Importance of cucurbits in the epidemiology of Papaya ringspot virus type P

Authors


E-mail: jrezende@usp.br

Abstract

Papaya ringspot virus type P (PRSV-P) systemically infects Carica papaya and species belonging to the family Cucurbitaceae. Attempts to recover PRSV-P from naturally infected cucurbit plants grown near or among diseased papaya trees have shown conflicting results worldwide. This study aimed to evaluate the natural infection of cucurbit species grown among and near papaya trees infected with PRSV-P in Brazil. Natural infection of cucurbits with PRSV-P occurred in zucchini squash but not in watermelon and cucumber. However, several attempts to recover PRSV-P from numerous Cucurbita pepo cv. Caserta (zucchini squash) plants grown 5–80 m from diseased papaya trees in the field failed. Mechanical inoculations of Cucurbita pepo cv. Caserta, Cucurbita maxima cv. Exposição (pumpkin), Cucumis sativus cv. Primepack Plus (cucumber) and Citrullus lanatus cv. Crimson Sweet (watermelon) with five Brazilian PRSV-P isolates showed that zucchini squash was the most susceptible species followed by watermelon and cucumber, while pumpkin was not infected. The results confirmed the variable susceptibility of cucurbit species to experimental and natural PRSV-P infection. Given these facts, the control of the disease through roguing should focus mainly on diseased papaya plants, as has been practised successfully in Brazil for many years, and on those cucurbits particularly known to be susceptible to natural infection with PRSV-P.

Introduction

Papaya ringspot is caused by Papaya ringspot virus type P (PRSV-P), from the genus Potyvirus, family Potyviridae, and is the most economically important disease of papaya (Carica (Ca.) papaya). The virus is classified into two serologically indistinguishable strains with distinct biological characteristics (Gonsalves & Ishii, 1980). The watermelon strain (PRSV-W) systemically infects species in the family Cucurbitaceae, and the papaya strain (PRSV-P) is capable of systemically infecting species in the families Cucurbitaceae and Caricaceae, papaya being the most important natural host (Barbosa & Paguio, 1982b; Gonsalves, 1998; Lima et al., 2001; Rezende & Martins, 2005). The susceptibility of cucurbits to the experimental transmission of PRSV-P depends on the origin of the virus isolate and the species/varieties evaluated (Capoor & Varma, 1958; Conover, 1964a; Sánchez De Luque & Martínez, 1976, 1977; Lana, 1980; Barbosa & Paguio, 1982a; Gonzáles et al., 2003). Both PRSV strains are transmitted in a non-persistent manner via several aphid species that do not colonize papaya (Purcifull et al., 1984; Gonsalves, 1998).

Controlling papaya ringspot disease is not easy. Several strategies have been investigated for effective and durable disease control. In Espírito Santo and Bahia, which are the major papaya-producing states in Brazil, the systematic roguing of diseased plants has been the single most effective strategy to control the disease over the last 25 years (Ventura et al., 2004). In addition, growers are advised to avoid planting cucurbits near or within papaya plantations because they are potential sources of PRSV-P inoculum (Adsuar, 1950; Costa et al., 1969; Sánchez De Luque & Martínez, 1977; Magdalita et al., 1990; Rezende & Martins, 2005; Chin et al., 2007). However, little is known about the actual importance of cucurbit plants in the epidemiology of papaya ringspot because of limited research efforts to study the natural infection of cucurbits and the contradictory results obtained (Conover, 1964b; Barbosa & Paguio, 1982b; Gonsalves, 1998). For example, biological and serological studies involving 454 cucurbit plants growing in northeastern Thailand near PRSV-P-infected papaya trees showed that 31% of the cucurbits were infected with PRSV-W (Gonsalves, 1998). Similar observations were made in Hawaii, USA (Gonsalves, 1998). In Florida, studies involving the recovery of PRSV-P from various cucurbits occurring near or among diseased papaya trees yielded negative results (Conover, 1964b). Barbosa & Paguio (1982b) did not recover PRSV-P from Momordica (Mo.) charantia, Cucurbita (Cr.) pepo and Melothria fluminensis located in or near papaya crops infected with PRSV-P in Pernambuco state, Brazil. However, Chin et al. (2007) recovered PRSV-P from Mo. charantia in Jamaica. In the Philippines, Magdalita et al. (1990) recovered PRSV-P from Diplocyclos palmatus plants located in abundance near papaya crops affected with the virus. Noa-Carrazana et al. (2006) detected PRSV-P in one sample of Cucurbita moschata present in a papaya field at Yucatán, Mexico. In Brazil, A. S. Costa (J. A. M. Rezende, ESALQ/USP, Brazil, personal communication) obtained PRSV-P from pumpkins that were present in an infected papaya orchard.

