Investigations on annual spreading of viruses infecting cucurbit crops in Uttar Pradesh State, India

During 2018 an intensive study was conducted to determine the viruses associated with cucurbitaceous crops in nine agroclimatic zones of the state of Uttar Pradesh, India. Total of 563 samples collected and analysed across 14 different cucurbitaceous crops. The results showed the dominance of Begomovirus (93%) followed by Potyvirus (46%), cucumber green mottle mosaic virus (CGMMV-39%), Polerovirus (9%), cucumber mosaic virus (CMV-2%) and Orthotospovirus (2%). Nearly 65% of samples were co-infected with more than one virus. Additionally, host range expansion of CMV, CGMMV and polerovirus was also observed on cucurbit crops. A new potyvirus species, zucchini tigre mosaic virus, earlier not documented from India has also been identified on five crops during the study. Risk map generated using ArcGIS for virus disease incidence predicted the virus severity in unexplored areas. The distribution pattern of different cucurbit viruses throughout Uttar Pradesh will help identify the hot spots for viruses and will facilitate to devise efficient and eco-friendly integrated management strategies for the mitigation of viruses infecting cucurbit crops. Molecular diversity and evolutionary relationship of the virus isolates infecting cucurbits in Uttar Pradesh with previously reported strains were understood from the phylogenetic analysis. Diverse virus infections observed in the Eastern Plain zone, Central zone and North-Eastern Plain zone indicate an alarming situation for the cultivation of cucurbits in the foreseeable future.


Results
Symptomatology and dynamics of disease incidence in Uttar Pradesh. A total of 563 samples from 14 different types of cucurbit crops were collected from nine agro-climatic zones of Uttar Pradesh (Supplementary Fig. 1). The samples showed virus like symptoms including mosaic, yellowing, mottling with chlorotic spots, distortion of leaves, puckering, yellowing, vein clearing, upward curling of leaves, necrosis, and stunting of plant growth with reduction in leaf size across farmers' fields ( Fig. 1). Virus incidence was found in all agroclimatic zone of Uttar Pradesh and the incidence and distribution of the infected cucurbits viruses were varied based on the region of the collecting areas (i.e., the greater the growing area, the more the diseased samples collected). These three zones viz., Eastern, central and north-eastern plain zone accounts for more than 50 percent of total diseased samples ( Supplementary Fig. 2). Based on symptomatology average disease incidence varied between different agro-climatic zones ranging between 11 and 27% with an overall mean incidence of 24%. Our data revealed that the highest viral incidence was found in the Eastern and North-Eastern Plain zone (27%), followed by the central zone (26%) and the south-western zone (25%), respectively. The lowest disease incidence (11%) was recorded from the Bundelkhand zone (Fig. 2). Based on average field disease incidence, round melon (90%), satputia (72%), squash (56.07%) and watermelon (38.14%) were more prone to viral diseases (Table 1).
Virus detection and analysis of distribution pattern. All samples collected during the study were subjected to PCR for Begomoviruses and RT-PCR for RNA viruses such as potyviruses (PRSV and ZYMV), Cucumovirus (CMV), Tobamovirus (CGMMV), Orthotospovirus (PBNV and WBNV), Potexvirus, Crinivirus and poleroviruses. Out of 563 collected cucurbits samples, 95.4% (537/563) were found to be infected with viruses (either single or mixed infection). Despite showing virus like symptoms, the remaining 4.6% of samples tested negative for viruses. Universal and virus-specific primer pairs were used to test samples for eight virus genera, out of which samples were found to be positive for six genera (Potyvirus, Cucumovirus, Tobamovirus, Polerovirus, Orthotospovirus and Begomovirus). None of the samples were found to be positive for Potexvirus and Crinivirus. The relative frequency of viruses infecting cucurbit samples in our study showed dominance of Begomovirus (93%) followed by Potyvirus (46%), CGMMV (39%), poleroviruses (9%), CMV (2%) and Orthotospovirus (2%).
