Simultaneous detection and quantification by multiplex qPCR of 'Candidatus Phytoplasma asteris' and 'Candidatus Phytoplasma fraxini' in a plant host and insect vectors

Phytoplasmas are bacteria transmitted by insects that can cause plant diseases. In Bogotá 'Candidatus Phytoplasma asteris' and ' Candidatus Phytoplasma fraxini', infect 11 species of urban trees, weeds, grass, potato and strawberry. A set of primers, that amplify both phytoplasmas species were designed and used for absolute and relative qPCR quantification of the 16SrRNA gene. The primers AJ-16Sr-F/AJ-16Sr-R allowed the amplification of ‘Ca. P. asteris’, ‘Candidatus Phytoplasma palmae’, ‘Ca. P. fraxini’ and ‘Candidatus Phytoplasma phoenicium’, not of ‘Candidatus Phytoplasma pruni’. Absolute qPCR detected phytoplasmas between 1 × 109 and 1 × 103 copies/μL DNA extract. Two species-specific hydrolysis probes, AJ-16SrI-Cy5.5 and AJ-16SrVII-TexRed, were designed to detect 'Ca. P. asteris' and 'Ca. P. fraxini' respectively, using the AJ-16Sr-F/AJ-16Sr-R primers. For relative quantification, the 18SrRNA gene was used as normalizer. Relative qPCR detected phytoplasmas between 1 × 109 and 1 × 103 copies/μL DNA extract. Multiplex reactions allowed the specific quantification of 'Ca. P. asteris', 'Ca. P. fraxini' in comparison to the normalizer. qPCR methods were validated on natural hosts Andean oak trees and leafhoppers. The relative quantification values were higher for 'Ca. P. fraxini' (x̅ RQ = 3203.1 ± 2622,9 n = 14) compared with 'Ca. P. asteris' (x̅ RQ = 14.9 ± 24,5 n = 6) in oak tree samples. In the leafhoppers, the relative quantification values ranged between RQ = 26.5 and RQ = 294,927.3 for 'Ca. P. fraxini’ and RQ = 34.8 and RQ = 1722.2 for 'Ca. P. asteris'. In conclusion, although absolute qPCR allowed the quantification of phytoplasmas by comparing Cq (quantification cycle) values of samples with a standard curve, it did not allow to differentiate between 'Ca. P. asteris' and 'Ca. P. fraxini'. In contrast, relative qPCR assays using specific hydrolysis probes allowed the specific detection and quantification of each phytoplasma, in individual and mixed infections in insect vectors and plant hosts.


