Identification of ‘Candidatus Phytoplasma’ species in “huanglongbing” infected citrus orchards in the Caribbean

“Huanglongbing” (HLB) is one of the most devastating diseases of citrus orchards worldwide. Samples from 183 citrus plants of different cultivars and rootstock/cultivar combinations, showing HLB symptoms in three Caribbean countries (Cuba, Jamaica, and Guadeloupe-France), were collected to verify the possible co-infection of ‘Candidatus Phytoplasma’ and ‘Candidatus Liberibacter’ species. The 64% of the samples resulted positive to the ‘Ca. L. asiaticus’ and the 27% to diverse ‘Ca. Phytoplasma’-related species, moreover about the 14% of the samples infected with ‘Ca. Liberibacter’ were also found positive to phytoplasmas, indicating the presence of mixed infection especially in the orchards located in Cuba. Moreover, in one of the samples from Jamaica mixed phytoplasma infection was detected. Moreover the detection of only phytoplasmas in 11 symptomatic citrus samples collected from Cuba and Guadeloupe without ‘Ca. Liberibacter’ detection, confirmed that the symptomatology cannot be the sole criterium to discriminate between the presence of the two pathogens, and molecular detection is necessary to identify single or mixed infections. Diaphorina citri insects collected from Cuba and Guadeloupe resulted infected with ‘Ca. L. asiaticus’ confirming its active role in the dissemination of the pathogen. Only one insect of the Cicadidae family, collected in Guadeloupe, was found positive for phytoplasma presence. Considering that the phytoplasmas belonging to some ‘Candidatus species’ were detected in the three countries in different citrus varieties, a relevant role as phytoplasma reservoir can be attribute to citrus orchards.


Introduction
"Huanglongbing" (HLB) is one of the most devastating citrus diseases in the world (Bové 2006), mainly present with severe outbreaks in Asian and American continents (CABI 2019). It was detected also in several areas of the Caribbean, and, in particular, Cuba at the end of 2006 (Luis et al. 2009), Jamaica in 2009 (Brown et al. 2011), and Guadeloupe in 2012 (Cellier et al. 2014). Citrus industry is present in 140 countries worldwide covering an area of 10.3 million hectares, with a production of about 167 million tonnes in 2016 (FAOSTAT n.d.). In the Caribbean countries the impact of HLB on the citrus industry is so relevant that both production and export of Cuba and Guadeloupe economy are severely affected.
Despite the lack of fulfilment of Koch's postulates, HLB occurrence has been primarily associated with 'Candidatus Liberibacter asiaticus', 'Ca. L. africanus' and 'Ca. L. americanus' (Jagoueix et al. 1994;Teixeira et al. 2005) presence in symptomatic citrus plants. However, phytoplasmas belonging to diverse ribosomal groups were also reported in several citrus species showing HLB-like symptoms with or without 'Ca. Liberibacter' detection (Martinello-Sanches et al. 2016). Since in Cuba and in the Caribbean only 'Ca. L. asiaticus' has been reported associated with HLB presence, the aim of the study was to confirm whether phytoplasmas are also associated with HLB-like symptoms in single or mixed infection taking into account preliminary investigations carried out in Cuba, Jamaica and Guadeloupe, .

Plant and insect material
A total of 103 symptomatic citrus samples were collected in Cuba (Table 1), six in Jamaica, and 74 in Guadeloupe (Table 2) during extensive surveys carried out in 2017-2019 in diverse growing areas (Fig. 1). The citrus plants surveyed in Cuba enclosed "Valencia" sweet orange, grapefruit, "Eureka" lemon, Mexican lime, Persian lime, and "Nova Tangelo" orange of different ages (Table 1). In Guadeloupe, the majority of the tested citrus was grafted onto diploid and tetraploid rootstocks, and the age of the trees varied between 2 and 6 years, only in a few cases 10 or 20 year-old trees were tested (Table 2). In Jamaica the six samples examined were collected from an industrial orchard of "Valencia" sweet orange five years old located in the area of Bog Walk (Saint Catherine Parish) and from a garden plant of the ornamental variety Citrus aurantium in Montego Bay (Fig. 1). Leaf samples were collected mainly based on the presence of asymmetric leaf mottling symptoms (Fig. 2); two additional asymptomatic samples per Country were also collected as negative controls. Insect vectors or putative vectors were collected in the surveyed orchards in Cuba and Guadeloupe and, after gross morphological identification, tested to verify 'Ca. Liberibacter' and phytoplasma presence as described below.
