SEARCH

SEARCH BY CITATION

Keywords:

  • disease detection;
  • diagnostics;
  • fragment analysis;
  • Mollicutes;
  • phytoplasma;
  • T-RFLP profile

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A terminal restriction fragment analysis (T-RFLP) technique was developed for the simple and rapid detection and diagnosis of phytoplasmas in plants. The selected primers amplified part of the 23S rRNA gene to provide improved resolution between the taxonomic groups compared to conventional restriction enzyme analysis of the 16S rRNA. Using the restriction enzymes Bsh12361 and MseI on the PCR products, and fragment analysis in the range 68–640 bp, the technique was tested on 37 isolates from 10 of the 16Sr groups. Distinct and unambiguous T-RFLP profiles were produced for nine of the 10 taxonomic groups, such that almost all isolates within a group shared the same profile and could be distinguished from isolates in other groups. The technique also identified the presence of mixtures of phytoplasmas from different groups in samples. Furthermore, the primers were devised to amplify a terminal restriction fragment (TRF) product of a specific defined size (461 bp) from the host plant chloroplast DNA, so that there was a built-in internal control in the procedure to show that the absence of a phytoplasma peak in a sample was the result of no detectable phytoplasma being present, not the result of PCR inhibition. This method offers the possibility of simultaneously detecting and providing a taxonomic grouping for phytoplasmas in test samples using a single PCR reaction.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Phytoplasmas are wall-less bacteria in the class Mollicutes that inhabit plant phloem and are known to cause disease in hundreds of plant species worldwide (Liefting et al., 2004). Not all plant species infected with phytoplasmas have disease symptoms, but infected plants normally show symptoms such as virescence, phyllody, yellowing, witches’ broom, leaf roll and generalized decline (Bertaccini et al., 2005). New disease reports are published frequently, and the list of hosts is growing and includes economically important food, fibre, forage, fruit and ornamental plants.

Traditional methods of phytoplasma detection use symptomatology known to be associated with the strain in question, and microscopy to locate the organisms. However, the accuracy of such methods is variable and lacks distinction between strains, or even groups. Since it is not possible to isolate and study phytoplasmas in pure culture, molecular methods, in particular PCR, are preferred for detection and diagnosis in plants and insect vectors. Most phytoplasma taxonomy has been based upon the 16S rRNA gene, which is highly conserved throughout phytoplasma groups, is present in two copies and is easy to amplify. Phylogenetic analysis of the 16S rRNA initially resulted in two tentative taxonomic systems (Lee et al., 1998; Seemüller et al., 1998). In 2004, the IRPCM Phytoplasma/Spiroplasma Working Team − Phytoplasma Taxonomy Group proposed to accommodate phytoplasmas within the genus ‘Candidatus Phytoplasma’ gen. nov., in which the formal descriptions and species composition fit closely to the 16S groups of Lee et al. (2000) (IRPCM Phytoplasma/Spiroplasma Working Team − Phytoplasma Taxonomy Group, 2004; Firrao et al., 2005). However, it is of interest to determine how a group system based on 16S rRNA similarity compares when constructed with other genes, such as the 16S/23S spacer region, the 23S rRNA, or less well-conserved genes such as secA. Usually the spacer region located between the 16S and 23S rRNA genes shows greater variation than the 16S rRNA gene, since this region has fewer evolutionary constraints (Barry et al., 1991), and sequence analysis of the 16S/23S spacer region for phytoplasmas results in a classification similar to that derived from 16S rRNA data, but with more detailed subdivisions (Wang & Hiruki, 2005). However, whilst there are more than 200 sequences available in databases of phytoplasma 16S rRNA genes and spacer regions, there are very few sequences available of phytoplasma 23S rRNA or other genes, so comparisons of these have not been possible.

As a result of the low titres of phytoplasmas present in infected plant material it is often appropriate to use nested PCR, and in the most widely used test, phytoplasma-specific primer pair P1/P7 are used initially, followed by R16F2 and R16R2, which yield a fragment 1·2 kb in length (Anfoka et al., 2003). The efficiency of nested PCR has shown that it can reamplify the direct PCR product in dilutions of 1:60 000 (Khan et al., 2004). Nested PCR can involve using group-specific primers for the second round of amplification, for example R16(I)F1/R1 group-I-specific, R16(III)F2/R1 group-III-specific and R16(V)F1/R2 group-V-specific primers (Anfoka & Fattash, 2004). However, a system has not yet been devised to identify all the taxonomic groups, and this approach requires more than one PCR step, increasing the chances of contamination between samples, and does not provide the rapid and simple diagnostic tool required. An alternative to the use of group-specific primers is to digest the 16S PCR products with specific restriction endonucleases, for example AluI, HaeIII or RsaI. The pattern of cut DNA is viewed using agarose or acrylamide gel electrophoresis and can provide a more informative analysis of the phytoplasma present (Lee et al., 2002), although it can prove difficult to distinguish between some of the taxonomic groups using this approach.

