Location

The study was carried out in the Experimental Field of the Superior Agricultural College of the State of Guerrero (CSAEGro), in Cocula, Guerrero, Mexico, located between the coordinates 18° 26’ 27.20’’ north latitude and -99° 65’ 07.98’’ longitude west, at 635 meters above sea level. Experimental cultivation of pumpkin var. Gray Zucchini was established between two tributaries of water, the San Juan River and Las Juntas irrigation water channel.

Virus detection of pumpkin curly leaf and phylogenetic analysis

From 15 symptomatic pumpkin leaves, DNA was extracted with the DNeasy Plant kit (QIAGEN^®). Degenerated oligonucleotides prV324 (5-gccyatrtayagraagccmag-3’) and CoPR (5-gangsatghgtrcadgccatata-3’) were used to detect begomoviruses in general, and SLCV in particular, these oligonucleotides bind to begomovirus DNA and amplify fragments of approximately 570 bp (^{Wyatt and Brown, 1996}; ^{Yongping et al., 2008}). The PCR reactions were performed following a program with an initial temperature of 94 °C for 4 min, followed by 35 cycles of 94 °C 60 s, 50 °C 45 s and 72 °C 45 s; with a final extension temperature of 72 °C for 6 min.

A Thermo™ thermocycler (Thermo Scientific, Wilmington, DE) was used for DNA amplification and the visualization of the amplified products was carried out by electrophoresis in 1% agarose gels at 62 volts for 5 min, followed by 100 volts by 40 min, and were observed in a UVMR light transilluminator (Labnet, Edison, NJ). Two representative samples of the PCR products were sequenced, and the sequences obtained were compared with the database of the Gene Bank (GenBank) of the National Center for Biotechnological Information (NCBI) (http://www.ncbi.nlm.nih.gov/).

The consensus sequences were edited and assembled with the CAP (Contig Assembly Program) option of the BioEdit 7.2.5 Software (Tom ^{Hall Ibis Biosciences) (Hall, 2004)}. In the evolutionary analysis, all consensus sequences were aligned with the ClustalW program (^{Thompson et al., 1994}) included in the MEGA 7 software (^{Kumar et al., 2016}). The phylogenetic reconstructions of the data were performed using the maximum parsimony method, using the Subtree-Pruning-Regrafting algorithm, search option (level=1) with the initial tree by random addition (10 repetitions) and the missing spaces were considered as complete deletions. To calculate the confidence values of the tree clades, a bootstrap test was performed with 1 000 repetitions (^{Felsenstein, 1985}). The sequence obtained was deposited in the NCBI GenBank database.

SLCV virus pathogenicity

The SLCV transmission test, by whitefly (B. tabaci), was performed on pumpkin (C. pepo) var. Gray Zucchini during the period February-March 2015. This was carried out in cages made with two expanded polystyrene glasses of 1 L capacity, with windows covered with organza fabric.

On March 15, adults of B. tabaci were collected using an oral sucker, in a pumpkin crop with virosis syndrome established in the experimental field of the Superior Agricultural College of the State of Guerrero. 30 insects were placed per cage, which contained a pumpkin plant. For this, five repetitions and two witnesses were included. They were inspected for 21 days to detect the appearance of symptoms of virosis. From the infected leaves, obtained from the cages, one was selected and DNA was extracted and by PCR the virus was detected using the degenerate oligonucleotides prV324/CoPR.

Population fluctuation of B. tabaci and spatio-temporal analysis of the severity of SLCV

In February 2015, pumpkin var. Gray Zucchini in a batch of 2 880 m², where a census was carried out in 20 quadrants with 10 rows and 20 plants per row. Traps were distributed in the four cardinal points and in the center of the lot by an experimental design of randomized complete blocks with two repetitions.

The types of traps were: a) 18 × 24 cm yellow sticky trap placed on a metal support 1 m high and 1 cm in diameter; and b) yellow tray trap of 17 and 12 cm of upper and lower diameters and 10 cm deep with the capacity to store 1 L of water. The traps were inspected 12 times, every three and four days, except in the last evaluation where the interval was 10 days. The number of adults of whiteflies in them was counted.

The data of these variables were analyzed with the SAS 9.4 statistical program. The severity of virosis was calculated using the ^{Van Der Plank (1963)} scale and five evaluations were performed every eight days. To determine the effect generated by this phytopathological problem during the entire pumpkin growing cycle and compare the epidemic of virosis, the area under the disease progress curve (ABCPE) was calculated using the severity percentage data (^{Campbell and Madden, 1990}). The data analysis included maps of the spatial distribution, such as response surfaces and contour curves (isolines) obtained using the kriging interpolation method, using the SAS 9.4 statistical program.

