were designed, one for 16SrII and one for 16SrXII, each with six primers (external primers F3 and B3, internal primers FIP and BIP, and loop primers F-Loop and B-Loop). Primers were designed by eye, based on the sequences of the IGS and 23S rRNA sequences (Hodgetts et al., 2008; Tomlinson et al., 2010a) and are listed in Table 2. The LAMP assays were initially tested on DNA from a range of isolates (see Results), and assays subsequently contained either sweet potato little leaf (16SrII) or stolbur (16SrXII) DNA as a positive control and water as a negative control. As a further control to confirm that DNA samples supported isothermal amplification, universal primers for the plant COX sequence were used as described in Tomlinson et al. (2010b).

Table 2.   LAMP primers designed for 16SrII- and 16SrXII-group phytoplasmas
NameSequence 5′–3′


Symptoms observed in the field

Symptoms recorded in the field varied from plant to plant within crop and between locations (Table 1). The most commonly observed symptoms in papaya were bright yellowing of most upper young leaves at an early stage, leaf crinkling, pale yellowing, stunted plant growth and dieback at the late stages of infection (Fig. 2a). No observations were made on flowers, but root decay/rot and vascular discoloration was evident in some samples. All commercial varieties sampled (aged 1–3 years) appeared to succumb to the syndrome and symptoms were more commonly observed during the main rainy season (June–September).

Figure 2.

 Plants sampled and tested for phytoplasma infection. (a) Papaya plant showing dieback symptoms at late stage of infection; (b) declining citrus plants with mosaic, yellowing, twig dieback, small fruits and stunted growth; (c) gummosis on citrus trunk and branches; and (d) bark scaling, branch and twig dieback on citrus; (e) Parthenium hysterophorus weed with phyllody, stunting and yellowing symptoms; (f) solanaceous weed (Datura stramonium) showing mosaic/mottling, reduced leaf size and leaf curling.

In citrus (commercial grafted varieties of sweet orange, mandarin, lemon and lime aged 8–30 years, and non-commercial sour oranges) a range of symptoms were recorded. The most commonly encountered symptoms were leaf curling, distorted leaves, leaf shading, midrib and whole-leaf yellowing, leathery leaves, small necrotic spots, interveinal yellowing and mosaic, twig and branch dieback, stunted growth, poor stand, small-sized fruit and pale green fruit colour (Fig. 2b), declining trees, gummosis on trunk and branches (Fig. 2c) and bark scaling (Fig. 2d). Short roots and stem pitting was also evident on both rootstocks and budwood, but no observations were made on flowers. No symptoms were observed on grapefruit.

Phyllody, yellowing of leaves and stunting were common symptoms in P. hysterophorus weeds (Fig. 2e). Other weeds and crop plants sampled had one or more of the following symptoms: longitudinal yellow stripe on leaves and along midribs, interveinal mosaic/mottling (Fig. 2f), leaf curling, chlorophyll clearing and leaf-tip yellowing. Insect infestations were observed in citrus trees, with the most frequently recorded insects being whiteflies, scale insects, leaf miners, mites and aphids, in this order.

PCR tests

DNA was extracted from 46 papaya samples, 15 citrus samples and 38 other plant samples collected in Ethiopia as detailed in Table 1. Initial phytoplasma assays were performed with the secA gene primers, based on those described in Hodgetts et al. (2008), but with the second-round primers modified to provide a new fully nested PCR rather than the seminested approach described in Hodgetts et al. (2008). These primers have been shown to amplify consistently from all phylogenetic groups of phytoplasmas in the University of Nottingham collection apart from 16SrII-B and 16SrXI (J. Hodgetts & M. Dickinson, unpublished observations). These primers detected phytoplasma in 18 papaya plants and one P. hysterophorus plant, but not in any other samples (Table 1). DNA preparations that tested negative with the secA gene primers were subsequently tested with the 16S rRNA gene primers. These primers produced PCR products from two orange plants and three additional P. hysterophorus plants (Table 1).

Sequencing and phylogenetic analysis

PCR products derived with the secA primers from 18 papaya plants were sequenced and all had sequences of 16SrXII phytoplasmas that were 99% identical to each other. A representative sequence was deposited in GenBank under accession number HM240286. Sequences were then compared to known secA sequences as published in Hodgetts et al. (2008) with the addition of the 16SrXII-B ‘Ca. Phytoplasma australiense’secA sequence (Tran-Nguyen et al., 2008). Results of the alignments showed that the papaya phytoplasmas were in 16SrXII subgroup A, the proposed ‘Ca. Phytoplasma solani’ group (Fig. 3). Analysis of the P. hysterophorus secA gene sequence (parthenium 4·17) showed this phytoplasma to be in 16SrII-D.