In general, the incidence of Begomovirus was higher in all the zones irrespective of the cucurbit crops. Diverse virus infection was observed only in samples tested from the Eastern Plain zone, Central zone and North-Eastern Plain zone while samples collected from the rest of the zones were positive for only potyviruses and CGMMV. Surprisingly, Polerovirus was primarily restricted to the Eastern plain zone, Central zone and North-Eastern Plain zone whereas Orthotospovirus and CMV were detected only among a few samples collected from the Eastern Plain zone and the North-Eastern Plain zone, respectively (Fig. 3). Potyvirus and CGMMV were the most frequent viruses detected among 64.3% samples associated with 12 different cucurbit crops ( Supplementary Fig. 3).

Expanded host range of viruses and new reports.
Viruses detected in this study have already been documented and characterized from different parts of India. From this study, expansion of viral host ranges on different cucurbits was observed. Infection by poleroviruses such as cucurbit aphid borne yellow virus (CABYV) on squash; melon aphid borne yellow virus (MABYV) on ivy gourd; Luffa aphid borne yellow mosaic (LABYV) on sponge gourd, bitter gourd and pumpkin has been recorded for the first time from India. In addition, CMV on sponge gourd, ivy gourd and ridge gourd and CGMMV on long melon were also documented. Besides, Zucchini tigre mosaic virus (ZTMV) was observed newly in India on pumpkin, bottle gourd, bitter gourd, squash and cucumber.

Mixed infection.
Virus-infected cucurbit samples showed a preponderance of mixed infection over single infection (Fig. 4). Among the tested samples, 366 samples (65%) were found to be co-infected with more than one virus (Fig. 4A). Among single infection (169 samples), infection dominated with Begomovirus (90%) followed by Orthotospovirus (6%), Potyvirus (2%) and CGMMV (2%) (Fig. 4B) Sequence analysis of viruses. In order to characterize the viruses infecting cucurbits in Uttar Pradesh state at the nucleotide level, selected samples were sequenced. Based on the sequencing, a CMV isolate infecting bitter gourd had 94.7% identity with the isolates reported earlier from Malaysia on cucumber (JN054637). The coat protein gene of the CGMMV isolate infecting cucurbits (bitter gourd, sponge gourd, cucumber, bottle gourd, long melon and snake gourd) was observed with more than 98% nucleotide identity with the CGMMV isolates recorded on various cucurbit crops characterised earlier from India. BLASTn analysis of nucleotide sequences amplified using a universal Polerovirus primer pair showed that three Polerovirus species (CABYV, MABYV and LABYV) were associated with different cucurbits of Uttar Pradesh. Similarly, sequence analysis using the BLASTn programme of the Nib region showed association of three potyvirus species (PRSV, ZYMV and ZTMV). In the coat protein region, PRSV showed more than 85% identity with isolates reported from India (EU475877) and China (KY933061); ZYMV showed > 98% identity with a Cucurbita pepo isolate (JN183062) reported from Iran; and ZTMV sequences showed > 85% identity with the sequences reported from France www.nature.com/scientificreports/    www.nature.com/scientificreports/ countries (clade A) and the second clade (B) clustered particularly with Indian isolates (Fig. 5c). Furthermore, phylogenetic tree generated based on the coat protein region of PRSV formed separate clades from the previously reported isolates from various countries including India (except an Indian PRSV-W strain, EU475877) ( Fig. 5d). Being a new virus to India, ZTMV isolates were found to have a close relationship with France isolates (Fig. 5c). With reference to ZYMV, all the study isolates were grouped with isolates reported from Asia, Europe, Africa and South America sharing their common ancestry (Fig. 5e). Similar analysis of WBNV isolates infecting round melon and watermelon typically displayed the same centre of origin with northern India isolates reported earlier and are clearly distinct from the southern India isolates (Fig. 5f). Phylogenetic analysis of polerovirus sequence showed 2 distinct clades of poleroviruses, one comprising CABYV and MABYV and the other comprising LABYV (Fig. 5g).