The aim of this work was to evaluate the natural infection of some species/varieties of cucurbits planted among or near to PRSV-P infected papaya plantations, in order to indirectly characterize the importance of those plants in the epidemiology of the disease. Moreover, this study aimed to examine the susceptibility of these same species/varieties to experimental infections with this potyvirus.

Materials and methods

Isolates of PRSV-P

Five isolates of PRSV-P obtained from Florianópolis (SC), Linhares (ES), Petrolina (PE), Piracicaba (SP) and Rinópolis (SP), Brazil, were maintained in papaya cv. Golden in the greenhouse and renewed through mechanical transmission to new plants every 3 months. To prepare the inoculum, PRSV-P infected leaves were ground in a mortar containing 0·02 m phosphate buffer, pH 7·0, and 0·02 m sodium sulphite, diluted 1:20 (w:v). Mechanical inoculations were performed on leaves previously dusted with silicon carbide abrasive (carborundum). The leaves of the negative control were mock-inoculated. Plants were maintained in a greenhouse until the evaluations. The analysis of partial nucleotide sequences of the coat protein (CP) gene was conducted to identify the virus isolates.

RT-PCR and nucleotide sequencing

Total RNA was extracted from infected papaya leaves using Trizol® LS (Invitrogen) according to the manufacturer’s protocol. The cDNA was synthesized using reverse transcription with 3 μL total RNA, 1 μL CP antisense primer (100 μm; 5′-AGGGCTACCCTCACTGTAAAATAG-3′), 1 μL 10 mm dNTP mix and 12 μL DEPC-treated water. This reaction was heated to 65°C for 5 min and then maintained at 2°C for 2 min. Subsequently, 5 μL of 5× reverse transcriptase buffer (250 mm Tris-HCl, pH 8·3, 375 mm KCl, 15 mm MgCl2), 2 μL 0·1 m dithiothreitol (DTT) and 200 units of M-MLV reverse transcriptase were added, and the reaction was incubated at 37°C for 50 min, 70°C for 15 min and then maintained at 4°C. The cDNA was used in a PCR reaction containing 3 μL cDNA, 2·5 μL 10× PCR buffer (600 mm Tris-SO4, pH 8·9, 180 mm ammonium sulphate), 1·5 μL 50 mm MgSO4, 0·25 μL CP sense primer (100 μm; 5′- CACATGTRTTTCACCAGT-3′), 0·25 μL CP antisense primer, 0·25 μL 10 mm dNTP mix, 1·2 units Taq DNA polymerase and 14·3 μL milli-Q water. The thermal cycler was programmed for 3 min at 94°C followed by 35 cycles of 94°C for 1 min, 54°C for 1 min and 72°C for 1 min, ending with 72°C for 10 min and subsequent cooling to 4°C. RT-PCR products were visualized on a 0·8% agarose gel stained with SYBR Safe™ DNA gel stain (Invitrogen).