Figure 3.

 Dendrograms, constructed by the neighbour-joining method, showing the phylogenetic relationships amongst the papaya and Parthenium hysterophorus phytoplasmas based on DNA sequences of the secA gene. GenBank accession numbers for previously published sequences are shown in brackets. Bootstrap values > 50% (expressed as percentages of 1000 replications) are shown, and branch lengths are proportional to the number of inferred character state transformations. Bar, substitutions per base.

PCR products amplified from orange and P. hysterophorus using the 16S rRNA primers were also sequenced and these sequences were analysed using blast searches (Altschul et al., 1990), since there are many 16S rDNA sequences available on the database unlike for the secA gene. The analysis showed that the three P. hysterophorus samples amplified with these primers but not the secA primers belonged to 16SrII-B, whilst the two orange samples contained Bacillus spp. DNA and no phytoplasma (Table 1). The secA genes of these three P. hysterophorus samples were subsequently amplified using the seminested approach described in Hodgetts et al. (2008) and the 16SrII-B designation was confirmed (Fig. 3).

LAMP assays

To confirm the results and develop a more rapid and robust diagnostic test, separate LAMP assays were developed for 16SrII- and 16SrXII-group phytoplasmas. These primers were tested on DNA from phytoplasmas of all taxonomic groups 16SrI–16SrXIV (with the exception of 16SrVIII) in the collection at the University of Nottingham (see Hodgetts et al., 2008), including isolates of 16SrII-C, II-B, II-D and 16SrXII-A and XII-E. The 16SrII LAMP assay detected all 16SrII-group isolates tested but no phytoplasmas from any of the other taxonomic groups, and similarly, the 16SrXII assay only detected isolates from the 16SrXII groups (results not shown). Furthermore, the assays were tested on isolates of Bacillus spp. and gave no positive reactions to these. When the 16SrXII assay was tested on samples from papaya, citrus and other plant species (Fig. 4), it only gave positive reactions from papaya isolates that secA PCR had shown to be positive (Table 1). However, the LAMP assay was able to detect phytoplasma in three additional papaya samples that had been negative with PCR (Table 1). The 16SrII LAMP assay only gave positive reactions with P. hysterophorus samples that had been PCR-positive. One advantage of the real-time LAMP assay is that the result can be easily observed in real-time on the Genie machine without the need for gel electrophoresis and results can be confirmed by annealing curve analysis of the products (Table 3), which in this case showed that the annealing temperatures for the test samples were the same as those for the reference strains (86·6 ± 0·2 for 16SrXII and 82·9 ± 0·2 for 16SrII). To confirm that samples that were negative with the phytoplasma LAMP assays supported isothermal amplification, all negative samples listed in Table 1 were tested with the plant COX primers and showed positive amplification (data not shown).

Figure 4.

 Real-time LAMP analysis data as shown on the Genie machine screen. The x-axis shows time in minutes, whilst the y-axis shows fluorescence. Results for eight samples are shown, tested with the 16SrII primers (wells A1–A8) and the 16SrXII primers (lanes B1–B8). The samples were (1) papaya 1·13, (2) papaya 4·14, (3) papaya 3·61, (4) citrus 16·1, (5) orange 2·2, (6) orange 15·1, (7) 16SrII positive control (soybean phyllody), (8) 16SrXII positive control (stolbur). Amplifications can be seen for sample A7 (Tp = 13 min), B1 (Tp = 19 min) B2 (Tp = 25 min, B3 (Tp = 20 min) and B8 (Tp = 24 min) All other samples were negative.

Table 3.   Annealing temperature analysis for the LAMP assay shown in Figure 4
WellaPeak positionValueWidthHeight
  1. aWells are as shown in Figure 4, with A7 the positive control for the 16SrII assay, well B8 the positive control for the 16SrXII assay and wells B1–B3 are the positive papaya samples.



A range of diagnostic techniques for phytoplasmas have been developed since the advent of PCR. Numerous PCR primer combinations have been devised to amplify the 16S rRNA gene, some of which are universal primers that work on DNA from all phytoplasma phylogenetic groups, whilst others are group-specific (Smart et al., 1996; Firrao et al., 2005). However, diagnostics based on these primers can be problematic, with occasional false positives, particularly through amplification of any Bacillus spp. that might be present in a plant sample (Harrison et al., 2002; Nejat et al., 2009), and the number of steps, particularly if nested PCR is being employed, increases the chances of contamination of samples. In addition, it is important to guard against false negatives, since there is no internal control built into the diagnostic test to confirm that a negative result is caused by a lack of phytoplasma and not PCR inhibition. More recently, alternative methods, such as an oligonucleotide array system based on 16S rRNA gene sequences (Nicolaisen & Bertaccini, 2007) and terminal restriction fragment length polymorphism (T-RFLP) analysis based on the 23S rRNA gene (Hodgetts et al., 2007), have been developed to identify samples from different subgroups and to build internal controls into the diagnostic tests.