DNA extraction DNA extraction was performed from 1 g of leaf midribs using CTAB (Murray and Thompson 1980) and phenolchloroform (Prince et al. 1993) based methods. The CTAB method was also applied to extract the DNA from the specimens of insect vectors or potential vectors collected in and near citrus orchards in Cuba (extracted in batches of 5) and Guadeloupe (extracted as single specimen).

PCR, RFLP and sequence analyses
The amplifications for the CTAB extraction method were carried out using 1 μl of the extracted DNA diluted 1: 30 in sterile distilled water, and 1 μl of a 20 ng/μl solution of the chloroform/phenol extracted DNA. The PCR amplification for 'Ca. Liberibacter' species detection was performed with primers OA1/OI2c and OI1/ OI2c (Jagoueix et al. 1996), while for the 'Ca. Phytoplasma' species detection the primers R16mF2/ R16mR1 (Gundersen and Lee 1996), U5/U3 (Lorenz et al. 1995), and M1/M2(=758f/1232r) (Gibb et al. 1995) were employed in direct and nested-PCR assays, respectively. Further nested-PCR assays following amplification with primers P1/P7 (Deng and Hiruki 1991;Schneider et al. 1995), 16S503f/LY16Sr (Harrison et al. 1999), and R16(I)F1/R1 (Lee et al. 1994) specific for ribosomal groups 16SrIV and 16SrI, -II and -XII respectively, were also performed. Phytoplasma positive controls from micropropagated collection (Bertaccini 2014) or from the published phytoplasma strain DNAs listed in Fig. 3, were amplified separately to avoid cross contamination of the samples. No positive control was used for the 'Ca. Liberibacter' detection. Negative controls were samples from asymptomatic citrus plants, devoid of nucleic acid template or containing sterile distilled water instead of the DNA template. The 'Ca. Liberibacter' and 'Ca. Phytoplasma' species identification was achieved by direct amplicons sequencing and/or RFLP analyses on 16S rDNA amplicons for the phytoplasma ribosomal group/subgroup identification (Lee et al. 1998). Sequence obtained after alignment were used to verify their clustering with reported 'Candidatus Phytoplasma' species (Bertaccini and Lee 2018). The evolutionary history was inferred using the Neighbor-Joining method (Saitou and Nei 1987) with the Maximum Composite Likelihood method in MEGA6 (Tamura et al. 2004(Tamura et al. , 2013. 'Ca. Liberibacter' species sequences were verified by BLAST analyses and sequence identity. For both pathogens the sequences of representative strains were deposited in GenBank.

Phytoplasma isolation, purification, and molecular identification
Randomly selected symptomatic and asymptomatic leaf samples (Tables 1 and 2) plus the six samples from Jamaica were employed for phytoplasma isolation in artificial media, following previously reported protocols (Contaldo et al. 2016(Contaldo et al. , 2019, with slight modifications. Approximately 100 mg of fresh midribs from each sample were placed in sterile mortars and mixed using a sterile pestle with 1 mL of succinate-citrate-phosphate buffer (SCP) (disodium succinate, 1 g/L; trisodium citrate, 1 g/L; K 2 HP0 4 , 1.5 g/L; and KH 2 P0 4 , 1 g/L; pH 7.0) (Sigma-Aldrich, St Louis, USA) (Minsavage et al. 1994). Immediately afterward, 100 μL were transferred into 8-mL monovette urine tubes (Sigma-Aldrich, St Louis, USA) and incubated at 25 ± 1°C. The tubes were examined daily for an acidification-induced colour change from orange to yellow. Non inoculated tubes were prepared following the same procedure. After the colour change, 100 μL of medium was inoculated onto plates containing 8 mL of CBs solid medium (Contaldo et al. 2016). Single colonies 36 h-old were transferred with sterile toothpicks into 8-mL monovette containing 2 mL of CBl for the purification steps. A gentle filtration of the liquid culture through 0.8-μm membrane filters (Whatman, Maidstone, UK), followed by 10-3/10-4 serial dilutions in CBl as described (Contaldo et al. 2019) was then performed. DNA extractions from 48 to 72 h-colonies and from CBl medium after filtration steps were carried out. PCR amplification of 16S rRNA coding gene, using the primer pair R16F2n/ R2 (Gundersen and Lee 1996) nested with U5/U3 primers was performed under reported cycling conditions (Contaldo et al. 2012). For each PCR reaction, DDSW (distilled, deionized sterile water) and non-inoculated media were used as negative controls. Phytoplasma identity was verified by RFLP and/or direct sequencing.