Generation of risk map of virus disease incidence using ArcGIS. Virus distribution and incidence
proportion for the four viruses were mapped using the functionalities of ArcGIS 9.1 software and the results are presented in Fig. 6. The maximum area under overall virus disease incidence fall under the category of 10-25% followed by 25-50% (Fig. 6A). The highest incidence was predicted with Polerovirus (50-89%) (Fig. 6B) whereas 25-50% incidence for CGMMV (Fig. 6C), Begomovirus (Fig. 6D) and Potyvirus (Fig. 6E) were observed. Since the survey points for CMV and Orthotospovirus are 3 and 2, these viruses were not mapped. The optimum numbers of power values were 7.59, 1, 5.28, 1 and 1 for Polerovirus, CGMMV, Potyvirus, Begomovirus and overall disease incidence, respectively. The green areas in the interpolated maps indicate very low to low disease incidences while the areas with orange to red colouration indicated moderate to high disease incidences. Around 75% area of Uttar Pradesh state fall under 10-25% disease incidence category for Begomovirus, Potyvirus and CGMMV whereas majority area of the polerovirus infection falls under 25-50% disease incidence category (Table 3).

Discussion
In The next to Begomovirus, Potyvirus was detected among 254 samples collected from 10 cucurbit crops (except watermelon, satputia, musk melon and round melon). Due to the diverse nature of potyviruses prevailing in the cucurbits ecosystem, a single universal primer pair (PNIbF1/PCPR1) could not amplify the majority of samples. Therefore, different sets of primer pairs were used for potyvirus detection (Table 4). Earlier studies from India have documented only the occurrence of PRSV and ZYMV on cucurbits 5,15 . ZYMV isolates from the current study were closely related to previously reported Indian isolates regardless of host, geographic location and year of collection as documented by Chikh-Ali et al. 16 in Syria. In the present study, in addition to these two potyviruses, occurrence of ZTMV has also been documented for the first time in India. Though diverse potyviruses were infecting cucurbits in India, much diversity was observed only with PRSV, but ZYMV and ZTMV seemed to have low diversity within the species. The widespread distribution of PRSV in single and mixed infections among tested samples indicated that it is an already established virus rather than a recent introduction. In contrast, CGMMV was detected among 223 samples in 11 different crops except for satputia, ridge gourd and round melon. CGMMV isolates in this study showed less diversity irrespective of the host and location of sample collection. Due to its seed transmission nature, single strains of CGMMV might have spread across the country. As  www.nature.com/scientificreports/ different cucurbit crops are cultivated side by side in the Indian subcontinent and harvested manually, mechanical transmission likely plays a major role in the expansion of virus host range. Besides, CMV was the least distributed virus detected only among 12 samples of bitter gourd, sponge gourd and ridge gourd. Though CMV has wide host range, the reason for its limited distribution among cucurbits remains unclear. Similarly, limited occurrence of CMV in cucurbits from commercial growers among the several viruses tested has been documented earlier 15,27 . Despite the fact that Orthotospovirus has been well documented on various cucurbits in the Indian subcontinent, only 10 samples of round melon and watermelon crops were found to be infected with Orthotospovirus in the UP state. There have been several reports on the occurrence of Orthotospovirus in southern India but not from northern India. Recently, research from our laboratory documented for the first time in India Orthotospovirus infection on cucurbits including PBNV on bitter gourd 20 and WBNV on round melon 28 . This study further reinforces the previous findings of Orthotospovirus being an emerging threat to the cultivation of vegetable crops in northern India. Three species of cucurbit infecting poleroviruses (CABYV, MABYV and LABYV) were identified and characterized in pumpkin, bitter gourd, sponge gourd, squash and ivy gourd. A previous study from southern India has documented the infection of bitter gourd and teasel gourd by CABYV 29,30 . Findings from the present study further confirm the existence of        The main aim of risk mapping of ArcGIS is to find the hazard level of the virus which is causing threat to crop cultivation 33,34 . It provides map showing probability of the occurrence of disease in the unexplored area based on the data collected from few pockets of the state. This gives clarity about the hot spot and cold spot for each virus separately as well as for the overall virus diseases infecting cucurbits. To our knowledge, this is the first of its kind in generating the risk map for virus diseases infecting cucurbits.