Other methods that have recently been developed for phytoplasma diagnostics include real-time PCR (Christensen et al., 2004; Torres et al., 2005) and heteroduplex mobility assays (HMA). Recently, 62 phytoplasma isolates collected from North America, Europe and Asia were analysed by HMA of the 16S/23S spacer region (Wang & Hiruki, 2005). This study demonstrated that HMA was sensitive enough to differentiate between closely related phytoplasmas, and new isolates could be assigned to specific subgroups, particularly within the 16SrI group.

Terminal-restriction fragment length polymorphism (T-RFLP) analysis is a direct DNA-profiling method that usually targets rRNA (Klamer et al., 2002). This genetic fingerprinting method uses a fluorescently labelled oligonucleotide primer for PCR amplification and the digestion of the PCR products with one or more restriction enzymes. This generates labelled terminal restriction fragments (TRFs) of various lengths depending on the DNA sequence of the bacteria present and the enzyme used to cut the sequence. The results of T-RFLP are obtained through TRF separation by high-resolution gel electrophoresis on automated DNA sequencers. The laser scanning system of the DNA sequencer detects the labelled primer (Sakai et al., 2004) and from this signal the sequencer can record corresponding fragment sizes and relative abundances. Resulting data is very easy to analyse, being presented as figures for statistical analysis and graphically for rapid visual interpretation.

Because of its simplicity, T-RFLP is one of the most widely used molecular methods for bacterial ecology studies (Wu et al., 2004) and has potential for rapid, simple diagnosis of phytoplasma samples. This paper reports the use of T-RFLP on phytoplasmas, and shows that using primers based on the 23S ribosomal gene it is possible to resolve almost all of the 16S groups. Through the design of primers that amplify chloroplast DNA, an internal control was built into the system to guard against any potential problems of PCR inhibitors.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Phytoplasma DNA samples

The phytoplasma samples used in this study, their group classification and their various sources are detailed in Table 1. Most of the samples were maintained in the Madagascan periwinkle (Catharanthus roseus), except for Napier grass stunt [maintained in Napier grass (Pennisetum purpureum)] and Cape St Paul wilt, which was obtained from trunk borings of coconuts (Cocos nucifera) in Ghana. Healthy controls, comprising DNA from uninfected P. purpureum and C. roseus, were employed throughout the study. DNA was extracted from small quantities of plant material using the cetyl trimethyl ammonium bromide (CTAB) method of Doyle & Doyle (1990) or the small-scale DNA extraction method of Zhang et al. (1998).

Table 1.  Names, descriptions and sources of the 37 samples used in this study, along with the sizes of the phytoplasma TRFs following amplification with MJD5 and 23Srev and digestion with either MseI or Bsh12361
SampleDescription16Sr groupCandidatus’ speciesSourceMseIBsh12361
  1. The abbreviated names and descriptions for the samples are given along with their 16Sr group and putative ‘Candidatus’ species. Samples were provided as purified DNA or as infected Madagascan periwinkle or Napier Grass plants. Sizes of the TRFs are given in base pairs.