The RT-PCR fragments for the CP gene were purified using the Wizard SV Gel and PCR Cleanup System (Promega) according to the manufacturer’s recommendations. The sequences were obtained using direct sequencing at Macrogen. The nucleotide sequences were aligned using multalin (Corpet, 1988; http://multalin.toulouse.inra.fr/multalin/) and electropherogram and quality analysis-phred-phrap (Togawa & Brigido, 2003; http://bioinformatica.cenargen.embrapa.br/phph/) to obtain the consensus sequences. The nucleotide and deduced amino acid consensus sequences were compared with the corresponding sequences of other isolates of PRSV-P deposited in the GenBank database (http://www.ncbi.nlm.nih.gov) using the blast program (Altschul et al., 1997).

PTA-ELISA

To confirm virus infection, plate-trapped antigen enzyme-linked immunosorbent assay (PTA-ELISA; Mowat & Dawson, 1987) was used, using polyclonal antiserum against PRSV-P virion. When necessary, PTA-ELISA was also used to detect the following viruses that can naturally infect cucurbit plants: Papaya ringspot virus type W (PRSV-W), Zucchini yellow mosaic virus (ZYMV) and Zucchini lethal chlorosis virus (ZLCV). Polyclonal antisera from the plant virology laboratory, ESALQ/USP, were used to detect PRSV-W and ZYMV, and Dr Alice K. Inoue-Nagata, Embrapa Vegetables, Brazil, kindly provided the antiserum against ZLCV. A 1:10 dilution of the extracts from healthy (negative control) and infected (positive control) plants were included in the tests in duplicate wells. The reaction was considered positive when the mean absorbance value was greater than three times the mean absorbance value for the extract of healthy plants.

Experimental inoculation of cucurbits with PRSV-P

The tests were conducted in a greenhouse in Piracicaba (SP), using zucchini squash (Cr. pepo cv. Caserta), pumpkin (Cucurbita maxima cv. Exposição), cucumber (Cucumis (Cm.) sativus cv. Primepack Plus) and watermelon (Citrullus (Ci.) lanatus cv. Crimson Sweet). Papaya plants (Ca. papaya cv. Golden) were included as the control for the infectivity of PRSV-P isolates. The plants were grown in pots containing a sterilized substrate composed of a mixture of soil and organic matter. The plants were regularly fertilized with approximately 0·5 g of ammonium sulphate per pot (2 L) and periodically sprayed with acaricides (avermectin B1 or 80% sulphur).

Cotyledons of cucurbit species were mechanically inoculated with the PRSV-P isolate and one plant from each species was mock-inoculated. Cotyledon samples were harvested from all plants at 7 days after inoculation (dai) for PRSV-P detection, using PTA-ELISA and biological recovery of the virus to papaya plants. Approximately 14–21 dai, the upper leaves of cucurbit plants were evaluated by observing the expression of symptoms and then subjected to virus detection using PTA-ELISA and biological recovery of the virus. The experiment was duplicated for all species/varieties of cucurbits and virus isolates.

The biological recovery test consisted of the mechanical inoculation of 30-day-old papaya plants with extracts from cucurbit plants inoculated with PRSV-P. The papaya plants were maintained in the greenhouse and evaluated for 30 dai through expression of symptoms and PTA-ELISA.

Natural infection of cucurbits planted within PRSV-P infected papaya plantations

The cucurbit plants were exposed to natural infection with PRSV-P in Piracicaba, Rinópolis and Linhares.

Piracicaba (SP)

Four experiments using different numbers of zucchini squash, cucumber and watermelon plants (Table 2) were performed in Piracicaba in the experimental field of the Department of Plant Pathology and Nematology, ESALQ/USP (Fig. 1) from April 2009 to August 2010. Cucurbit plants were grown among rows of papaya trees infected with PRSV-P in plots E and F (Fig. 1). The zucchini squash plants were obtained by directly sowing pits previously prepared with mineral fertilizer 04-14-08 (250 g per pit) and tanned cattle manure (200 g per pit), which were applied 15 days before planting. Spacing was 1 m between the plants. After germination, thinning was performed resulting in one plant per pit. Cucumber and watermelon seedlings were produced in polystyrene trays lined with a substrate of pine bark and transplanted 10 days after emergence in pits prepared in the manner described above. The plant spacing was 2 m.