Real-time PCR assays have also been developed for both generic and specific phytoplasma detection, and these assays have the advantage of being more easily automated and less labour intensive than conventional PCR, such that appropriate controls can be conducted more easily (Christensen et al., 2004; Hodgetts et al., 2009). However, these assays are relatively slow compared to LAMP, and more recently a LAMP assay was developed for 16SrI and 16SrXXII phytoplasmas which, when combined with novel DNA extraction techniques, can result in a diagnosis within 1 h of sampling (Tomlinson et al., 2010a). The present study has refined the LAMP technique using enzymes and equipment supplied by Optigene UK, such that the analysis is undertaken in real-time based on the detection of a fluorescent dye incorporated into dsDNA generated during the amplification, hence assays can be undertaken even more quickly and without the need for any gel electrophoresis. The reaction mixtures are very easy to assemble, requiring just the enzyme mix, primer mix, water and DNA, and because the reaction and analysis is then in a closed tube, there is less potential for contamination of samples through opening and closing of tubes, steps that would be required for nested PCR and/or gel electrophoresis. Furthermore, LAMP assays have been shown to be less prone to enzyme inhibitors in DNA preparations than PCR, and this is supported by the fact that three papaya samples that were negative in nested PCR were positive with the LAMP assay (Table 1). In addition, all the samples that were positive with PCR amplified with the correct LAMP assay, and the COX control LAMP primers worked on all the plant samples.

A drawback of the LAMP assay is that knowledge of which phytoplasma is to be tested for is required, so that the correct assay can be performed. In this study, it was initially assumed that the phytoplasma present in papaya and citrus would be from group 16SrII. This assumption was based on a previous study in Ethiopia (Arocha et al., 2007), in which the samples were collected from farms in the Melkassa area of the East Shewa zone in September 2005, and 11 out of 12 papaya and six out of six citrus samples were reported to contain 16SrII-C ‘Ca. Phytoplasma aurantifolia’ phytoplasma, most similar to the causal agent of papaya yellow crinkle in Australia. In the present study, secA gene primers were initially used in a nested PCR to amplify from the papaya and citrus samples because these primers have been shown to work on most taxonomic groups. Additionally, sequence analysis of this region of the secA gene is relatively easy since it is a shorter sequence than the 16S rDNA, yet the resolution of phytoplasmas into their taxonomic groups is at least as good as through use of 16S rDNA and RFLP analysis, because of the much greater sequence diversity in the secA gene (Hodgetts et al., 2008). Sequencing of the PCR products from the present study revealed that the phytoplasmas in all papaya samples tested belonged to 16SrXII, and no phytoplasma was found in any citrus samples – sequencing of products subsequently amplified with P1/P7 from citrus only revealed the presence of Bacillus spp. Furthermore, the secA sequencing clearly placed the papaya phytoplasmas from Ethiopia into the 16SrXII-A ‘Ca. Phytoplasma solani’ subgroup (Fig. 3). These results are surprising, especially as 16 of the positive papaya samples came from the East Shewa zone where the 16SrII phytoplasma was previously reported. It is possible that there was a 16SrII phytoplasma present in 2005 which has now been replaced by the 16SrXII-A phytoplasma. Results from Australia suggested that both 16SrII and 16SrXII phytoplasmas can be found associated with disease of papaya. White et al. (1998) extracted DNA from three individual plants at a commercial plantation in central Queensland during January and February 1995 and found two were infected with 16SrII phytoplasmas (those showing yellow crinkle and mosaic symptoms) and one with a 16SrXII phytoplasmas (the plant showing dieback symptoms).