Introduction
Phytoplasmas are a diverse group of phytopathogenic bacteria that lack cell walls, are mainly transmitted by hemipteran insect vectors, and are associated with diseases in many crops, ornamental and woody plants worldwide (Lee et al. 2000;Bertaccini et al. 2018).Phytoplasmas are fastidious bacteria classified in the class Mollicutes and provisional genus 'Candidatus Phytoplasma.'The term 'Candidatus' is used for taxa of nonculturable bacteria (IRPCM 2004).The difficulties in cultivating phytoplasmas have hampered their detection and species identification for a long time and have restricted their investigation to ecological studies, electron microscopy, and serological techniques (Nejat and Vadamalai 2013).
Thanks to the development of molecular detection techniques and particularly nested PCR of the 16S rRNA gene, it is now easier to detect phytoplasmas in infected plant and insect samples and to establish their phylogenetic relationships (Gundersen et al. 1994;Gundersen and Lee 1996).Standard nested PCR uses two pairs of primers that anneal to conserved regions within the 16S rRNA -23Sr RNA genes.Primer pairs frequently used for the initial amplification are P1/P7 or P1A/P7A (Deng and Hiruki 1991;Lee et al. 2004) and for nested PCR R16mF2/R16mR1 and R16F2n/R16R2 (Gundersen and Lee 1996;Hodgetts and Dickinson 2009), among others.Nested primers for the specific detection of phytoplasma groups have also been designed (Lee et al. 1994).
There are two classification systems for phytoplasmas based on the PCR amplification of the 16S rRNA gene.In the first one, a 16Sr RNA amplicon of 1200 bp obtained by nested PCR is analyzed by Restriction Fragment Length Polymorphism (RFLP), cutting with 17 restriction enzymes.Different restriction patterns generate different similarity coefficients that allow classification into groups and subgroups (Lee et al. 1998;Wei et al. 2008).In 2004, the taxonomically accepted taxon 'Candidatus' used for unculturable bacteria was introduced as a second classification system for phytoplasmas (IRPCM 2004).This scheme has been recently revised, so if the percentage of identity of a query 16S rRNA gene phytoplasma sequence is equal to or less than 98.65% in comparison to known sequences, it is considered a new 'Candidatus Phytoplasma' species (Bertaccini et al. 2022).
Although nested PCR is widely used for phytoplasma detection and classification, it has some disadvantages.It requires two consecutive amplification reactions which increases the contamination risk and extends the analysis time.Additionally, it can generate bands of nonspecific sizes that complicate the interpretation of the results (Gori et al. 2007;Fránová 2011).Furthermore, given the low concentration of phytoplasma in plants and insects, conventional PCR is frequently not sensitive enough for its detection, especially in samples collected in the field where the amount of phytoplasma DNA can be less than 1% of the total DNA (Bertaccini 2007;Firrao et al. 2007).
The quantitative real-time PCR (qPCR) is a more specific and sensitive method for phytoplasma detection than conventional PCR.It allows the accurate quantification of phytoplasma DNA, less analysis time, and reduces the crosscontamination risk of nested PCR (Baric et al. 2006;Satta et al. 2017).In a comparative study, nested PCR and qPCR of the 16Sr RNA gene were used to detect Aster yellows phytoplasma (AYp) in the leafhopper Macrosteles quadrilineatus (Hemiptera: Cicadellidae).In tests conducted over two years, 2.2% of the samples tested by nested PCR were positive, whereas 4.6% were positive by qPCR.Furthermore, qPCR analysis was performed twice as fast as nested PCR tests (Demeuse et al. 2016).Additionally, qPCR allowed the quantification of phytoplasmas in low titers in insects and plants (Monis and Giglio 2006).Hydrolysis probes labelled with different fluorogenic dyes allowed the simultaneous detection of different phytoplasma strains in a single reaction.For relative comparison, plant or insect host DNA was used as an internal control in the same PCR (Linck et al. 2017;Herath et al. 2010).For instance, a multiplex qPCR assay was developed to detect and quantify the 16SrII and 16SrIX phytoplasma groups in sesame plants, using specific primers and hydrolysis probe that anneal to the 16S rRNA gene.The 18S rRNA gene was used as an internal control to estimate the relative amounts of phytoplasmas present in the samples (Ikten et al. 2016).Furthermore, qPCR amplification of other genes different from 16S rRNA has also been used for phytoplasma quantification.In Rubus sp.plants infected with 'Candidatus Phytoplasma rubi', secY and plant 18S rRNA genes allowed the specific detection and relative quantification of phytoplasma DNA in comparison with plant host DNA (Linck et al. 2017).
In Cundinamarca, Colombia, 'Ca.P. asteris' and 'Ca.P. fraxini' infect at least 11 urban tree species, grasslands, weeds and crops such as potato and strawberry in urban and rural areas (Perilla-Henao and Franco-Lara 2014;Franco-Lara 2019;Franco-Lara et al. 2020, 2022;Varela-Correa and Franco-Lara 2020), making phytoplasmas diseases a threat for this ecosystem.Infected trees can serve as a permanent source of inoculum for wild tree species and other plants growing in this area.The Andean oak Quercus humboldtii (Fagaceae) is a native species of the Colombian Andes and is an essential urban tree in Bogotá, with approximately 15,000 trees planted in the city.In a recent survey where 238 Andean oaks of Bogota were tested, 94% of the trees were infected with p 'Ca. P. asteris' or 'Ca.P. fraxini', in single and mixed infections (Lamilla et al. 2022).To date, two insect species of the area, leafhoppers Amplicephalus funzaensis and Exitianus atratus (both Hemiptera: Cicadellidae) have been confirmed as vectors in transmission trials.Both species are polyphagous and capable of transmitting 'Ca.P. asteris' and 'Ca.P. fraxini' either in single or mixed infections (Perilla-Henao et al. 2016;unpublished results).
No effective chemical treatments are currently known to control diseases caused by phytoplasmas (Namba 2019;Bertaccini 2022).Early detection with rapid and sensitive molecular methods is critical to prevent the pathogen spread.In this work we standardized a qPCR method that allows the simultaneous detection of 'Ca.P. asteris' and 'Ca.P. fraxini' in single and mixed infections, in plants and leafhoppers, using 18S rRNA gene as a normalizer.