Figure 1.

 Sketch of the experimental area of the Department of Plant Pathology and Nematology, ESALQ/USP, Piracicaba (SP), consisting of plots A to F, where the cucurbit plants were exposed to natural infection with PRSV-P.

At 40–60 days after exposure in the field, recently emerged leaves were collected from each plant separately. Extracts from individual samples or composites of three to five samples were mechanically inoculated to 30-day-old papaya plants for the biological recovery of PRSV-P. Composite samples were diluted (1:10) as individual samples. The papaya infection was evaluated as described above by observing the expression of symptoms and by using PTA-ELISA. Cucurbit samples were also tested using PTA-ELISA for the detection of PRSV-W, ZYMV and ZLCV.

Rinópolis (SP)

Three independent experiments were performed in commercial papaya orchards that were 100% infected with PRSV-P in the region of Rinópolis from April to November 2009. For these experiments, only zucchini squash cv. Caserta was planted among the rows of infected papaya plants. The evaluations were performed as described above.

Linhares (ES)

A similar experiment was conducted in an experimental field containing PRSV-P-infected papaya in the region of Linhares from 5 October to 20 November 2009. Evaluations were performed as described for the other experiments in Piracicaba and Rinópolis.

Natural infection of zucchini squash with PRSV-P planted at different distances from infected papaya trees

These experiments were conducted in the experimental field at the Department of Plant Pathology and Nematology, ESALQ/USP in Piracicaba (Fig. 1). Two experiments were conducted: from 30 March to 20 May 2007 and from 10 July to 20 September 2007. For each assay, 340 plants of zucchini squash cv. Caserta were grown in plot A located 5–25 m from 70 PRSV-P infected papaya plants grown in plot C (Fig. 1). The zucchini squash plants were obtained by direct seeding as described before. The plants were spaced 1 m apart. After germination, thinning was performed, leaving one plant per pit.

After approximately 60 days of exposure, recently emerged leaves from all plants were collected and grouped in sets of 10. The extract from each group (34 groups in total), diluted 1:10, was mechanically inoculated to three 30-day-old papaya plants cv. Golden in the greenhouse. Papaya plants were inoculated with PRSV-P isolates from infected papaya and used as controls. The infection of the papaya plants was evaluated as described before, by observing the expression of the symptoms and by using PTA-ELISA and RT-PCR for the detection of the CP gene.

Five other experiments were conducted from October 2009 to August 2010 using zucchini squash cv. Caserta plants grown in plots A, B, C and D at different distances (approximately 5, 15, 20, 35, 70, 75 and 80 m) from PRSV-P infected papaya plants located in plots E and F (Fig. 1). The plants were grown in the same way as previous experiments. The infection of zucchini squash plants was detected using biological recovery tests for PRSV-P in papaya plants followed by PTA-ELISA.

Monitoring of aphids

From May to November 2009, the aphids were monitored by capturing in a Moericke yellow pan trap. The trap was placed at a height of 18 cm from the soil in the centre of plots E or F, which were cultivated with PRSV-P infected papaya plants. The aphids were collected twice a week and preserved in 70% alcohol for subsequent quantitation and identification using a stereoscopic microscope. The identification of genera and/or species was based on the characteristics described by Jadot (1975) and Blackman & Eastop (2000). Aphids were also monitored in plots A and B from 28 February to 1 April 2010.