Because of the possibility that 16SrII phytoplasmas might have been present in the samples but as mixed infections with 16SrXII phytoplasmas, all the papaya plants were initially used with both the 16SrII and 16SrXII LAMP assays and no evidence was found of either mixed infection or 16SrII in any papaya samples. These results are in agreement with the findings on Nivum-Haamir-dieback of papaya in Israel, which is also in 16SrXII (Gera et al., 2005). Interestingly, Gera et al. (2005) implied that their phytoplasma most closely resembled the Australian papaya dieback 16SrXII-B phytoplasma based on sequence analysis. However, analysis here of their sequence suggests that their phytoplasma is in fact more similar to the 16SrXII-A stolbur group, and that the 16SrXII-B and 16SrXII-C phytoplasmas are geographically isolated to Australasia. The evidence presented clearly indicates that the 16SrXII isolate causing papaya dieback in Australia is from a different subgroup to the ones occurring in Ethiopia and Israel, but that 16SrXII phytoplasmas are associated with all these disorders.

Phytoplasmas were not found to be associated with all the papaya plants tested in Ethiopia (Table 1). In particular, phytoplasmas were not detected at the higher-altitude locations such as West Shewa and West Hararghe. This may be because the vector (which has not been identified) is not present at these higher altitudes, or it may reflect a failure to recognize the specific symptoms associated with the phytoplasma, and confusion with other symptoms, possibly caused by other biotic and abiotic factors.

The results from citrus are also notable, since no evidence was found of phytoplasmas, despite sampling from leaf samples, as was previously done (Arocha et al., 2007). This may reflect the uneven distribution of phytoplasmas in plants, loss of phytoplasma in the intervening years or a previous misdiagnosis. Apparently, there are no reports of citrus decline being caused by a phytoplasma in other countries, and most evidence suggests that the syndrome commonly referred to as ‘decline’ is in fact the result of nutrient deficiency and soil quality (Srivastava & Singh, 2009). The only confirmed phytoplasma disease of citrus is lime witches’ broom, which is caused by a 16SrII-C ‘Ca. Phytoplasma aurantifolia’ phytoplasma in countries such as Oman and the United Arab Emirates. However, whilst this phytoplasma could be experimentally grafted into rough lemon it could not be transmitted to sweet orange, sour orange, grapefruit or mandarins (Chung et al., 2006), suggesting that these are not hosts for the phytoplasma. It is therefore concluded that the citrus decline in Ethiopia is not caused by a phytoplasma, but by some other biotic or abiotic factors.

Phytoplasmas belonging to 16SrXII groups have also been found in other hosts, such as potato, in some parts of the world. Therefore, a range of weed plants from in and around the papaya plantations, including solanaceous plants, was tested to determine whether these might be alternative hosts and reservoirs for the papaya dieback phytoplasma. However, no 16SrXII phytoplasmas were found in any of these plants, despite the focus being on plants showing phytoplasma-like symptoms. Phytoplasmas were found associated with P. hysterophorus plants, and this agrees with previous findings in Ethiopia (Taye et al., 2004; Janke et al., 2007). With the secA gene primers, phytoplasma was only found in one P. hysterophorus plant, and this was found to be a 16SrII-D phytoplasma based on sequencing. However, when the other P. hysterophorus plants were tested with the 16S rDNA primers, phytoplasmas belonging to 16SrII-B were detected. These results were subsequently confirmed by the LAMP assay. This confirms the previous observations that the nested primers designed for the secA gene are able to amplify from phytoplasmas in nearly all the taxonomic groups tested, but less well from 16SrII-B. Producing universal sets of primers for phytoplasma diagnostics remains a problem. P1/P7 followed by nested PCR with primers such as R16F2n/R16R2 works well for most phytoplasma phylogenetic groups, although these primers were found to be unreliable for groups such as 16SrXI and XIV (unpublished observations). Martini et al. (2007) designed semi-universal primers for the rp genes which work as long as the right primer combinations are used for particular groups, and the nested secA primers used in this study have similar constraints. It is therefore important to have a range of diagnostic tools available for phytoplasmas to ensure correct and accurate diagnosis, and the group-specific LAMP assays developed in this and the previous work (Tomlinson et al., 2010a) provide a quick, cheap and reliable confirmatory diagnostic system.


The laboratory work of Berhanu Bekele at the University of Nottingham, UK was done under the sponsorship of the Rothamsted International (RI)-African Fellows Program (AFP). Phil Swarbrick was funded through BBSRC/DFID Grant no. BB/F004044/1 and Leverhulme/Royal Society Africa Award Grant AA090003, and Petra Nikolić was funded on a Short Term Scientific Mission as part of EU COST Project FA0807. Field work in Ethiopia was supported by the Ethiopian Institute of Agricultural Research (EIAR), Plant Protection Research Centre (PPRC). We acknowledge Dr Taye Tessema and Mr Takele Negewo for their help in the identification of weeds. Phytoplasma-infected plant material is held in the UK under Defra Plant Health Licence no. PHL 173B/5244.