Results and discussion
The most common symptoms observed in all the citrus plants from which samples were collected were asymmetric leaf mottle and reduction of the plant growth (Fig. 2). Reduced flowering and fruits displaying a color inversion as they begin to mature before eventually the tree dies were also observed in the majority of the surveyed orchards. In Cuba, 100 symptomatic samples out of the 103 resulted positive for the 'Ca. L. asiaticus' presence, and the sequence of sample L3 in Table 1 showing 100% identity with available sequences of this bacterium was deposited in G e n B a n k u n d e r t h e a c c e s s i o n n u m b e r MW065487. Moreover 39 samples after nested PCR assays or isolation in artificial media were detected as positive for phytoplasmas tentatively assigned to eight ribosomal groups (16SrI, 16SrII, 16SrIV, 16SrVII, 16SrX, 16SrXI, 1SrXII and 16SrXV) as determined by 16S rRNA gene RFLP and/or sequence analyses (Table 1, Fig. 3, and data not shown). After phylogenetic analyses the sequenced citrus phytoplasma strains clustered with seven 'Ca. Phytoplasma' species and one ribosomal group (Fig. 4). The two samples from Fig. 2 Symptomatology observed in the citrus samples in A, B and C Guadeloupe; in D Cuba in which 1, sour orange; 2, grapefruit; 3, sweet orange; 4, Volkamer lemon; 5, Persian lime and 6, trifoliate orange Jagüey Grande, Matanzas and Miramar, la Habana of C. sinensis and C. aurantifolia, respectively, that were negative to HLB were positive to 'Ca. Phytoplasma' presence. The varieties in which the mixed infection of 'Ca. L. asiaticus' and 'Ca. Phytoplasma' was detected were C. latifolia, C. paradisi, C. reticulata x C. sinensis and C. sinensis. A total of 105 specimens of Diaphorina citri was collected in Cuba in the same citrus orchards surveyed (Fig. 1) and on Murraya exotica and Triphasia trifolia. All D. citri specimens resulted positive to the 'Ca.
Liberibacter' presence and negative for the 'Ca. Phytoplasma' presence.
In Jamaica, out of six symptomatic samples, five were positive to HLB, while the samples from C. aurantium plant were positive in nested PCR to phytoplasmas enclosed in the 16SrI and 16SrXII groups present in mixed infection (Fig. 3, and data not shown).
Only seven of the 27 samples processed for isolation resulted phytoplasma positive in the colonies obtained in the artificial medium. In particular, samples 13/18, 24/18, 4/18, C4L2, C4L7, and C4L11 from Guadeloupe, and OC2 and OC5 from Cuba, showed colour changes referable to microorganism growth after approximately 10-12 days from isolation. Afterward from the subsequent plating, colonies morphologically similar with those previously reported for phytoplasmas were obtained from all the isolation tubes but OC5 (Fig. 5) (Contaldo et al. 2012(Contaldo et al. , 2016(Contaldo et al. , 2019. Phytoplasma DNA amplification was performed on a total of 16 DNA aliquots from liquid and/or solid media of these samples, and the results revealed the presence of the same phytoplasmas detected in the corresponding plant materials (16SrI, 16SrIV, 16SrVI and 16SrXII) (Figs. 3   1353; 1078; 872; 603; 310; 281; 271; 234; 194; 118 and 72 and 4). In particular the phytoplasma isolation from the citrus samples from Guadeloupe was achieved from three symptomatic and three asymptomatic plants of polyploid genotypes grafted on polyploid rootstocks: 24/18-Persian lime 3x grafted on Volkamer lemon 2x, 4/18-Jackson orangelo grafted on 2x Citrumelo 4475/4x, 13/18-Navelate sweet Orange 2x grafted on Flor AG4x, C4L7-Rode Red Valencia sweet orange grafted on 2x Citrumelo 4475/2x, C4L2-Tangor Ellandale 2x grafted on Flhor AG1 4x, and C4L11-Tangor Ellandale 2x grafted on Volkamer lemon/2x, respectively. Four of these samples were negative to the presence of HLB (Table 2). Only one sample from an orange tree collected in Cuba resulted positive to phytoplasma presence in both liquid media and colonies (sample OC2, Table 1). The same sample was positive in plant leaf detection for 16SrXII and 16SrIV phytoplasmas, the same detected in the culture; in the detection from aliquots of liquid medium this sample resulted also positive for phytoplasmas of 16SrX group (data not shown).
