Three strains of bean yellow mosaic virus : symptoms and accumulation in eight pea cultivars ( . Pisum sativum L . )

A pea mosaic strain and a bean strain of bean yellow mosaic virus (BYMV) were isolated from naturally infected pea and broad bean plants and named BYMV-Ps and BYMV-Vf, respectively. A third strain of BYMV isolated from Gladiolus (BYMV-G) was obtained from Denmark which was distinguished from the two above strains serologically and by its symptoms in test plants. BYMV-Ps and BYMV-Vfcaused yellow mosaic symptoms and green mosaic symptoms, respectively, in eight pea cultivars tested, but the concentration of BYMV varied among the cultivars. BYMV-G caused mild mosaic or vein clearing in peas. A need to improve resistance to BYMV in the Finnish pea varieties was recognized.


Introduction
Bean yellow mosaic potyvirus (BYMV) occurs worldwide and infects several economically important legumes, including peas (Pisum sativum L.), beans (Phaseolus spp.) and broad beans (Vida faha L.) and non-leguminous plant species, such as those in the family Iridaceae and the genus Fagus. In legumes, BYMV is transmitted by several species of aphids in a non-persistent manner, and also experimentally in sap of infected plants (Bos 1970 a, Tapio 1970, Cockbain 1983, Hampton 1984, Jayasena and Randles 1985, Bays and Demski 1986, Winter and Nienhaus 1989, Provvidenti 1991. Many strains of BYMV have been reported which may be distinguished by host range, aphid transmission, serology or RNA sequence heterology (Bos 1970 a, Jones and Diachun 1977, Reddick and Barnett 1983, Schmidt and Zobywalski 1984, Herrera and Sepulveda 1986, Barnett et al. 1987, Scott et al. 1989, Hopps and Mclaughlin 1990. The pea strain of BYMV, formerly considered a distinct virus from the bean strain of BYMV and called pea mosaic virus (PMV), causes bright yellow mosaic symptoms, whereas the bean strain of BYMV causes green mosaic symptoms in pea leaves (Bos 1970b, Tapio 1970). Both strains of BYMV infect broad beans inducing similar symptoms and cytoplasmic and nuclear inclusions (Bos 1969, Tapio 1970, Cock-BAIN 1983. The bean and pea strains of BYMV occur naturally in Finland, and these strains may cause a yield loss of up to 62% and 30%, respectively, in peas depending upon the age at which the plants are infected (Tapio 1970 Kielpinski andBlixt 1982, Hovinen 1988, S. Hovinen pers. com.). However, breeding for resistance to viruses was not particularly emphasized, and only resistance to pea seed-bome mosaic virus has been incorporated into a few breeding lines (Hovinen 1990). Pea plants with yellow mosaic symptoms are commonly found in experimental fields of APBF (S. Hovinen pers. com.). In the present study, two strains of BYMV were isolated from naturally infected pea and broad bean plants. Eight pea cultivars were inoculated with the above Finnish BYMV isolates and one Danish BYMV isolate. These cultivars included some of the recently released varieties ofAPBF as well as one old Finnish and two Dutch cultivars commonly grown in Finland. The objective was to determine whether any resistance to BYMV from foreign genetic materials had been incorporated into new pea varieties by breeders in addition to the characters being deliberately selected.
Experiments were done in a screenhouse and a glasshouse under natural daylight at the University of Helsinki, Viikki (60°13'N) during August mid-September in 1992. Seeds were sown into a mixture of steam-sterilized peat and washed sand (10:1 v/v), and the plants were watered daily with a solution containing 0.1% NPK fertilizer (5-7-6). Plants in the screenhouse were sprayed weekly with dimethoate, and the glasshouse was fumigated weekly with nicotine to prevent any contamination by aphids. Daily means of the minimum and maximum temperatures were 10°C and 19°C in the screenhouse and 16°C and 28°C in the glasshouse.