Overall, the present work outlined the current status of cucurbits viruses in UP state and provided conclusive evidence of viral diversity and the potential presence of new viruses currently emerging in this area. The distribution pattern of different cucurbit viruses in this region will help to identify the hot spots for viruses and will facilitate to devise efficient and eco-friendly integrated management strategies for management. Additionally, we identified many new viruses such as CMV on sponge gourd, bitter gourd and ridge gourd; CGMMV on long melon; Polerovirus on squash, pumpkin, ivy gourd and sponge gourd; and ZTMV on pumpkin, bottle gourd, bitter gourd, squash and cucumber. Further studies are needed to explore and characterize new and unidentified viruses which have been reported from different parts of world.   Detection of RNA viruses. Total RNA was extracted from 100 mg of each symptomatic and apparently healthy leaf samples using TriReagent (Sigma Aldrich, USA). Total RNA was subjected to reverse-transcription polymerase chain reaction (RT-PCR) for the preparation of cDNA using the RevertAid First Strand cDNA synthesis kit (Thermo Scientific, USA) according to the manufacturer's instructions. Prepared cDNA was used for virus detection through PCR amplification using specific primers (Table 4).
Cloning, sequencing and sequence analysis. Amplified PCR products of representative samples were purified using the QIAquick Gel Extraction Kit (Qiagen) and cloned in the pGEM-T Easy Vector System (Promega Corp.) 36 . Plasmid DNA preparations were obtained using Wizard Plus Minipreps DNA Purification (Promega Corp). Two clones were selected from each sample for sequencing and sequencing was performed at the Delhi University-South campus, New Delhi (India). Sequences were analyzed using the Basic Local Alignment Search Tool (BLAST) for the identification of virus at species level 37 . The top three to five hits against each database were included in the analysis for each sequence. The molecular evolutionary genetics analysis software (MEGA, version 7) was employed to determine the phylogenetic relationship of the study isolates with the earlier reported viruses using the neighbour joining method 38 .
Application of ArcGIS for disease mapping. The GPS (geographical positioning system) data and disease incidence data of the explored areas were collected to be used for mapping the epidemiological distribution and incidence proportion of viruses infecting cucurbit crops. The spatial variability maps of Polerovirus, CGMMV, Potyvirus, Begomovirus and overall disease incidence in nine agro-climatic zones were prepared using Geostatistical Analyst extension in ArcGIS 9.1 software. Inverse distance weighted (IDW) method was used for interpolation. The IDW is a simple interpolation technique which is based on the assumption that the variable values at unmeasured locations are influenced most by nearby observation points and less by distant points. This technique assumes that each measured location has a local effect, which reduces with distance by means of the utilization of a power parameter 39 . The formula of IDW 40 is given as follows: www.nature.com/scientificreports/ where Ẑ (S 0 ) is the predicted value at location S 0 , n is the number of measured sample points surrounding the prediction location, λ i is the weight assigned to each measured point, Z(S i ) is the observed value at the location S i . The formula to calculate the weights is given as follows: p is power parameter which reduces with increasing distance, d i0 is the distance between the prediction location (S 0 ) and each of the measured locations (S i ).
The optimal power value of IDW was estimated using root mean square error of cross-validation (RMSECV). Power value with the lowest RMSECV was selected for IDW interpolation 41 . A variable search radius with maximum of 15 sample points and minimum of 10 sample points was used.