Results

Identity of PRSV-P isolates

DNA fragments of 863 bp (nucleotides 9258–10120), representing part of the CP gene of each PRSV-P isolate, were obtained using RT-PCR. The deduced proteins were 286 amino acids long. The analysis of the nucleotide sequences revealed 95–99% identity among the isolates of PRSV-P (JQ755424, JQ755425, JQ755426, JQ755427 and JQ755428) used in this work. When compared with similar sequences of PRSV-P isolates from different parts of the world (accession numbers HQ424465, AY162218, AF344644, AF344647, DQ339581, DQ104823, AY903266, AY841757, AF344642, U14738, AF344640, AF344645 and AF344643), the nucleotide and deduced amino acid sequences shared 88–99% and 88–100% identity, respectively (data not shown).

Among the cucurbit species tested, zucchini squash (Cr. pepo cv. Caserta) was the most susceptible, with 85–93% plants showing symptoms (Table 1). The watermelon (Ci. lanatus cv. Crimson Sweet) plants were also susceptible to infection with the same virus isolates. In the first experiment, 55% of the plants exhibited symptoms, and 100% of the plants exhibited symptoms in the second experiment. None of the hybrid cucumber (Cm. sativus cv. Primepack Plus) plants showed symptoms of virus infection in the first experiment, although the biological recovery of the virus was positive for 25% of the plants. In the second experiment, it was possible to identify cucumber plants with symptoms (40%) when inoculated with PRSV-P isolates from Linhares and Piracicaba. The infection was confirmed in many plants using PTA-ELISA and biological recovery of the virus to papaya plants. PRSV-P was not identified in pumpkin (Cr. maxima cv. Exposição) by monitoring symptom development, PTA-ELISA or biological recovery.

Table 1. Reaction of four varieties of cucurbit species to infection with five isolates of Papaya ringspot virus - type P (PRSV-P), by mechanical inoculation, evaluated using PTA-ELISA and the biological recovery of the virus to papaya plants
SpeciesIsolates of PRSV-PExperiment IExperiment II
Symp/Inoc.aPTA - ELISABRTSymp/Inoc.PTA - ELISABRT
CTLCTLCTLCTL
  1. C: cotyledons; TL: top leaves; nt: not tested; BRT: biological recovery tests to papaya plants.

  2. aNumber of plants with symptoms/number of inoculated plants.

  3. bComposite samples (3–5 plants).

Citrullus lanatus Florianópolis2/30/32/30/32/33/30/33/30/30/3
Linhares2/31/32/31/32/33/30/30/30/30/3
Petrolina0/60/60/60/60/63/30/33/30/31/3
Piracicaba3/30/33/30/31/33/30/33/30/30/3
Rinópolis3/30/33/30/33/33/30/32/30/32/3
Total10/181/1810/181/188/1815/150/1511/150/153/15
Cucumis sativus Florianópolis0/30/30/30/30/30/30/30/30/30/3
Linhares0/30/30/33/31/33/31/31/30/31/3
Petrolina0/80/80/80/82/80/30/30/32/30/3
Piracicaba0/30/30/32/30/33/33/33/30/31/3
Rinópolis0/30/30/30/32/30/30/30/31/30/3
Total0/200/200/205/205/206/154/154/153/152/15
Cucurbita pepo Florianópolis2/2nt2/2nt1/33/30/33/30/30/3
Linhares3/30/33/30/33/33/30/33/30/31/3
Petrolina6/90/96/91/2b7/92/30/32/31/33/3
Piracicaba3/30/33/31/33/33/30/33/30/32/3
Rinópolis3/30/30/33/33/32/2nt2/2nt1/2
Total17/200/1814/205/1117/2013/140/1113/141/127/14
Cucurbita maxima Florianópolis0/30/30/30/30/30/30/30/30/30/3
Linhares0/30/30/30/30/30/30/30/30/30/3
Petrolina0/110/110/110/110/110/30/30/30/30/3
Piracicaba0/30/30/30/30/30/30/30/30/30/3
Rinópolis0/30/30/30/30/30/30/30/30/30/3
Total0/230/230/230/230/230/150/150/150/150/15

In four independent experiments in a papaya orchard at ESALQ/USP, Piracicaba, PRSV-P was detected in the zucchini squash by bioassays (Table 2). The natural infection of zucchini squash was also observed in two out of three experiments in Rinópolis, but not in Linhares (Table 3). No cucumber and watermelon plants were infected with PRSV-P in Piracicaba.