CHRYMChrysanthemum yellowsI-AP. asterisDNA, A. Bertaccini, Bologna129305
DIVDiplotaxis virescenceI-BP. asterisDNA, A. Bertaccini, Bologna129305
DAYDwarf aster yellowsI-BP. asterisPlant, P. Jones, Rothamsted129305
CACTAster yellows cactusI-BP. asterisDNA, A. Bertaccini, Bologna129305
EAYEuropean aster yellowsI-BP. asterisDNA, A. Bertaccini, Bologna129305
KVEClover phyllodyI-CP. asterisDNA, A. Bertaccini, Bologna114305
AYAApricot chlorotic leaf rollI-FP. asterisDNA, A. Bertaccini, Bologna129305
AY2192Aster yellowsI-LP. asterisDNA, A. Bertaccini, Bologna129305
AVUTAtypical aster yellowsI-MP. asterisDNA, A. Bertaccini, Bologna129305
CLPCleome phyllodyII-AP. aurantifoliaDNA, A. Bertaccini, Bologna129456
FBPFaba bean phyllodyII-CP. aurantifoliaPlant, R. Mumford, CSL, York111456
FBPSACrotalaria saltiana phyllodyII-CP. aurantifoliaPlant, P. Jones, Rothamsted111456
SOYPSoyabean phyllodyII-CP. aurantifoliaPlant, P. Jones, Rothamsted111456
TBBAustralian tomato big budII-DP. aurantifoliaPlant, P. Jones, Rothamsted129456
SPLLSweet potato little leafII-DP. aurantifoliaPlant, P. Jones, Rothamsted129456
PEPPicris echioides phyllodyII-EP. australasiaeDNA, A. Bertaccini, Bologna129456
CoPCotton phyllodyII-FP. australasiaeDNA, A. Bertaccini, Bologna111456
CXPeach X diseaseIII-AP. pruniDNA, A. Bertaccini, Bologna129450
GVXGreen Valley XIII-AP. pruniDNA, A. Bertaccini, Bologna129450
PYLVPeach western XIII-AP. pruniDNA, A. Bertaccini, Bologna129450
APIEuscelidius variegatusIII-BP. pruniDNA, A. Bertaccini, Bologna129450
JRIPoinsettia branching factorIII-HP. pruniDNA, A. Bertaccini, Bologna129450
LDGGhanaian Cape St Paul wiltIV-CP. cocosnigeriaeDNA, J. Nipah, Takoradi, Ghana420450
ULWElm witches’ broomV-AP. ulmiDNA, A. Bertaccini, Bologna213234
FD-CFlavescence doréeV-CP. vitisDNA, A. Bertaccini, Bologna213234
FD-DFlavescence doréeV-CP. vitisDNA, A. Bertaccini, Bologna213234
RuSRubus stuntV-EP. ulmiDNA, A. Bertaccini, Bologna213455
PWBPotato witches’ broomVI-AP. trifoliiDNA, A. Bertaccini, Bologna425237
BLLBrinjal little leafVI-AP. trifoliiPlant, P. Jones, Rothamsted425237
CPSCatharanthus phyllodyVI-CP. trifoliiDNA, A. Bertaccini, Bologna425237
ASHY-1Ash yellowsVII-AP. fraxiniDNA, A. Bertaccini, Bologna328234
AP-15Apple proliferationX-AP. maliDNA, A. Bertaccini, Bologna129230
GSFY-1German stone fruit yellowsX-BP. prunorumDNA, A. Bertaccini, Bologna129230
PDPear declineX-CP. pyriDNA, A. Bertaccini, Bologna129230
BVKFlower stuntingXI-CP. oryzaeDNA, A. Bertaccini, Bologna129456
NGSNapier grass stuntXIP. oryzaePlant, P. Jones, Rothamsted129456
STOLStolbur of pepperXII-AP. solaniDNA, A. Bertaccini, Bologna412441

16S ribosomal DNA analysis

Amplifications of 16S rRNA were performed in 25-µL reactions using Ready To Go PCR beadsTM (Amersham Pharmacia Biotech) containing 15 ng template DNA and 100 ng of each primer, in an MJ Research PTC200 thermocycler. The phytoplasma universal primer pairs P1 (5′-AAGAGTTTGATCCTGGCTCAGGATT-3′) (Deng & Hiruki, 1991)/P7 (5′-CGTCCTTCATCGGCTCTT-3′) (Smart et al., 1996) were used for first round PCR and the reaction conditions were 95°C for 3 min, followed by 30 cycles of 94°C for 30 s, 53°C for 90 s and 72°C for 90 s, followed by a final extension step of 72°C for 10 min. One microlitre of the P1/P7 reaction product was used as the template in the second-round PCR using primer pair R16F2 (5′-GAAACGACTGCTAAGACTGG-3′)/R16R2 (5′-TGACGGGCGGTGTGTACAAACCCCG-3′) (Gundersen & Lee, 1996). For the second round, reaction conditions were 95°C for 3 min followed by 35 cycles of 94°C for 30 s, 56°C for 90 s and 72°C for 90 s, followed by a final extension step of 72°C for 10 min. Following PCR, 10 µL of PCR products were digested with 0·5 U of either HaeIII, RsaI or AluI restriction enzymes (New England Biolabs) overnight at 37°C in 20-µL reaction volumes. Digests were separated on 1·4% agarose gels in 1 × TBE buffer containing ethidium bromide, and visualized under UV light.