Primer and probe design
The AJ-16Sr-F/ AJ-16Sr-R primers that amplify the 16S rRNA gene were designed for absolute and relative quantification from sequences of 'Ca.P. asteris' (16SrI), 'Ca.P. fraxini' (16SrVII) and nine other phytoplasma species.The sequences were obtained from the GenBank database (https:// www.ncbi.nlm.nih.gov/ genba nk/) and were aligned using the Geneious Prime 2020 software (Biomatters) (Fig. 1).Two hydrolysis probes were designed that hybridise to internal sequences of the 16S rRNA amplified region.The AJ-16SrI and AJ-16SrVII probes specifically hybridise to 'Ca.P. asteris' and 'Ca.P. fraxini' (16SrVII), respectively (Table 1).Primers and specific probes for each phytoplasma were designed using the Geneious Prime 2020 software (Biomatters).The eukaryotic 18S rRNA gene was used as an endogenous control to normalize the amount of DNA and for the relative quantification of phytoplasmas in infected tissues.The EUKF/EUKR primers and the universal hydrolysis probes S5 probe (López-Andreo et al. 2005) that hybridize to the 18S rRNA gene of Andean oak and leafhopper nuclear DNA.Primers and hydrolysis probes were synthesized by Macrogen Inc. (Korea).
The primer specificity was tested in silico with the Primer-BLAST tool (www.ncbi.nlm.nih.gov/ tools/ primerblast).The secondary structure and Gibbs free energy of primers and probes were estimated with the mFold option of the UNAFold Web Server (www.unafo ld.org).
In initial tests, cloned phytoplasma DNA was used to optimize the qPCR conditions.A Maize bushy stunt phytoplasma (MBS) ('Ca.P. asteris') or Ash yellows phytoplasma (AshY) ('Ca.P. fraxini') 16S rRNA gene inserts, cloned into the pMiniT 2.0 plasmid (New England Bio-labs®), were used as positive controls.DNA extracts of phytoplasma-free Andean oak trees, previously tested by nested PCR, were used as negative controls, and double distilled water was used as a blank (No Template Control, NTC).Later, DNA extracts of naturally infected Andean oak trees and leafhoppers previously tested by nested  PCR, were used to validate the standarized conditions of multiplex absolute and relative qPCR tests.The nested PCR conditions for the detection of 'Ca.P. asteris' and 'Ca.P. fraxini with primers P1A/P7A (Lee et al. 2004) followed by primers R16F2n/R16R2 and RFLPs analysis (Gundersen and Lee 1996), have been described elsewhere (Lamilla et al. 2022).
To obtain a calibration curve to estimate the concentration of the phytoplasmas 16S rRNA gene, the plasmids described above were amplified by nested PCR using the P1A/P7A and R16F2n/R16R2 primers as described by Lamilla et al. (2022).The obtained amplicons of 1250 bp were used as concentration standards.The amplicons were purified using the Monarch® kit (New England BioLabs®) and quantified with a Qubit® fluorometer® (Invitrogen).The copy number of the PCR amplicons was calculated using the following formula: number of copies = (amount in ng * 6.022 × 10 23 )/ (amplicon length in bp * 1 × 10 9 * 650) (URI Genomics & Sequencing Center 2004).Aliquots of the amplicons of both phytoplasma species were subjected to nine serial dilutions.For MBS, 1.7 X 10 9 to 1.7 X 10 1 total copies, and for AshY, 2.33 X 10 9 to 2.33 X 10 1 copies were obtained.For the standard curve, each dilution was amplified in duplicate by qPCR, and a Cq value was assigned to the total number of copies of each dilution.Additionally, 20 ng of total DNA obtained from non-infected Andean oak trees or leafhopper specimens was added to each dilution to simulate the detection conditions of infected samples.The absolute amount of 'Ca.P. asteris' or 'Ca.P. fraxini' present in different samples was calculated by comparing the Cq values with the standard curve.
Assays to optimize the qPCR conditions for the relative quantification of 'Ca.P. asteris' and 'Ca.P. fraxini' also were conducted.The same dilutions described above were used to determine the detection limit for both phytoplasma species.Known concentrations of phytoplasma DNA of both species were added in various combinations to phytoplasmafree Andean oak tree or leafhopper DNA extracts (20 ng).The relative concentration of phytoplasma DNA was calculated by comparing the Cq value of each phytoplasma with that of the 18S rRNA gene used as a normalizer.The samples were tested in multiplex reactions where both phytoplasma and the 18S rRNA gene were amplified and detected with specific fluorescent probes.The relative quantification (RQ) calculations were estimated with LineGene 9600 PCR software version V1.0.07 2014 (Bioer®).

