Genetic diversity of phytoplasma strains infecting chrysanthemum varieties in India and their possible natural reservoirs

Symptoms typical of phytoplasma infection such as phyllody, virescence, witches’ broom and yellowing were observed in 12 varieties of Chrysanthemum morifolium in floral nurseries and experimental fields at New Delhi, Karnataka, Maharashtra and Andhra Pradesh, India, during surveys made from 2015 to 2017. Disease incidence ranged from 15 to 30%. Phytoplasma presence was confirmed in all symptomatic chrysanthemum varieties by molecular identification assays. Sequence comparison, phylogenetic and in silico RFLP analyses of 16S rDNA sequences allowed the identification of the chrysanthemum infecting phytoplasma strains into different ribosomal groups and subgroups, namely 16SrI, 16SrII-D, 16SrVI-D and 16SrXIV. Detection of phytoplasma strains of 16SrII-D subgroup were also confirmed in symptomatic Chenopodium album and Parthenium hysterophorus plants grown in and around the surveyed chrysanthemum fields at New Delhi, whereas 16SrVI-D phytoplasma strains were detected in symptomatic Cannabis sativa weed and leafhopper Hishimonus phycitis individuals collected from the symptomatic chrysanthemum fields at New Delhi. This is the first report on the presence of 16SrVI and 16SrXIV groups of phytoplasmas in chrysanthemum plants. Studies on genetic diversity of phytoplasmas infecting the major chrysanthemum varieties in India and their epidemiological aspects had previously not been reported. The detection and identification of phytoplasmas in different chrysanthemum varieties could contribute to increase the awareness among farmers in the management of these diseases.


Introduction
Chrysanthemum (Chrysanthemum morifolium Ramat) occupies 3rd and 5th positions in the cut flower and pot plant trades; respectively (Anonymous 2015). It is used as cut flower and loose flower due to its attractive form and colour of flower. In India, it is grown commercially on a large area of 16.63 thousand ha with a production of 186.06 thousand MT (Anonymous 2014). Flower crops are affected worldwide by several biotic and abiotic stresses and phytoplasmaassociated diseases are the major threat to commercial cultivations and are responsible of severe economic losses (Chaturvedi et al. 2010;Bellardi et al. 2018).
The occurrence of phytoplasmas in chrysanthemum has been reported from Japan, Italy, China, Iran and India (Shiomi and Sugiura 1983;Conti et al. 1988;Raj et al. 2007;Chung 2008;Min et al. 2009;Rani et al. 2014;Yadav et al. 2015). Although chrysanthemum has been grown at a commercial scale in different states of India, very little is known about the phytoplasma strains infecting this crop (Rao et al. 2017). Therefore, the genetic diversity of phytoplasmas infecting the major chrysanthemum varieties in India was studied. In addition, molecular identification of these phytoplasmas presence in weeds that may act as alternative hosts and potential insect vectors responsible of their spread were also made.

Materials and methods
Surveys of gardens, nurseries and experimental fields at ICAR-Indian Agricultural Research Institute (ICAR-IARI), New Delhi, ICAR-Indian Institute of Horticultural Research (IIHR), Bangalore, Karnataka, Maharashtra (Bandra and Pune) and Andhra Pradesh (Rajahmundry) were made from October 2015 to February 2017 (Fig. 1). Symptoms of phytoplasma infections were recorded on 12 chrysanthemum varieties from 5 surveyed locations. The disease incidence on each variety was evaluated on the basis of visual inspection by counting the number of symptomatic plants in relation to the total number of plants inspected using formula: Three leaf samples each of symptomatic chrysanthemum varieties from all the four locations and three weed species (Chenopodium album, Parthenium hysterophorus and Cannabis sativa) from IARI, New Delhi were collected (Table 1) Asymptomatic chrysanthemum plant samples were collected from the fields of the same area but at distance from the symptomatic fields with no symptoms and used them as negative controls. Leafhopper species from the surveyed chrysanthemum fields at New Delhi were also collected at 15 days intervals from October 2016 to February 2017 by the sweeping net method. The captured leafhoppers were identified by morphology as per taxonomy manual on identification of leafhoppers (Viraktamath and Meshram 2019) at Division of Entomology, ICAR-Indian Agricultural Research Institute, New Delhi. DNA from plants and insects was extracted employing CTAB-based extraction procedures as described (Ahrens and Seemüller 1992;Maixner et al. 1995). PCR amplification was performed with the universal phytoplasma primer pair P1/P7 (Deng and Hiruki 1991;Schneider et al. 1995) followed by primer pair 3Far/3Rev (Manimekalai et al. 2010). PCR reactions were carried out following the protocol described by Rao et al. (2014). Five microlitres of PCR product was subjected to electrophoresis in a 1.0% (w/v) agarose gel, stained with ethidium bromide and observed under a UV transilluminator. All PCR products (~ 1.3 kb) were purified using the Wizard ® SV Gel and PCR Clean-up System (Promega, Madison, USA). PCR products were ligated in pGEM ® T vector (Promega) and cloned in Escherichia coli (DH5-α) following the manufacturer's instructions, the cloned products were outsourced for sequencing using M13Fwd/M13Rev universal primer pair in both directions at Agrigenome, Kerala, India. The sequences were assembled using DNA baser V.4 software online tool and submitted to GenBank. A database search was performed by BLASTn analysis at NCBI (www.ncbi. org). The 16S rRNA gene sequences were aligned with phytoplasma group/subgroup representatives available in GenBank using ClustalW (Thompson et al. 1994) and used to construct a phylogenetic tree by the Neighbor-joining method with 1000 replications for each bootstrap value using MEGA 7.0 software (Kumar et al. 2016). The 16Sr DNA sequence (Acc. No. AB680603) of Acholeplasma laidlawii was used as out group to root the phylogenetic tree.