Terminal-restriction fragment length polymorphism

T-RFLP was conducted using primers for the 23S rRNA gene. The fluorescently labelled primer used was 23Srev* (5′-TTCGCCTTTCCCTCACGGTACT-3′) (Anthony et al., 2000) and was labelled with D4 Beckman dye (Sigma Proligo). Two alternative forward primers were designed based on alignments between phytoplasma 23S ribosomal DNAs and chloroplast sequences and were MJD2 (5′-GTGGATGCCTTGGCACTAAGAG-3′) and MJD5 (5′-GGCACTAAGAGCCGATGAAGG-3′). PCR was performed in 25-µL reactions using Ready To Go PCR beadsTM containing 15 ng template DNA and 100 ng of each primer, in an MJ Research PTC200 thermocycler, with PCR settings of 94°C for 2 min, followed by 35 cycles of 94°C for 30 s, 56°C for 1 min and 72°C for 90 s, completed with a final extension step of 72°C for 15 min. After amplification, 10 µL of PCR product was digested with either MseI (T'TAA) or Bsh12361 (CG’CG) (Helena Biosciences) in 20-µL reaction volumes containing 1U of restriction enzyme for 2 h at 37°C. One microlitre of the digest was then diluted in 5 µL of water, and 1 µL of this was added to 38·5 µL of sample loading solution (Beckman Coulter) and 0·5 µL of size standard-600 (Beckman Coulter). The samples were overlaid with mineral oil and separated on a CEQ 8000 DNA analysis system. Fragments between 60 and 640 bp were considered and analysis between replicates indicated that terminal restriction fragments (TRFs) that differed by more than 1·25 bp could be considered to be different.

Sequence analysis

Samples for sequencing of the 23S rRNA gene were amplified with primers MJD5 and 23Srev. PCR products were gel-purified with the Qiagen gel purification kit, and sequenced directly using an ABI sequencing big dye terminator kit. Sequences were edited and analysed using the programs seqed, lineup and pileup in the Wisconsin (GCG) package.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

16S ribosomal RNA analysis

To confirm the presence of phytoplasma DNA and which 16Sr subgroup the phytoplasmas belonged to, nested PCR was undertaken on all 37 samples in the collection using primers P1 and P7, followed by R16F2 and R16R2. The collection included a number of samples provided as purified DNA by Professor Assunta Bertaccini, University of Bologna, Italy, and others where the DNA was prepared from infected plant material at the University of Nottingham. Following PCR the samples were digested with AluI, HaeIII or RsaI and analysed by agarose gel electrophoresis. The results for representative samples from the different subgroups following digestion with AluI are shown in Fig. 1. No PCR products were amplified from water controls or from DNA from C. roseus and P. purpureum control plants.

image

Figure 1. Restriction digest profiles of phytoplasma 16S rRNA; 1·4% agarose gel electrophoresis of AluI-digested R16F2/R16R2 PCR products for phytoplasmas from different 16Sr groups. Lanes: M, 1-kb ladder; 1, AY2192 (16SrI); 2, SPLL (16SrII); 3, JRI (16SrIII); 4, ULW (16SrV); 5, FD-C (16SrV); 6, PWB (16SrVI); 7, ASHY-1 (16SrVII); 8, AP-15 (16SrX); 9, GSFY (16SrX); 10, BVK (16SrXI); and 11, STOL (16SrXII). Sizes are indicated in base pairs (bp). For expansion of phytoplasma codes, see Table 1.

Download figure to PowerPoint

The digest patterns obtained confirmed the group classifications for all the isolates used as far as possible. However, none of the enzymes was able to separate group I and XII phytoplasmas, whilst discrimination between groups II, III and XI and between groups V, VI and VII was difficult and required high-resolution gel electrophoresis and use of either RsaI or HaeIII (results not shown). Furthermore, when DNA from more than one isolate was combined prior to restriction digestion, to simulate the results of a mixed infection, the restriction profiles were difficult to interpret because of the multiple bands present (results not shown).

T-RFLP primer design

By making use of 16S and 23S rRNA sequences held in databases, attempts were made to design primers that could be used for T-RFLP analysis. Predictions based on 16S rRNA and the 16S/23S spacer sequences failed to result in any primer/enzyme combinations that would give good resolution between the different taxonomic groups. For example, use of a labelled R16R2 primer followed by digestion with AluI would result in fragments of 151 bp for 16Sr groups I, II, III, VI, VII, VIII, IX, X and XI, a fragment of 74 bp for group V and a fragment of 139 for groups XII and XIII.

In a previous study on microbial contamination of salad vegetables (H. Meakin, S. Rossall, CER Dodd, WM Waites & MJ Dickinson, unpublished data), it was shown that the primers designed by Anthony et al. (2000) for the 23S rRNA gene gave better resolution between bacterial species in T-RFLP than primers for the 16S rRNA gene. Analysis of the 23S rRNA genes available in databases for phytoplasmas confirmed the potential for using this gene, and alignments were made between these sequences for onion yellows (accession number CP000061), ‘Ca. P. luffae’ (AF086621) and peach western X (AF533231), along with the tomato chloroplast DNA (AM087200), resulting in the design of two alternative forward primers, MJD2 and MJD5, that could be tested in combination with a fluorescently labelled 23Srev primer (Anthony et al., 2000) in T-RFLP analysis.