In the Guadeloupe surveyed locations beside D. citri, specimens of the following insects were collected: Helochara spp., Lepidosaphes spp., Fiorinia proboscidaria, Coccus viridis, Icerya seychelles, Empoasca vitis, Zelus lungipe, Coccus viridis, Cicadidae, Muschidae, and Psyllidae. Only one specimen of D. citri collected in Vieux Habitants locality resulted positive for the 'Ca. L. asiaticus' presence, and the sequence was deposited in GenBank under the accession number MW065488; moreover the only unclassified Cicadidae collected from Capesterre Belle Eau resulted positive to 16SrXI phytoplasmas (data not shown). Thirty-seven citrus samples all collected from diverse localities in Guadeloupe and belonging to polyploid varieties grafted on diploid and tetraploid rootstocks tested negative to the presence of both pathogens ( Table 2). The asymptomatic samples resulted negative in all the PCR assays carried out (data now shown).
Phytoplasmas enclosed in groups 16SrI and 16SrXII were identified in the surveys carried out in all the three islands suggesting a possible primary dissemination with the propagation materials. The 16SrII, 16SrIV, 16SrVII, 16SrX and 16SrXI phytoplasma ribosomal groups were present in citrus in both Guadeloupe and Cuba samplings from experimental and small farms in the first case, and from larger orchards, in the second case. This could indicate that the surrounding cultivated species are influencing the phytoplasma population composition in citrus orchards, as indicated by the relevant presence of the lethal yellowing disease, that was severely affecting the Cuban palm plantations in the past years (Llauger et al. 2002;Paredes-Tomás et al. 2019). The identification of phytoplasmas enclosed in groups 16SrIII, 16SrVI and 16SrIX in Guadeloupe and in group 16SrXV in Cuba corroborates this hypothesis. In Guadeloupe the 16SrI phytoplasmas were recently detected in pomegranate and mango fruits from orchards (Castañeda-Alvarez et al. 2018) located near to citrus plot where the 16SrI phytoplasma was detected in the present study, which revealed the presence of natural source reservoirs of 16SrI group phytoplasmas in the area. Phytoplasmas of the same ribosomal group were reported from citrus in China and Mexico (Chen et al. 2009;Arratia-Castro et al. 2014). Moreover, the 16SrX phytoplasmas were detected in Guadeloupe in passion fruit trees located besides citrus plots . The consistent detection of three phytoplasma ribosomal groups in citrus only in Guadeloupe, could also be related to the variety of the diverse species cultivated in the island that are potential phytoplasma reservoirs. These phytoplasmas were already reported in the Caribbean areas and in some cases also in citrus crops (Teixeira et al. 2009;Caicedo et al. 2015, Mauricio-Castillo et al. 2015. In Cuba the presence of phytoplasmas in the 16SrI group was reported in papaya, beans, peppers, cabbage, beets, carrots, cassava, soybeans, strawberry and macadamia, 16SrII group in papaya, radish and medlar, and the 16SrX group in papaya (Arocha et al. 2007;Arocha et al. 2009;Zamora et al. 2012Zamora et al. , 2014Acosta-Pérez et al. 2013;Ferriol-Marchena et al. 2013;Pérez-López et al. 2013). All these species are cultivated as intercrop in citrus fields or are planted closely to citrus orchards.