Tree and leafhopper sampling
Total DNA extracts from Andean oak trees and adult leafhoppers infected with phytoplasmas were used to validate the designed primers and probes.Twenty-four tree samples collected in Bogotá, Colombia, were used.Secondary phloem from low branches of approximately 3 cm in diameter was sampled and used for DNA extractions.A. funzaensis (n = 10) and E. atratus (n = 10) were sampled from infected Andean oak trees.A shaking trap (Japanese umbrella) was used to collect the insects from the three lowest branches of the trees by vigorously shaking each branch for 1 min  4 Standard curve obtained by qPCR using primers AJ-16Sr-F/ AJ-16Sr-R on serial dilutions 1.7 X 10 9 to 1.7 X 10 3 on (A) ´Candidatus Phytoplasma asteris´ (16SrI) amplicons and (B) on 2.33 X 10 9 to 2.33 X 10 3 ´Candidatus Phytoplasma fraxini´ (16SrVII) amplicons, to which 20 ng of phytoplasma-free Andean oak DNA were added.Phytoplasma-free Andean oak DNA was used as negative control.NTC = No Template Control (Aissat and Moulai 2016).The leafhoppers were collected with a mouth aspirator, stored in 90% ethanol, and identified using taxonomic keys (Dietrich 2005;Zahniser and Dietrich 2008;Krishnankutty et al. 2016).

DNA extractions
The total DNA of Andean oaks trees was extracted from 2 g of the secondary phloem using the modified method of Prince et al. (1997).The DNA extracts were cleaned using the New England BioLabs® Monarch® kit and stored at -20 °C.The absence of PCR inhibitors in the extracts was verified by PCR using the rpsF/rpsR2 primers (Oxelman et al. 1997), which amplify an intronic region of the rps16 gene of the chloroplast ribosomal protein.DNA from A. funzaensis (n = 10) and E. atratus (n = 10) was extracted from whole adult individuals using an Invisorb® Spin Tissue Mini Kit (INVITEK) and stored at -20 °C.These extracts were tested by nested PCR and later used to validate the primers and probes designed in this work for qPCR.

Specificity of primers and probes
Two qPCR methods were developed to estimate the absolute and relative concentrations of 'Ca.P. asteris' and 'Ca.P. fraxini', using the primers AJ-16Sr-F/AJ-16Sr-R that anneal to a region of the 16S rRNA gene.Specificity tests conducted with these primers showed that they produced the expected 149 bp amplicons when DNA of 'Ca.P. asteris', Fig. 5 Standard curve obtained by multiplex qPCR with the AJ-16Sr-F/AJ-16Sr-R primers and AJ-16SrI andAJ-16SrVII probes on serial dilutions 1.7 X 10 9 to 1.7 X 10 3 on (A)´Candidaus Phytoplasma ast-eris´ (16SrI) amplicons and on (B) 2.33 X 10 9 to 2.33 X 10 3 ´Ca.P. fraxini´ (16SrVII) amplicons, added to 20 ng of phytoplasma-free Andean oak ADN.Phytoplasma-free Andean oak DNA was used as negative control.NTC = No Template Control 'Ca.P. palmae', Ca.P. fraxini', and 'Ca.P. phoenicium' were used as a templates, but not when that of 'Ca.P. pruni', phytoplasma-free Andean oak or leafhopper extracts were used (Fig. 2).In addition, in silico analyses performed with sequence alignments from 20 representative phytoplasma species (Table S1), including species evaluated from DNA extracts, showed that the AJ-16Sr-F/AJ-16Sr-R primers amplify the 16S gene of different phytoplasma species.