Viruses
To obtain isolates of BYMV, leaves of pea and broad bean plants showing yellow and green mosaic symptoms were sampled in the experimental fields of the Anttila Plant Breeding Farm (60°25'N) on July 28, 1992. Batches of c. 20 -100 plants with bright yellow mosaic symptoms were found scattered in many pea fields (Fig. 1),and all the samples from those plants reacted strongly when tested in DAS-ELISA with the two antisera available to BYMV. The virus isolate from the pea breeding line Hja 57839 which gave the highest ELISA absorbance values was selected for further studies and named BYMV-Ps. Similarly, the virus isolate from the broad bean breeding line Hja 62029 A, which produced the highest ELISA absorbance values for BYMV, was used in further experiments and named BYMV-Vf. One isolate of BYMV obtained from Gladiolus in Denmark by Dr. N. Paludan (isolate DK-17-3-4: Boye et al. 1990, Albrecht-SEN et al. 1991was kindly provided in freeze-dried leaves ofNicotiana benthamiana Domin. by Dr. M. Albrechtsen, Danish Research Center for Plant Protection, Lyngby, Denmark, and this is referred to as BYMV-G. BYMV-Ps and BYMV-G were maintained in the pea cv. Kalle, and BYMV-Vf was maintained in the broad bean cv. Ukko.

Antisera
Polyclonal antibodies raised in rabbits against BYMV-G and alkaline phosphatase (AP) conjugated antibodies were kindly supplied by Dr. M.
Albrechtsen, and were used in all serological tests. Rabbit polyclonal antibodies (ATCC PVAS-368 4-85) to BYMV (Hammond and Hammond 1989) were obtained from the American Type Culture Collection (ATCC), Rockville, USA, and conjugated with AP at our laboratory. The BYMV antibodies from M. Albrechtsen and ATCC were used in parallel ELISA tests for the detection of BYMV in the field samples.
Inoculations BYMV was sap-inoculated by grinding 1 g of BYMV-infected leaves in 5 ml of 0.001 M phosphate buffer, pH = 7.5, and rubbing the sap onto carborundum-dusted stipules of pea seedlings and leaves of the other test plant species when the plants were 10 days old. Four plants each of the eight pea varieties and test plant species were inoculated with each of the three virus isolates in two replicate experiments.
Virus detection BYMV was detected by double antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) (Clark 1981) and by immunosorbent electron microscopy (ISEM) (Roberts and Harrison 1979). Stipules or leaves of the three uppermost knots of the pea plants and the uppermost fully-expanded leaves of the other test plant species were sampled in duplicate. Samples were weighed and ground in four volumes (w/v) of extraction buffer for DAS-ELISA and in nine volumes (w/v) of 0.06 M phosphate buffer, pH = 6.5, for ISEM. In DAS-ELISA, the absorbances were recorded at 405 nm (A 405) using the ELISA reader (Titertek Multiscan) after 45 min of incubation with the substrate p-nitrophenyl. Inclusion bodies of BYMV were stained by immersing epidermal strips from the undersides of leaves of V. faba cv. Ukko in a solution containing 0.5% Trypan blue and 0.9% NaCl.

Statistical analysis
Analysis of variance was used for statistical analysis, and calculations of the least significant differences (LSD) of the A405 values were made where appropriate (Steel and Torrie 1981).

Results
The isolates BYMV-G, BYMV-Vf and BYMV-Ps were distinguished by the symptoms they caused in test plants (Table 1). In all the pea cultivars tested, BYMV-Ps caused bright yellow mosaic symptoms, whereas BYMV-Vf caused mainly green mosaic and sometimes mild yellow mosaic symptoms (Fig.  2). However, BYMV-G caused only very mild mosaic symptoms and vein clearing. Similarly, BYMV-Vf and BYMV-Ps caused severe green mosaic symptoms in the broad bean cv. Ukko, and the symptoms of BYMV-Vf were visible sooner than those of BYMV-Ps. In contrast, broad bean plants infected with BYMV-G became slightly pale and showed no other symptoms. BYMV-Vf and BYMV-Ps infected Trifolium pratense cv. Bjursele    2. Mean absorbance values from ELISA for detection of BYMV in plants each of eight pea cultivars grown in a screenhouse (a) and glasshouse (b) and also in P. vulgaris cv. Dufrix grown in a glasshouse (b) 18 days post-inoculation. The BYMV isolates were from Gladiolus (G), Vida faha (Vf) and Pisum sativum (Ps). The least significant differences are given at the risk level of 1% (LSDo.oi causing systemic vein chlorosis and yellowing, but no infection with BYMV-G was detected. The leaves of P. vulgaris L. cv. Dufrix inoculated with BYMV-G produced local ringspots and vein necrosis, but those inoculated with BYMV-Vf or BYMV-Ps remained symptomless. All the BYMV isolates infected Lupinus luteus L. causing systemic green mosaic symptoms, narrowing of leaflets and severe stunting. In Nicotiana tabacum L. cv. Samsun, all the BYMV isolates caused local symptomless infection which was detected by ELISA, but no systemic infection. The mean lengths of 50 particles of the BYMV-G, BYMV-Vf and BYMVPs isolates were 781 nm, 778 nm and 816 nm, respectively, in sap prepared for ISEM from systemically infected leaves of P. Ukko with all the BYMV isolates (Fig. 3).
The A405 values for the detection of BYMV-Vf and BYMV-Ps differed significantly between many pea cultivars (Table 2). However, symptoms among the cultivars infected with BYMV-Vf or BYMVPs were similar. Thus, the virus litres did not correlate with the severity of symptoms. The concentration of BYMV-G was clearly increased, whereas the concentration ofBYMV-Ps was slightly depressed, in pea plants grown in the higher glasshouse temperatures compared to plants grown in the cool screenhouse. In most of the plant species tested, the A405 values from ELISA for the detection of BYMV-G were lower than those for the other two BYMV isolates. However, sap-inoculation of BYMV-G and BYMV-Ps from L. luteus, pea cv. Kalle and P. vulgaris cv. Dufrix to C. amaranticolor Coste et Reyn resulted in similar numbers of local lesions, indicating that the litres of infective virus were similar.