Results
During survey of gardens, nurseries, experimental fields in four states of India, symptoms of leaf yellowing, phyllody, virescence, witches' broom and stunting were recorded on chrysanthemum varieties. The phyllody and witches' broom symptoms were recorded as most common symptoms (Figs. 2, 3). The symptoms recorded on chrysanthemum varieties in four states of India are listed in Table 1. The disease incidence on different chrysanthemum varieties were recorded from 15% in Indira variety at Kadiam, Rajahmundry, Andhra Pradesh to 30% in Pusa Anmol variety at IARI, New Delhi (Table 1). Symptoms were most pronounced at the flowering stage at all the surveyed locations.
Phytoplasma strains detected in C. album (GenBank Acc. No. MF040139) and P. hysterophorus (GenBank Acc. No.MF040141) plants showed a 99.9-100% 16S rDNA sequence identity with phytoplasma described in peanut witches' broom group whereas those present in C. sativa plants and H. phycitis leafhopper showed a 99.9-100% 16S rDNA sequence identity with 16SrVI group.
Phylogenetic analysis of 16S rDNA sequences showed that phytoplasma strains detected in different varieties of chrysanthemum, weed plants and H. phycitis leafhoppers were clustered with phytoplasmas enclosed in 16SrI, 16SrII, 16SrVI and 16SrXIV groups (Fig. 5). Subgroup assignment by comparison of virtual RFLP patterns derived from in silico digestions 17 restriction endonucleases indicated Pusa Centenary (GenBank Acc. No. KY612250), Ajay orange strain 1 (Gen-Bank Acc. No. KX641012) Ajay orange strain 2 (GenBank Acc. No. KY693690), Pusa anmol strain1 (GenBank Acc. No KY693687) and Pusa anmol strain 2 (GenBank Acc. No KY693688) Red spoon (GenBank Acc. No. MF040140), Chenopodium album (GenBank Acc. No. MF040139) and Parthenium hysterophorus (GenBank Acc. No. MF040141) belonged to 16SrII subgroup-D, However the RFLP pattern of phytoplasma strain Indira (GenBank Acc. No. KY693693) with the reference strain papaya mosaic disease phytoplasma (GenBank Acc. No. Y10096) showed variation for enzymes AluI, BfaI, HaeIII and HpaI could not be assigned to subgroups because of variation in the RFLP patterns with reference strain (Fig. 6a  and TaqI and could not be assigned to subgroups because of variation in the RFLP patterns with reference strain of 16SrVI-D subgroup (Fig. 7, Table 1  to subgroups because of variation in the RFLP patterns with reference strain (Fig. 8, Table 1). The Nayaka yellow (Gen-Bank Acc. No. KX641016) belonged to 16SrXIV group but could not assigned to subgroup because of variation in profilings of Mse1 and Taq1 (Fig. 9, Table 1).