Detection limit and sensitivity of qPCRs
To estimate the approximate detection limit of the AJ-16Sr-F/AJ-16Sr-R primers, absolute qPCR tests were performed on serial dilutions of 1 X 10 9 to 1 X 10 1 'Ca.P. asteris' and 'Ca.P. fraxini' DNA extracts.Both phytoplasmas were detected in concentrations of 1 X 10 9 to 1 X 10 3 copies/ μL DNA extract (Fig. 4).The Cq values obtained from the serial dilutions were used to construct a standard curve for the absolute quantification of phytoplasma DNA concentration.The correlation and efficiency of the standard curve for 'Ca.P. asteris' was 0.995 and 96.1 respectively, and for 'Ca.P. fraxini' 0.997 and 100, respectively.These values were calculated directly with the LineGene 9600 PCR software version V1.0.07 2014, using the formula E = 1-(10 ^ (-1/ slope)) *100.
In relative qPCR tests, 'Ca.P. asteris' and 'Ca.P. fraxini' were detected in dilutions from 1.7 X 10 9 to 1 X 10 3 copies/μL DNA extract in multiplex qPCR tests conducted on samples carrying one, or two phytoplasmas in different dilutions (Fig. 5).In samples containing DNA of the two phytoplasmas at different concentrations and mixed with phytoplasma-free Andean oak or leafhopper DNA, the estimated approximate detection limit for both phytoplasmas remained 1.7 X 10 3 copies/μL DNA extract.

Detection and quantification of phytoplasmas in Andean oak and leafhopper extracts
'Ca.P. asteris' and 'Ca.P. fraxini' approximate concentration was estimated by multiplex qPCR relative quantification on 24 Andean oak trees and 20 leafhoppers (n = 10 A. funzaensis and n = 10 E. atratus) DNA extracts.'Ca.P. fraxini' was detected in 14 trees, 'Ca.P. asteris' in six, and, both phytoplasmas were present in mixed infections in four trees (Table 2).In A. funzaensis, 'Ca.P. asteris' was detected in three DNA specimens, 'Ca.P. fraxini' in two, and both phytoplasmas were detected in one.In E. atratus, 'Ca.P. asteris' was detected in extracts from three individuals, and in one individual, both phytoplasmas were present (Table 2).