Discussion
The isolates BYMV-G, BYMV-Vf and BYMV-Ps were clearly distinguished from each other by the symptoms they caused in various test plant species. The symptoms in pea plants and the slightly longer particles of the BYMV-Ps isolate were similar to those of the pea mosaic strains of BYMV previously described, and the symptoms caused by BYMV-Vf in the test plants resembled those described for the bean strain ofBYMV (BOS 1970a,b, Tapio 1970. As with other potyviruses, the length of BYMV particles in plant extracts is affected by the concentration of magnesium ions in the extraction buffer (Govier and Woods 1971). On the other hand, BYMV-G differed from the other two BYMV isolates by its milder symptoms in most of the test plant species, and also serologically, as has been reported previously for other BYMV isolates (JONES and Diachun 1977). However, all the isolates reacted with the antibodies raised against BYMV (obtained from M. Albrechtsen and ATCC), but not with those raised against bean common mosaic virus (obtained from L. Bos and D.Z. Maat, Institute for Phytopathological Research (IPO), Wageningen, The Netherlands; data not shown). The intranuclear inclusion bodies produced by all the virus isolates were identical and similar to those previously described for BYMV (BOS 1969, CHANG et al. 1988. Therefore, all the virus isolates represented distinct strains ofa single vims species, namely BYMV.
All the recently released pea cultivars of APBF included in the present study were susceptible to BYMV, suggesting that little or no improvement in resistance to BYMV has occured during the last decades and since the previous studies by Tapio (1970). However, yield losses caused by BYMV in pea are potentially high in Finland, and the virus is readily spread by aphids from forage legumes to peas in the field, red clover acting as the main vims reservoir (Tapio 1970). Globally, BYMV is considered one of the most important pathogens ofpea, particularly in older cultivars lacking resistance to BYMV (Hampton 1984). Aphicides are ineffective in decreasing the spread of the nonpersistently transmitted viruses such as BYMV, but treatment of plants with mineral oil may be more successful (Cockbain 1983, Jayasena andRandles 1985). However, resistance to BYMV in pea remains to be developed for virus control in the field. Indeed, since 1965 the recessive gene pair mo/mo conferring resistance to BYMV in P. sativum has been routinely incorporated into new pea cultivars in many pea breeding programmes abroad, which is the primary reason to consider BYMV as a minor pathogen ofpea at the present (Marx andProvvi-DENTi 1979, Hampton 1984). As the domestic cultivation of pea is limited in Finland (7700 ha in , Hovinen 1990, breeding of peas in Finland should meet the requirements of an international market area to strengthen the economic basis of the breeding programme. Lack of resistance to viruses may be one of the obstacles reducing the adaptability of Finnish pea varieties for cultivation abroad. Genetically engineered virus resistance, such as coat protein mediated resistance to BYMV, is also becoming available in peas in the future (Hammond and Kamo 1991, Cecchini et al. 1992, Hull and Davies 1992.