SLCV detection and phylogenetic analysis

Severe symptoms of pumpkin leaf curl were observed at the experimental site. PCR and sequencing analyze revealed that the leaves with virosis symptoms obtained in the Cocula Valley, Guerrero Mexico, were infected with SLCV. The virus genome sequence obtained in this work was deposited in the NCBI GenBank (accession KX620945.1 DAAV-1).

^{El-Dougdoug et al. (2009)} identified the SLCV in C. pepo var. Eskandrani, using the same oligonucleotides as in the present investigation; reported that amplified DNAs were 480 bp in size. The phylogenetic analysis showed that the SLCV found in Guerrero, Mexico is closely related to the SLCV accessions found in Cairo (DQ285019) and The Imperial Valley of California (USA) (DQ285016), with bootstrap reliability value of 96%. In addition, the accessions of American and Middle Eastern origin differed from the group of Asian accessions (AM260206, EU47910 and EFQ99774) (74%) (Figure 1).

Figure 1 Phylogenetic tree obtained with the maximum parsimony method using the Subtree-Pruning-Regrafting algorithm. The reliability values of the nodes were formed with 1 000 bootstrap resamples. Bold access corresponds to the variant evaluated in this study. 

SLCV pathogenicity

The results of this test indicated that whitefly transmitted by pumpkin SLCV is pathogenic. Symptoms manifested at 8 days after the introduction of whitefly in the cages. SLCV infection causes leaf curl, growth retardation, epinastia, intervenal chlorosis and mottled leaves (Figure 2). These results coincide with what was reported by ^{Idris et al. (2006)}, who studied the SLCV virus in cucurbits, in Jordan and Egypt and found that the infection affected leaf size and reduced plant growth by 40%.

Figure 2 SLCV pathogenicity test in pumpkin var. Gray Zucchini plants. Healthy leaf (left) and leaves infected with the pumpkin curly leaf virus with different severity (right). 

Also, ^{Sobh et al. (2012)} reported that SLCV infection caused severe symptoms of pumpkin leaf curl (Cucurbita sp.) Camelia F1 variety, at 2 weeks after inoculation; that is, they were presented in double the time compared to the results of the present trial. Similarly, they agree with ^{Taha et al. (2016)}, which confirmed that the SLCV virus is transmitted by whitefly and induces severe growth retardation, leaf curly and causes serious losses in the production of cucurbitaceae. In addition, it was determined that the amplified product of the SLCV, obtained from infected pumpkin leaves of the experiment, had a size of 464 bp. The sequence was deposited in the GenBank of the NCBI (accession KX620948.1 DAAV-2).

Population fluctuation of B. tabaci

The number of whitefly adults showed significant differences in the majority of the samples, due to the type of trap. It was found that, at all sampling dates, the sticky trap was more effective than the tray with water, to capture these insects, whose averages ranged from 7.4 (sample 2) to 379 (sample 11).

The analysis of the cumulative sampling of all samples indicated that 2 082 adults (data not shown) were captured in the sticky trap, compared to the tray type trap (416 specimens). The location of the trap only significantly influenced the amount of insects captured in samples 1, 3 and 6. In the first samples (1 and 3), the averages indicate that the largest populations of flies were recorded at the West site; however, in the final evaluations, the highest insect density occurred in the traps of the East and North.

The highest accumulated amount of B. tabaci was obtained on the East side, with 538 individuals, the lowest value was 443 and was found on the South site (Table 1). On the other hand, there were significant differences between the sampling dates (Figure 3), since the populations varied in a range of 26 (date 2) to 437 insects (date 11). Pest management programs should be based on accurate knowledge of the main factors responsible for changes in the dynamics of the target insect population.

Figure 3 Number of whiteflies according to the sampling date. 

The biological cycle parameters estimated in different biotic and abiotic conditions provide the basic tools to understand the changes and behavior of pest species. The species B. tabaci is a poikilothermic organism, that is, the temperature influences the biological cycle; therefore, it is important to consider this factor to explain the ecology of the population. In this regard, ^{Drost et al. (1998)} reported that, in addition to temperature, host plants and biotypes influence the biology of B. tabaci.

The results obtained confirm what ^{Curnutte et al. (2014)}, who mention that oviposition, the survival of nymphs and the reproduction of B. tabaci; were significantly affected by the temperature, since the total reproductive capacity decreased 36.4% at 33 °C, also, they noted that 28 °C was the favorable temperature for whitefly development; however, they determined that the optimum temperature for the reproduction of B. tabaci fluctuates from 28 to 33 °C.

In this study, it was found that during the sampling the average temperature increased, which in combination with the phenology of the crop favored the increase in the reproductive rate of B. tabaci. It has been shown that temperatures from 30 to 32 °C favored the development of B. tabaci (^{Bonato et al., 2007}; ^{Curnutte et al., 2014}). In this regard, ^{Quintela et al. (2016)} found that the maximum population of the B. tabaci biotype B in corn (Zea mays) coincided with high temperatures in Brazil.