Table 2. An assessment of the natural infection of three species of cucurbits planted among the rows of PRSV-P infected papaya plants in Piracicaba, SP, Brazil, through the biological recoverya of the virus to papaya and PTA-ELISA
ExperimentSpeciesNo. of plantsExposure periodNo. infected papaya/No. inoculatedPTA-ELISA
  1. aThe results from the biological recovery test for the period of 9 April to 2 June 2009 was conducted with composite samples of 3–5 plants.

1 Cucurbita pepo 459 April to 2 June 20096/96/9
Cucumis sativus 110/30/3
Citrullus lanatus  90/30/3
2 Cucurbita pepo 2718 Sept to 17 Nov 200915/2715/27
Cucumis sativus 200/200/20
Citrullus lanatus 120/120/12
3 Cucurbita pepo  616 Feb to 29 March 20100/60/6
Cucumis sativus 130/130/13
Citrullus lanatus  60/60/6
4 Cucurbita pepo  926 June to 8 Aug 20102/92/9
Cucumis sativus 130/130/13
Citrullus lanatus 100/100/10
Table 3. An assessment of the natural infection of zucchini squash (Cucurbita pepo cv. Caserta) planted among the rows of PRSV-P infected papaya trees in Rinópolis, SP and Linhares, ES, Brazil, by the biological recovery of the virus to papaya and PTA-ELISA
ExperimentLocalityNo. of plantsExposure periodNo. infected papaya/No. inoculatedPTA-ELISA
  1. aThe biological recovery test and PTA-ELISA were performed with the composite samples from three plants.

1Rinópolis-SP3020 April to 17 June 20090/10a0/10a
2Rinópolis-SP4605 June to 23 July 20096/12a6/12a
3Rinópolis-SP4005 Sept to 05 Nov 20093/403/40
4Linhares-ES1405 Oct to 20 Nov 20090/140/14

None of the leaf extracts from the 965 zucchini squash plants analysed in the composite (exposure periods 30 March to 20 May 2007 and 10 July to 20 September 2007) and individual (other exposure periods) samples caused infection in the papaya plants used for the virus recovery tests. None of the leaf extracts from inoculated papaya plants reacted with antiserum to PRSV-P in PTA-ELISA (Table 4) or RT-PCR (data not shown).

Table 4. Assessment of the natural infection of zucchini squash (Cucurbita pepo cv. Caserta) planted near to PRSV-P infected papaya plants in Piracicaba, SP, Brazil, through the biological recovery of the virus to papaya and PTA-ELISA
Distance from the source of inoculum (m)Exposure periodNo. of evaluated plantsNo. infected papaya/No. inoculatedPTA-ELISA
  1. aThe biological recovery test and PTA-ELISA were conducted using composite samples from 10 plants.

  2. bThe biological recovery test was performed using composite samples from five plants.

5–256 Oct to 1 Dec 2009400/8b0/8
15–1726 June to 8 Aug 201050/50/5
20–2226 June to 8 Aug 201030/30/3
15–3530 March to 20 May 20073400/34a0/34
15–3510 July to 20 Sept 20073400/34a0/34
35–556 Oct to 1 Dec 2009600/12b0/12
35–5518 Oct to 30 Nov 2009600/600/60
70–7516 Feb to 29 March 2010140/140/14
75–8022 Apr to 4 July 2010910/450/45
80–8216 Feb to 29 March 2010120/120/12

Fifteen pumpkin plants identified as cv. Paulistinha exhibiting mosaic and planted approximately 10 m away from the PRSV-P infected papaya trees in Rinópolis were also analysed for virus infection. The biological recovery tests for the papaya test plants indicated that only one pumpkin plant was infected with PRSV-P. Twelve other cucurbit plants from an unidentified cultivar, planted far away from the PRSV-P infected papaya orchards, were also analysed, and the result was negative. The potyviruses ZYMV and PRSV-W were detected using PTA-ELISA in pumpkin plants from both collections (data not shown).