Evaluation of forward primers for T-RFLP analysis

The two alternative forward primers, MJD2 and MJD5, were used in combination with fluorescently labelled 23Srev in PCR reactions using DNA from healthy C. roseus and P. purpureum plants and from those infected with diplotaxis virescence (DIV) or Green Valley X (GVX) phytoplasmas. Following PCR, the resultant DNA was digested with either MseI or Bsh12361 prior to fragment analysis. Fig. 2a shows the results of this analysis for a healthy C. roseus plant compared with a DIV-infected plant.

imageimage

Figure 2. (a) T-RFLP profiles for Catharanthus roseus controls and diplotaxis virescence (DIV)-infected plants amplified with MJD2 and labelled 23Srev or MJD5 and labelled 23Srev, following digestion with either MseI or Bsh12361; (i) C. roseus amplified with MJD2/23Srev digested with MseI; (ii) C. roseus amplified with MJD5/23Srev digested with Bsh12361; (iii) DIV amplified with MJD2/23Srev digested with MseI; (iv) DIV amplified with MJD5/23Srev digested with MseI; (v) DIV amplified with MJD2/23Srev digested with Bsh12361; (vi) DIV amplified with MJD5/23Srev digested with Bsh12361. (b) Profiles for plants infected with different phytoplasmas amplified with MJD5 and labelled 23Srev following digestion with either MseI or Bsh12361; (i) FBP digested with MseI; (ii) FBP digested with Bsh12361; (iii) TBB digested with MseI; (iv) TBB digested with Bsh12361; (v) FD-C digested with MseI; (vi) FD-C digested with Bsh12361. (c) Profiles following digestion with MseI for a simulated mixed infection of DIV (16SrI) mixed with SOYP (16SrII) amplified with MJD5 and labelled 23Srev. Values on the x-axis are sizes in base pairs (as are the numbers on the peaks); y-axis gives signal strength in arbitrary units. For expansion of phytoplasma codes, see Table 1.

With primer MJD2, the C. roseus plants yielded a single TRF of 472 bp with both restriction endonucleases (Fig. 2a i). This was because, as predicted from the tomato chloroplast DNA sequence, neither of these enzymes cut the PCR product between the MJD2 and 23Srev primers, so this fragment is effectively a full-length PCR product from the plant chloroplast. With primer MJD5, a single TRF of 461 bp was generated, which was the predicted size based on the location of the MJD-5 primer 11 bp downstream of the MJD2 primer (Fig. 2a ii). TRFs of identical size were generated from healthy P. purpureum plants (results not shown). Comparisons between duplicate samples showed that TRFs differed by less than 1 bp between replicates. Therefore TRFs differing by more than 1·25 bp were considered to be different, in line with T-RFLP data from other studies (Edel-Hermann et al., 2004).

When these same primers were used on DNA from DIV-infected plants, the same TRFs were present. However, an additional TRF was also present in each sample. Using primer MJD2, this additional TRF was 129 bp when MseI was used (Fig. 2a iii) and 305 bp when Bsh12361 was used (Fig. 2a v); primer MJD5 resulted in the same 129-bp and 305-bp fragments (Fig. 2a iv and vi). In the database sequence for onion yellows, there was a predicted restriction site for MseI 129 bp upstream of the 23Srev primer, and a Bsh12361 site 305 bp upstream of this primer. Since DIV is in the same 16SrI group of phytoplasmas as onion yellows, these TRFs are consistent with them representing the amplified phytoplasma DNA. For GVX, the TRFs for MJD2 were 129 bp with MseI and 461 bp with Bsh12361, whilst for MJD5, the TRFs were 129 and 450 bp, respectively (results not shown).

The ratio of peak heights relative to each other in fragment analysis can be taken as an indication of the relative amounts of PCR products. With primer MJD2, the ratio of phytoplasma peak to plant peak was approximately 1:1 for DIV (Fig. 2a iii and v). However, with primer MJD5, the ratio was 3:1 (Fig. 2a iv and vi). These ratios represent the efficiency of the primers on two competing templates, plant and phytoplasma DNA. The 3:1 peak ratio in favour of the DIV phytoplasma DNA using primer MJD5 indicates that this primer was more effective at priming from phytoplasmas than MJD2 in the competitive PCR. These results were confirmed when other phytoplasmas were used, in that in all cases the phytoplasma peaks obtained using MJD5 were higher than those using MJD2 relative to the plant peaks (results not shown). It was therefore decided to use MJD5 as the forward primer for subsequent phytoplasma amplifications.

T-RFLP analysis of different phytoplasmas

Based on the results obtained using primers MJD5 and 23Srev, and because predictions based on the published 23S sequences for onion yellows, western X and ‘Ca. P. luffae’ indicated that MseI and Bsh12361 might give different-sized fragments for phytoplasmas from different taxonomic groups, T-RFLP was performed on DNA for all the phytoplasmas in the collection. Some representative results are shown in Fig. 2b, and the complete data are summarized in Table 1. In all samples, including those from Napier grass stunt and Cape St Paul wilt, a plant TRF of 461 bp was present, indicating that the PCR was successful. In addition a further single TRF was present in each sample.