The results of this work confirm the presence of mixed infection of phytoplasmas enclosed in different ribosomal groups with 'Ca. L. asiaticus' in almost all the citrus varieties with HLB symptomatology in Cuba, and in only three plants of diverse genotypes (Temple Tangor 2x and Nova tangelo 2x grafted on Citrumelo 4475 4x and Tangor Ellandale 2x grated on Flhor AG1 4x) in Guadeloupe. This type of mixed infection was reported in other citrus growing areas of the world (Chen et al. 2009;Teixeira et al. 2008;Arratia-Castro et al. 2014). In this survey the phytoplasmas were mainly detected in mixed infection with 'Ca. L. asiaticus'. In 11 samples, two in Cuba and 9 in Guadeloupe, the phytoplasmas were identified in symptomatic citrus plants where 'Ca. L. asiaticus' was not detected, and this confirms that the symptomatology alone cannot discriminate between the presence of the two pathogens, as already reported in Brazil (Martinello-Sanches et al. 2016). Moreover, the only phytoplasma presence in grapefruit was associated with the presence of HLB-like symptoms in China (Lou et al. 2014) and in Citrus reticulata in India (Das et al. 2016). The analysis carried out also indicates that 'Ca. L. asiaticus' has a higher titre than phytoplasmas since the latter were always detected by nested-PCR assays and/or after isolation in artificial media. However, the phytoplasma presence in citrus represents a source of infection for other crops, and made it more difficult to prevent their spreading to the surrounding natural environments or farmland.
In Guadeloupe, molecular tests have been carried out by French authorities (DAAF-SALIM) to identify the presence of 'Ca. L. asiaticus' in citrus orchards since 2013. A monitoring and information system of D. citri in the plots is also conducted as part of the epidemiosurveillance of the island (BSV Guadeloupe 2020). HLB is present in orchards and its spread is presently not controlled. Several of the plots where D. citri is present are symptomatic and the work carried out clearly indicates the need of surveillance also for 'Ca. Phytoplasma' species and their possible insect vectors.
In Cuba there is a production system of certified citrus material for the planting new orchards (López-Hernández et al. 2014), however there is a very high source of inoculum in the fields, so the plants are infected very early, and the management of HLB is not yet well established. In most of the commercial orchards there are populations of D. citri recorded on symptomatic trees, resulting in low citrus productivity in Cuba and Guadeloupe. These findings of widespread presence of both 'Ca. L. asiaticus' and various 'Ca. Phytoplasma' species and ribosomal groups clearly indicate the need for specific surveillance to apply better managements tools enclosing those aimed to contain the possible insect vectors presence after their appropriate identification.
Acknowledgements This work was carried out in the frame of EU H2020 R & I programme project "Insect-borne prokaryoteassociated diseases in tropical and subtropical perennial crops -TROPICSAFE" GA No 727459.
Author contributions All authors contributed to the study conception, design, material preparation, data collection and analysis. All authors read and approved the final manuscript.
Fundings Open access funding provided by Alma Mater Studiorum -Università di Bologna within the CRUI-CARE Agreement. The research leading to these results received funding from EU H2020 R & I programme project "Insectborne prokaryote-associated diseases in tropical and subtropical perennial crops -TROPICSAFE" under Grant Agreement No 727459.Data availabilityThe datasets generated during the current study are available in the GenBank database. Fig. 4 Evolutionary relationships of selected phytoplasma strains detected in citrus samples and reported as listed in Tables 1 and 2, using the Neighbor-Joining method. All the 'Candidatus Phytoplasma' reported species are enclosed in the analyses. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches (Felsenstein 1985). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The analysis was conducted in MEGA6. Acholeplasma laidlawii was used as outroot. The GenBank accession numbers are reported on the right of the 'Candidatus Phytoplasma' species and of the strains sequenced in the present work (in bold). Ribosomal subgroups are added for phytoplasmas used as reference (green bold) Fig. 5 Phytoplasma colonies from isolation using citrus tissues from sample orange OC2 from Cuba and resulted positive to 16SrXII and 16SrIV phytoplasmas (Table 1 and Fig. 4) (magnification 40 X)

Declarations
Ethical statement The authors declare no potential conflicts of interest. The research is not involving Human Participants and/or Animals. All the authors have read and approved the submitted manuscript text.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.