These findings are similar to those obtained in the present study, where the population of the pest increased in relation to the increase in temperature. On the contrary, ^{Bonato et al. (2007)} reported that the temperature of 32.5 °C was the optimum for the development of nymphs of B. tabaci (biotype Q); in addition, they determined that total fertility (eggs per female) varied from 105.3 (21 °C) to 41 (35 °C) and that longevity decreased as the temperature increased. Similarly, they argued that the association between temperature and biological cycle parameters is useful for predicting the effect of temperature on population dynamics.

Spatio-temporal analysis of the severity of SLCV

The average severity percentage (PS) recorded; through, the evaluation dates fluctuated between 5.14 and 73.28. It was determined that, from evaluation 5, the average PS remained stable, while the temperature registered an increasing trend, which was related to the severity of the SLCV (Figure 4).

Figure 4 Relationship of average temperature and severity of SLCV in pumpkin var. Gray Zucchini plants, in five evaluations in Cocula, Guerrero. 

The relationship between the severity and the temperature observed in this work coincides with that reported by ^{Ali et al. (2014)} who reported that temperatures between 37.16 and 37.78 °C led to the development of the disease; however, the minimum average temperature of 27.43 °C was favorable for the incidence of geminivirus in cotton and showed significant effect during two consecutive cycles of culture. The combined analysis, through the evaluation dates, indicated that quadrants 7 to 10 and 16 to 20 had a higher PS (47.88 to 54.87%).

It was found that ABCPE recorded significant differences (p< 0.0001) between quadrants and grooves; the highest incidence of disease occurred between quadrants 7 to 10 and 16 to 20, with values of 1431.72 to 1674.41; likewise, it was observed that the severity in quadrants 1 to 6 and 11 to 15 was less than 50%, while quadrants 17 and 19 had the highest severity of virosis in the pumpkin crop. The PS and ABCPE were lower in quadrants 2 to 5, with values of 39.9 to 40.23% and 1049.78 to 1134.66, respectively, indicating that the SLCV-induced epidemics were high intensity, monitored through changes in the severity of virosis (active and progressive infection) (Figure 5).

Figure 5 Percentage of severity of SLCV per quadrant in pumpkin var. Gray Zucchini plants, in different evaluations in Cocula, Guerrero. 

A pronounced gradient of the PS of SLCV was observed in the quadrants north and northeast of the study area, compared to the central and southeast parts that were little affected (Figure 6). In the contour plot, light areas indicate a lower intensity of severity, while dark areas indicate greater severity of virosis. An apparent shore effect attributable to the source of primary inoculum was perceived, perhaps by perennial weeds from the San Juan River bank; but no uniform distribution of the disease was detected in the culture.

Figure 6 Response surface map and plane of contour curves or isolines, of the dispersion and frequency of the severity distribution of the SLCV in pumpkin var. Gray Zucchini 

The distribution pattern of the SLCV in the study area revealed a strong spatial association between the severity of the virosis and the incidence of B. tabaci, since the first one increased in the western part during the crop cycle. Similar results reported a close spatial association between the incidence and severity of the virus in several cycles, which tended to increase towards a specific and localized part of the study area, during the spatial analysis and temporal change in the pandemic of a cassava geminivirus in northwestern Tanzania (^{Szyniszewska et al., 2017}).

In the present study, the spatial distribution of the incidence and severity of the SLCV provided information to explain the arrival and distribution of B. tabaci populations as a virus vector, from weeds present in areas adjacent to the San Juan River, where the humidity it is permanent and favors the permanent incidence of weeds that function as reservoirs of B. tabaci.

These observations have been supported by the report by ^{Mubin et al. (2009)}, who argue that weeds act as begomovirus reservoirs and are responsible for the appearance of virosis at the beginning of the crop cycle, becoming severe epidemics of cultivated plants; while, during the dry season, weeds play the crucial role in the spread of various viruses, they act in the redistribution of the primary inoculum and serve as a refuge for vector insects, which are an important part in the epidemiology of begomoviruses (^{Zaidi and Mansoor, 2017}).

In another study, it was mentioned that the most significant variation pattern in the incidence of whitefly was recorded among quadrants during the same season, in addition to the severity of a begomovirus, it is associated with environmental conditions, since during one of the cycles of experimentation, a fresh climate was presented and the severity of virosis was lower, compared to the cycle that had a warm and dry climate and therefore, the severity increased over the spatial and temporal distribution of cassava begomovirus (^{Szyniszewska et al., 2017}). The findings obtained in this research corroborate the results published by ^{Legg (2010)}, who confirmed that the changes in incidence of whitefly and the patterns of dissemination of virosis, are closely correlated.