Discussion
Different groups of phytoplasmas are reported earlier with different chrysanthemum species. Symptoms induced by aster yellows phytoplasma infection included stunt, yellowing, virescence and phyllody in C. frutescence in Italy (Bertaccini et al. 1992;Conti et al. 1988). The presence of phytoplasmas affiliated to aster yellows group in chrysanthemum was reported earlier from Italy, Korea and India (Marzachi and Bosco 2005;Chung 2008;Raj et al. 2007). Chrysanthemum yellows phytoplasma (16SrI-B subgroup) associated with witches' broom, vein clearing, dwarf and abnormal production of secondary shoots was reported in C. carinatum and C. morifolium from European and Mediterranean areas (Bertaccini et al. 1990;Conti et al. 1988;Saracco et al. 2005). The identification of 16SrII group phytoplasmas in C. grandiflorum was reported in Okinawa Prefecture, Japan (Naito et al. 2007) and India (Yadav et al. 2015).  China associated with 16SrI group, which was responsible for serious losses in flower quality.
Limited reports are available on identification of phytoplasma strains association with chrysanthemum varieties in India (Raj et al. 2007;Yadav et al. 2015). However, in the present study, two subgroups belonging to four groups of phytoplasmas (16SrI, 16SrII-D, 16SrVI-D and 16SrXIV) infecting 12 varieties of chrysanthemum were identified from four major chrysanthemum growing states of India. However, the restriction profiles of phytoplasma strain associated with chrysanthemum variety Johan weber from Delhi (Acc. No. KX641014), Nayak yellow from Pune (Acc. No. KX641016), Meera from Mumbai (Acc. No. KY472314) and Indira from Andhra Pradesh did not match completely with any of the earlier classified subgroups, hence they may be new strains and multilocus genes specific detection and real RFLP analysis are necessary to confirm the reported unusual variability.
Several weeds are reported as reservoirs of important phytoplasma strains which are suggested to play an important role in natural spread of phytoplasma strains and serve as natural alternative hosts, since they influence the population density of the vectors and act as source of inoculums (Pasquini et al. 2007;Duduk et al. 2018). In the present study, two phytoplasmas ribosomal groups (16SrII and 16SrVI) were identified in three weed species, that are also reported to be reservoir of different phytoplasma strains in different countries, viz., 16SrII-E subgroup in Italy (Tolu et al. 2006), 16SrIII (X-disease) group and 16SrXII group in Czech Republic (Safarova et al. 2011). In the present study, the report of Chenopodium sp. as a host of 16SrII group indicates its potentiality as natural reservoir. Earlier, P. hysterophorus weed was also reported as host of 16SrI and 16SrII group in India and China (Raj et al. 2009;Li et al. 2011;Mall et al. 2015;Yadav et al. 2016). Further, C. sativa and Achyranthes aspera weed species were also reported as host of 16SrI, 16SrII, 16SrVI and 16SrXIV groups of phytoplasmas in India (Rao et al. 2017), which suggests potentiality of these weed species as natural reservoirs of phytoplasmas. C. album, P. hysterophorus and C. sativa weed species that are prevalent in chrysanthemum fields at New Delhi and resulted positive for the 16SrII and 16SrVI groups of phytoplasmas in the present study may also play an important role for natural spread of phytoplasma strains.
Limited information is available on transmission of phytoplasma strains associated with ornamentals through insect vectors. Bosco et al. (2007) reported transmission of chrysanthemum yellows phytoplasma through three leafhopper vector species (Euscelis incisus, Euscelidius variegatus and Macrosteles quadripunctulatus) in Italy. Empoasca decipiens was proved to be an experimental vector in transmitting chrysanthemum yellows phytoplasma (16SrI) in Italy (Galetto et al. 2011a;. However, in the present study, the leafhopper, H. phycitis was positive to16SrVI group phytoplasmas. The identification of 16SrVI group of phytoplasma strain in H. phycitis and in two chrysanthemum varieties, Jaya and Johan Webber, in same field suggest it as vector, which needs further confirmation through transmission assays. This No. KX641014) digested using 17 different restriction endonucleases indicating that Jaya, White Nursery, Cannabis sativa and Hishimonas phycitis phytoplasma belonged to 16SrVI-D sub groups, but Johan Webber 'Chrysanthemum showed variation for RFLP pattern and could not be assigned subgroup leafhopper species has already been reported as vectors for 16SrI and 16SrII phytoplasma groups in India (Nabi et al. 2015;Gopala and Rao 2018), which indicates its potentiality to transmit different phytoplasma strains. In India, many crops are being cultivated in parallel with other agricultural crops in different seasons. The widespread occurrence of phytoplasma strains identified in chrysanthemum varieties in the study pose a serious threat of chrysanthemum cultivation in India and needs further studies on epidemiology and management.  No. KX641016) digested using 17 different restriction endonucleases indicating that this strain belonged to 16SrXIV group but could not be assigned subgroup 16SrXIV-A subgroup because of difference in MseI and Taq1 restriction profiles