Discussion
In this work, we standardized two specific, sensitive, and fast qPCR methods that allow 'Ca.P. asteris' and 'Ca.P. fraxini' detection and quantification in both plants and insect vectors.Since qPCR is a sensitive method that allows the detection of low concentrations of target DNA (Monis and Giglio 2006;Demeuse et al. 2016) it can be used as part of management schemes for early diagnostic of phytoplasma infections.It can also be used to determine phytoplasma concentration in other plant hosts and insect vectors, the correspondence of infection levels and presence of symptoms, and to study the interaction between phytoplasmas in mixed infections.
Nested PCR, followed by RFLP and sequencing analysis are methods frequently used for detection of phytoplasmas in plant and insect hosts (Bertaccini et al. 2018).Nested PCR has a higher risk of cross-contamination, is less specific and sensitive, and more time-consuming than qPCR (Lee et al. 1994;Gundersen and Lee 1996;Linck et al. 2017).
We describe a multiplex qPCR that allows sensitive detection of 'Ca.P. asteris' and 'Ca.P. fraxini' in samples where the two phytoplasmas are present.The primers AJ-16Sr-F/ AJ-16Sr-R were designed to amplify both 'Ca.P. asteris' and 'Ca.P. fraxini' and specific hydrolysis probes allow the specific detection of each phytoplasma and its quantification.For relative quantification, the 18S rRNA gene of the host plant or insect vectors was used as a normalizer.We demonstrated experimentally that the primers AJ-16Sr-F/ AJ-16Sr-R amplified 'Ca.P. palmae', 'Ca.P. phoenicium' in addition to 'Ca.P. asteris' and 'Ca.P. fraxini', but did not amplify 'Ca.P. pruni'.According to Primer Blast, 20 other species of phytoplasmas can be amplified with primers AJ-16Sr-F/AJ-16Sr-R, although this must be confirmed experimentally (Table S1).Therefore, in mixed infections of two or more phytoplasmas, such as those existing in Bogotá (Lamilla et al. 2022), hydrolysis probes or other discriminating method are needed to determine the involved species.
In our initial experiments, attempts were made to discriminate the two phytoplasmas using the Tm of the amplicons.However, the Tm values for these two species were very close and did not discriminate them.Thus, these primers allow the detection and absolute quantification of any of the phytoplasmas in single infections, but not when the two are present.Later, we designed hydrolysis probes that allowed the specific detection and quantification of 'Ca.P. asteris' and 'Ca.P. fraxini' in single or mixed infection.The analytical specificity of the probes was experimentally validated on naturally infected samples or prepared samples.Multiplex qPCR assays have been used to assess for different phytoplasma species, other pathogens, and host species (Baric et al. 2006;Hodgetts et al. 2009;Pelletier et al. 2009;Danet et al. 2010;Oberhansli et al. 2011;Malandraki et al. 2015;Ikten et al. 2016;Minguzzi et al. 2016;Linck et al. 2017;Mittelberger et al. 2020).In this study, we report a multiplex qPCR method for the simultaneous quantification of 'Ca.P. asteris' and 'Ca.P. fraxini' for the first time, using the 18S rRNA gene as a normalizer.The sensitivity for both absolute and relative qPCR quantification was 1 X 10 3 copies of phytoplasmas per μL DNA extract, in single and mixed infections, similar to values reported in other qPCR studies.For instance, the absolute quantification for of 16SrII and 16SrIX phytoplasma groups was 1.8 × 10 2 and 1.6 × 10 2 DNA copies respectively (Ikten et al. 2016), and for the 16SrV group, 1 × 10 2 DNA copies (Linck et al. 2017).However, the actual total number of cells per sample is half because phytoplasmas have two rRNA operons in their genomes (Schneider and Seemüller 1994).Using 18S rRNA gene as a normalizer for multiplex qPCR with the EUKF/ EUKR primers and the S5 probe, we efficiently amplified Andean oak tree and leafhopper DNA as expected, since they were designed for the universal detection of eukaryotes (López-Andreo et al. 2005).
To validate the qPCR methods, DNA extracts of naturally infected trees and leafhoppers were used.The presence of phytoplasmas had been established in the Andean oak samples, but no in the insects.Using both the absolute and relative qPCR methods, all the trees were positive for phytoplasmas and the same phytoplasma species, either in single or mixed infections, were detected.On the contrary, not all the insects were infected with phytoplasmas.Of ten A. funzaensis tested, 'Ca.P. asteris' was detected in three specimens by nested PCR, however, by qPCR two additional specimens were positive for 'Ca.P. fraxini' and one for both phytoplasmas.Of ten E. atratus tested by nested PCR, three were positive for 'Ca.P. asteris' but when tested by qPCR one additional specimen resulted positive for 'Ca.P. asteris' and another for both phytoplasmas.(Table 2).These results indicate that qPCR was more sensitive than nested PCR as has been observed (Ikten et al. 2016;Linck et al. 2017).
The number of tree and insect samples used to validate the methods was low, but some tendencies are observed in relation to the concentration of phytoplasmas in them.For instance, RQ values vary in several orders of magnitude between for the same species of phytoplasmas in different plant or insect samples.The average RQ values of both phytoplasmas were higher in the insect vectors than in Andean oak trees.In average, titres of 'Ca.P. fraxini' were higher than those of 'Ca.P. asteris in Andean oaks and in A. funzaensis, but no comparison was possible for E. atratus because the number of samples were too low.More studies are need to determine the relationship between these phytoplasmas and their insect and plant hosts.

Table 2
Phytoplasmas detected in DNA extracts of Andean oak trees and insect vectors at four locations in Bogotá by A. nested PCR and B. multiplex qPCR

Table 3
Absolute and relative quantification by qPCR of ´Candidatus Phytoplasma asteris´ and ´Candidatus Phytoplasma fraxini´ on DNA extracts of trees and insects Hyphens indicate absence of samples from those locations