T-RFLP analysis for simulated mixed infections

In order to simulate a plant infected with more than one phytoplasma, DNA samples from different infected plants were mixed prior to PCR, restriction digestion and fragment analysis. The results for such an analysis where samples of DIV (16SrI) were mixed with soyabean phyllody (SOYP; 16SrII) are shown in Fig. 2c. The results clearly demonstrated that both phytoplasmas were detected in the mixture, and that they were readily identified as belonging to different taxonomic groups, with MseI TRFs of 129 bp (DIV) and 111 bp (SOYP).

Confirmation that the T-RFLP products are phytoplasma 23S ribosomal DNA

To confirm that the TRFs obtained using primers MJD5 and 23Srev were phytoplasma ribosomal DNA sequences, PCR was performed on Napier grass stunt DNA using MJD5 and an unlabelled 23Srev primer. The resultant PCR products were then cloned and clones of putative phytoplasma origin were identified by restriction digestion with MseI. One of these clones was sequenced (accession number DQ836329), and the sequence was clearly identified as that of a phytoplasma, with 90% identity to onion yellows at the nucleotide level. Furthermore, this sequence had the predicted MseI site 129 bp upstream of the 23Srev primer.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Because of the unculturable nature of phytoplasmas, the standard method for identifying their presence in plants and insects has been to use PCR, often the nested approach using primers based on the 16S rRNA sequence (Anfoka et al., 2003). Such techniques require the use of appropriate controls to guard against false positives, and also to ensure that the lack of a PCR product is not caused by PCR inhibitors present in the plant or insect DNA preparation. These controls generally involve spiking DNA preparations with known phytoplasma DNA, to show that amplification can proceed. The taxonomic status of the phytoplasma is then determined by restriction enzyme digestion of the PCR product, although this does not give good discrimination between all the taxonomic groups, or by DNA sequencing (Lee et al., 2002). Such techniques can become cumbersome and time consuming, particularly when large numbers of plant or insect samples require screening for the presence of phytoplasmas, and the multiple PCR steps involved increase the potential for contamination of samples.

Alternative methods that have recently been developed include heteroduplex mobility assay (HMA) of the 16S/23S spacer region (Wang & Hiruki, 2005), where PCR products from phytoplasmas are annealed prior to electrophoresis in acrylamide gels. This technique has proved very useful for classification of phytoplasmas into groups and subgroups, but is not designed to be a rapid diagnostic method as samples have to be checked following the initial PCR and prior to annealing to confirm that amplification has occurred. This study was therefore aimed at developing a simple and easy-to-interpret test to detect phytoplasmas and give a good indication of their taxonomic group, yet requiring fewer steps than nested PCR/enzyme digestion and less expensive than sequencing.

Initially, the possibility was investigated of developing a T-RFLP test based on conventional phytoplasma primers and the 16S rRNA gene or 16S/23S spacer. However, no primer/enzyme combinations could be identified that would give good resolution between the different taxonomic groups. In addition, previous work on microbial contamination of salad vegetables showed that primers based on the 23S rRNA gave better resolution between bacterial species in T-RFLP than primers for the 16S rRNA gene, and also had the potential to simultaneously amplify plant DNA as a built-in internal control (Meakin et al., unpublished data). As a result, the 23S rRNA sequences for onion yellows, western X and ‘Ca. P. luffae’ were aligned with the tomato chloroplast DNA to result in the design of two alternative forward primers MJD2 and MJD5, which could be tested in combination with a fluorescently labelled 23Srev primer (Anthony et al., 2000) in T-RFLP analysis.