The infection of cucurbit plants with the potyviruses PRSV-W and ZYMV, transmitted by aphids, varied from 16% to 42% and from 20% to 88%, respectively, during the different experiments. Natural infection with ZLCV, a tospovirus transmitted by Frankliniella zucchini, was also detected. Eleven genera and/or aphid species, some of them recognized as vectors of PRSV-P (Table 5), were collected from yellow water traps in plots E and F. Aphis spp., Liphaphis erysimi and Myzus persicae were the most abundant species captured. A total of 1266 and 687 aphids, belonging to the same species indicated in Table 5, were collected in traps placed in plots A and B.

Table 5. The total number of aphids captured during the exposure tests of cucurbit plants to infection with PRSV-P in the field at ESALQ/USP, Piracicaba, SP, Brazil, 2009
Genus/speciesMayJuneJulyAugSeptOctNov
  1. aThe species identified as efficient vectors of PRSV-P in Brazil.

Acyrthosiphum sp.1010000
Aphis spp.a4475149139147165
Liphaphis erysimi 40123404831
Macrosiphum spp.a0010031
Myzus persicaea 7171727217
Pentalonia nigronervosa 0111111
Rhopalosiphum spp.8120110
Toxoptera aurantii 1001150
Toxoptera citricidaa 20044110
Uroleucon spp.161211710
Other6094946489029
Total14320145122268328234

Discussion

The susceptibility of different cucurbit species to PRSV-P is widely known and has been reported in several studies conducted worldwide. The results of the present work concerning the experimental transmission of five isolates of PRSV-P from different regions of Brazil to four species of cucurbits (Table 1) are consistent with those reported in the Brazilian and international literature. The zucchini squash cv. Caserta was the most susceptible species to infection with the five isolates of the virus, followed by watermelon cv. Crimson Sweet and cucumber cv. Primepack Plus. The pumpkin cv. Exposição was resistant to infection with isolates of PRSV-P. In Venezuela, López (1972) also found that Cr. pepo cv. Early Prolific Straightneck was the most susceptible to infection with PRSV-P among the 10 mechanically inoculated species of cucurbits, but that Cr. maxima cv. Big Tom was resistant to infection with the potyvirus. Sánchez De Luque & Martínez (1976, 1977) reported high susceptibility of Cr. pepo (cvs Small Sugar and Senator) to mechanical inoculation with PRSV-P in Colombia. However, these authors found that the susceptibility to infection varied according to the method of inoculation, which was increased when the plants were inoculated by the aphid M. persicae.

The zucchini squash, cucumber and watermelon plants experimentally infected with the five Brazilian isolates of PRSV-P generally did not exhibit severe symptoms typically induced by PRSV-W such as mosaic, blisters and distortions (Purcifull et al., 1984). Notably, there was some remission of symptoms when the plants were older. The mild symptoms in several species of cucurbits infected with isolates of PRSV-P was also reported in experiments conducted in India (Capoor & Varma, 1958), Colombia (Sánchez De Luque & Martínez, 1976) and Brazil (Barbosa & Paguio, 1982a; J. A. M. Rezende, data not shown), although cases of severe symptoms in some species of cucurbits were also observed (Capoor & Varma, 1958; López, 1972).

The recovery of the different isolates of PRSV-P from experimentally infected cucurbit species to papaya plants through mechanical inoculation was possible in many cases. The recovery was always more efficient when using extracts from new leaves than in cotyledons inoculated with the virus (Table 1). The papaya plants developed severe symptoms of mosaic and leaf deformation, accompanied by oily streaks on the stem, which are characteristic symptoms of PRSV-P in this species.