Because it was assumed that there would be more copies of the chloroplast genome than the phytoplasma in a given DNA sample, it was important to design primers more homologous to phytoplasma sequences than the chloroplast sequence. However, there was no way of predicting how effective these primers would be in competitive PCR using DNA from infected plants, nor could the expected sizes of the TRFs or the most appropriate enzymes to use for most phytoplasmas be predicted. Such data were therefore produced experimentally (Table 1), revealing that repeatable and consistent results could be obtained and that there was good resolution of the different 16Sr groups. For example, all isolates in group I, with the exception of clover phyllody, gave the predicted 129-bp fragment with MseI and the 305-bp fragment with Bsh12361. The result for clover phyllody presumably reflects some divergence in the rRNA sequence for this strain compared with other group-I isolates, which is consistent with analysis of the 16S rRNA sequence of clover phyllody (accession number AY265217). Consistent TRFs were found for all isolates tested in the 16SrIII, 16SrVI, 16SrX and 16SrXI groups. In the 16SrII group, two patterns were found when MseI was used, but all samples gave the same-sized TRF with Bsh12361, whilst in group 16SrV, there was a consistent TRF with MseI of 213 bp, but variation when Bsh12361 was used. Based on this data, a test is proposed in which MseI is used initially on samples to resolve groups IV, V, VI, VII, VIII (based on a predicted TRF of 207 bp from the loofah witches’ broom sequence) and XII. Any samples which give a 129-bp TRF can then be digested with the alternative enzyme Bsh12361, which would resolve groups I, (II or XI), III and X. Only groups II and XI would not be resolved from each other using this strategy, apart from those isolates within group II that give the 111-bp TRF with MseI. Although these enzyme combinations give good resolution between the 16Sr groups, no attempts have yet been made to resolve isolates into the different ‘Candidatus Phytoplasma’ species (Firrao et al., 2005). The simplest way to do this would be to sequence the 23S rRNA genes for all the different species and identify more appropriate enzymes to use. The availability of these sequences would also enable users to adapt the test and select different enzymes to suit specific requirements.

Nested PCR of the 16S rRNA gene was shown to detect the presence of very low levels of phytoplasmas in test samples (Khan et al., 2004). The samples tested with the 23S rDNA primers in this study were mostly from glasshouse-grown Madagascan periwinkle or from an in vitro collection, in which titres of phytoplasmas were higher than in many naturally infected hosts. However, the Napier grass stunt and Cape St Paul wilt samples were from natural hosts, where 16S rRNA nested PCR had been required for detection. These samples gave good results from a single round of PCR using the 23S rRNA primers, indicating that the primers can detect phytoplasmas at low titres, and more field samples are currently being tested to assess their sensitivity further.

There have been some reports of plants being infected with more than one phytoplasma (Zhang et al., 2004), and in such cases the use of restriction enzyme digestion of 16S rRNA PCR products and gel electrophoresis results in complex patterns that are difficult to interpret. Because T-RFLP results in only a single peak for each phytoplasma, results from mixed infections can be interpreted more easily using this technique (see Fig. 2c). In addition, many natural hosts for phytoplasmas are colonized by populations of endophytes and other bacteria. The 23S rRNA primers are not specific for phytoplasmas and will amplify DNA from these bacteria to give additional PCR products, and on some occasions these were detected from the test plants in this study (results not shown). However, these bacteria are not closely related to phytoplasmas and give different-sized TRFs, easily resolved by T-RFLP, e.g. Pseudomonas fluorescens gives TRFs of 200 bp and 135 bp with MseI and Bsh12361, respectively.

The T-RFLP technique and primers developed in this study amplify both phytoplasma and plant chloroplast DNA in a single PCR reaction from test samples. Using the restriction endonucleases MseI or Bsh12361 on these PCR products results in a TRF of 461 bp for the plant chloroplast DNA, and the presence of this peak alone would indicate that there is no phytoplasma present in a test sample, whilst the absence of this peak would indicate that the PCR has been unsuccessful and that there are PCR inhibitors present in the DNA sample. If phytoplasmas or other bacteria are present in the DNA sample, these will result in additional peaks and the TRFs for these would give a good indication of the 16Sr grouping of any phytoplasmas present.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors would like to thank Dr Phil Jones, Rothamsted Research, UK, and Professor Assunta Bertaccini, University of Bologna, Italy, for providing samples and for useful discussions. Phytoplasmas were held under Defra Plant Health Licence no. PHL 173B/5244 (12/2005).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Anfoka GH, Fattash I, 2004. Detection and identification of Aster yellows (16SrI) phytoplasma in peach trees in Jordan by RFLP analysis of PCR-amplified products (16S rDNAs). Journal of Phytopathology 152, 2104.
  • Anfoka GH, Khalil AB, Fattash I, 2003. Detection and molecular characterisation of a phytoplasma associated with big bud disease of tomatoes in Jordan. Journal of Phytopathology 151, 2237.
  • Anthony RM, Brown TJ, French GL, 2000. Rapid diagnosis of bacteremia by universal amplification of 23S ribosomal DNA followed by hybridisation to an oligonucleotide array. Journal of Clinical Microbiology 38, 7818.
  • Barry T, Colleran G, Glennon M, Dunican LK, Gannon F, 1991. The 16/23S ribosomal spacer region as target for DNA probes to identify eubacteria. PCR Methods and Applications 1, 516.
  • Bertaccini A, Franova J, Botti S, Tabanelli D, 2005. Molecular characterization of phytoplasmas in lilies with fasciation in the Czech Republic. FEMS Microbiology Letters 249, 7985.
  • Christensen NM, Nicolaisen M, Hansen M, Schulz A, 2004. Distribution of phytoplasmas in infected plants as revealed by real-time PCR and imaging. Molecular Plant-Microbe Interactions 17, 117584.
  • Deng SJ, Hiruki C, 1991. Genetic relatedness between two nonculturable mycoplasmalike organisms revealed by nucleic acid hybridization and polymerase chain reaction. Phytopathology 81, 14759.
  • Doyle JJ, Doyle JL, 1990. Isolation of plant DNA from fresh tissue. Focus 12, 135.
  • Edel-Hermann V, Dreumont C, Pérez-Piqueres A, Steinberg C, 2004. Terminal restriction fragment length polymorphism analysis of ribosomal RNA genes to assess changes in fungal community structure in soils. FEMS Microbiology and Ecology 47, 397404.
  • Firrao G, Gibb K, Streten C, 2005. Short taxonomic guide to the genus ‘Candidatus phytoplasma’. Journal of Plant Pathology 87, 24963.
  • Gundersen DE, Lee I-M, 1996. Ultrasensitive detection of phytoplasmas by nested-PCR assays using two universal primer pairs. Phytopathologia Mediterranea 35, 14451.
  • IRPCM Phytoplasma/Spiroplasma Working Team–Phytoplasma Taxonomy Group, 2004. Candidatus Phytoplasma’, a taxon for the wall-less, non-helical prokaryotes that colonize plant phloem and insects. International Journal of Systematics and Evolutionary Microbiology 54, 124355.
  • Khan JA, Srivastava P, Singh SK, 2004. Efficiency of nested-PCR for the detection of phytoplasma causing spike disease of sandal. Current Science 86, 15302.
  • Klamer M, Roberts MS, Levine LH, Drake BG, Garland JL, 2002. Influence of elevated CO2 on the fungal community in a coastal scrub oak forest soil investigated with terminal-restriction fragment length polymorphism analysis. Applied and Environmental Microbiology 68, 43706.
  • Lee I-M, Gundersen-Rindal DE, Davis RE, Bartoszyk IM, 1998. Revised classification scheme of phytoplasmas based on RFLP analyses of 16S rRNA and ribosomal protein gene sequences. International Journal of Systematic Bacteriology 48, 115369.
  • Lee IM, Davis RE, Gundersen DE, 2000. Phytoplasma: phytopathogenic mollicutes. Annual Reviews of Microbiology 54, 22155.
  • Lee ME, Grau CR, Lukaesko LA, Lee IM, 2002. Identification of aster yellows phytoplasmas in soybean in Wisconsin based on RFLP analysis of PCR-amplified products (16S rDNAs). Canadian Journal of Plant Pathology 24, 12530.
  • Liefting LW, Shaw M, Kirkpatrick BC, 2004. Sequence analysis of two plasmids from the phytoplasma beet leafhopper-transmitted virescence agent. Microbiology 150, 180917.
  • Sakai M, Matsuka A, Komura T, Kanazawa S, 2004. Application of a new PCR primer for terminal restriction fragment length polymorphism analysis of the bacterial communities in plant roots. Journal of Microbiological Methods 59, 819.
  • Seemüller E, Marcone C, Lauer U, Ragozzino A, Goschl M, 1998. Current status of molecular classification of the phytoplasmas. Journal of Plant Pathology 80, 2336.
  • Smart CD, Schneider B, Blomquist CL, Guerra LJ, Harrison NA, Ahrens U, Lorenz K-H, Seemuller E, Kirkpatrick B, 1996. Phytoplasma-specific PCR primers based on sequence of the 16S-23S rRNA spacer region. Applied and Environmental Microbiology 62, 298893.
  • Torres E, Bertolini E, Cambra M, Montón C, Martín MP, 2005. Real-time PCR for simultaneous and quantitative detection of quarantine phytoplasmas from apple proliferation (16SrX) group. Molecular and Cellular Probes 19, 33440.
  • Wang K, Hiruki C, 2005. Distinctions between phytoplasmas at the subgroup level detected by heteroduplex mobility assay. Plant Pathology 54, 62533.
  • Wu M, Song L, Ren J, Kan K, Qian P-Y, 2004. Assessment of microbial dynamics in the Pearl River Estuary by 16S rRNA terminal restriction fragment analysis. Continental Shelf Research 24, 192534.
  • Zhang J, Hogenhout SA, Nault LR, Hoy CW, Miller SA, 2004. Molecular and symptom analyses of phytoplasma strains from lettuce reveal a diverse population. Phytopathology 94, 8429.
  • Zhang Y-P, Uyemoto JK, Kirkpatrick BC, 1998. A small-scale procedure for extracting nucleic acids from woody plants infected with various phytopathogens for PCR assay. Journal of Virological Methods 71, 4550.