When zucchini squash cv. Caserta, watermelon cv. Crimson Sweet and cucumber cv. Primepack Plus plants were grown among rows of PRSV-P-infected papaya plants in Piracicaba, only the zucchini squash plants became infected with PRSV-P. The infection was confirmed through virus recovery to the papaya plants by mechanical inoculation (Table 2). The susceptibility of zucchini squash to natural infection with PRSV-P was also observed in the experiments conducted in Rinópolis, but not in Linhares (Table 3). However, PRSV-P was not recovered from any cucumber or watermelon plants grown among the diseased papaya plants in Piracicaba, suggesting that these species appear to have field resistance to the potyvirus, because they were susceptible to infection through mechanical inoculation. Additional exposure tests of these two species in the field, followed by virus recovery through aphid vectors or specific molecular methods for the detection of PRSV-P, might confirm the presumed field resistance. When plants of zucchini squash cv. Caserta were grown at distances of 5–80 m from PRSV-P infected papaya plants in Piracicaba, the potyvirus was not recovered from the analysed plants (Table 4).

A large number of cucurbit plants exposed to natural infection in Piracicaba and Rinópolis were infected with PRSV-W and ZYMV, which were detected using PTA-ELISA (data not shown). The sources of PRSV-P inoculum were always present in these areas. Moreover, in the experimental field in Piracicaba, the occurrence of several species of aphids was confirmed during the period of some cucurbit exposure tests, including the species M. persicae, Aphis gossypii, Aphis fabae, Aphis coreopsidis and Toxoptera citricida, reported as vectors of PRSV-P in Brazil (Rezende & Martins, 2005). These data show that all cucurbit plants were continuously exposed to the inoculum of all three potyvirus (PRSV-P, PRSV-W and ZYMV) and the aphids that can transmit them.

The natural infection of cucurbits with PRSV-P assessed using virus recovery tests to papaya test plants, through mechanical inoculation, has been extremely variable, as reported by several authors (Conover, 1964b; Barbosa & Paguio, 1982b; Magdalita et al., 1990; Chin et al., 2007) and as demonstrated in the results obtained in this work. However, the degree of infection also depends on the cucurbit species and/or variety because the susceptibility of the cucurbits to PRSV-P is variable.

Considering the hypothesis that PRSV-P might have arisen from PRSV-W, as suggested by Bateson et al. (1994), experimentally demonstrated by Chen et al. (2008), and partially corroborated by Castillo et al. (2011), one can assume that the PRSV-P virus is widely adapted to papaya in all regions of the world where it occurs and does not need cucurbits for its perpetuation (Gonsalves, 1998). Therefore, these plants might have little importance as a source of inoculum of PRSV-P. Costa et al. (1969) and Barbosa & Paguio (1982b) showed that the primary source of inoculum of PRSV-P for new papaya crops is the infected papaya trees. A similar case was reported for the Cocoa swollen shoot virus (CSSV) in Ghana (Owusu, 1983). According to the author, the first epidemics originated from wild hosts of the virus. Once CSSV was spread widely throughout the cocoa plantations, the wild hosts were of little importance as sources of inoculum.

Data from this study confirm previous results of various authors on the variability in susceptibility of species/varieties of cucurbits to experimental infection with PRSV-P. However, this study clearly shows that susceptibility to natural infection with this potyvirus does not necessarily follow the experimental infection, because some species/varieties (cucumber and watermelon) showed field resistance, making them unimportant sources of inoculum for PRSV-P. Given these facts, the control of the disease through systematic roguing should focus mainly on diseased papaya plants, as has been practised successfully in Brazil for many years, and on those cucurbits particularly known to be susceptible to natural infection with PRSV-P.

Acknowledgements

The authors thank Associação Brasileira dos Exportadores de Papaya (BRAPEX) for supporting this work. This work was financially supported through grants from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação de Amparo à Pesquisa do Espírito Santo (FAPES) and the Financiadora de Estudos e Projetos (FINEP).

Ancillary