Distribution of Citrus Tristeza Virus in Grapefruit and Sweet Orange in Florida and South Africa*, ** R

Home , Citrus tristeza virus

Distribution of Citrus Tristeza Virus in Grapefruit and Sweet Orange in Florida and South Africa*, ** R. F. Lee, S. M. Garnsey, L. J. Marais, J. N. Moll, and C. 0. Youtsey

ABSTRACT. The distribution of citrus tristeza virus (CTV) in field trees was determined by enzyme-linked immunosorbent assay on individual flushes collected from citrus trees in Florida and South Africa. In Florida, most CTV isolates were unevenly distributed in grapefruit trees, especially in late summer. However, one mild isolate with some cross-protecting ability (T26) was consistently distributed throughout the trees regardless of season. Citrus tristeza virus was more evenly distributed in sweet orange trees in Florida, but occasionally very young flush tissue was found CTV-free. In South Africa, CTV was found evenly distributed throughout both grapefruit and sweet orange trees. South African isolates were highly invasive in recently inoculated plants. The even distribution of CTV within trees is a trait which may be important for effective cross protection by mild CTV strains. Index words. ELISA, cross protection, stem pitting.

Citrus tristeza virus (CTV), a of effective cross protection. Selection member of the closterovirus group, is of cross-protecting CTV strains has the most economically important cit- thus far been an empirical process. rus virus worldwide (1). There are The distribution of the mild strain of many strains of CTV which differ in virus within the plant host may play their biological activity (1, 3). Quick an important role in its ability to cross decline of trees on sour orange protect the host against subsequent rootstock induced by CTV has killed challenge by severe isolates (9). millions of trees worldwide, but this We used the enzyme-linked im- aspect of tristeza can be controlled by munosorbent assay (ELISA) to study the use of CTV-tolerant rootstocks. the distribution of CTV within Strains of CTV which cause severe grapefruit and sweet orange trees in stem pitting on scions cause losses Florida where severe stem-pitting even in trees on CTV-tolerant strains of CTV are not yet present rootstocks. For these severe stem pit- and in trees in South Africa where se- ting strains of CTV, the use of mild vere stem pitting strains are endemic. strains for cross protection is proba- Results are presented in this paper bly the most effective control strat- and discussed in relation to cross pro- egy (5, 7). With mild strain cross protection success. tection, a plant previously infected with a mild virus strain will show in- METHODS AND MATERIALS terference or delay in the expression Virus isolates. For some experi- of severe symptoms after being chal- ments, Marsh grapefruit trees on lenge-inoculated with a severe strain sour orange rootstock were inocu- of the same virus. Severe stem-pit- lated with the previously charac- ting strains of CTV have been con- terized CTV isolates T3, T4, or T26 trolled successfully by use of mild (11). These trees were in an experi- strain cross protection in Brazil (5) mental field plot in central Florida. and in South Africa (7). Isolate T3 causes decline of sweet While mild CTV strains are com- orange on sour orange rootstock and mon, few of these strains are capable causes seedling yellows (SY). Isolate

*Florida Agricultural Experiment Station Journal Series No. 7952. **Mention of a trademark, warranty, proprietary product, or vendor does not constitute a guaran- tee by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products or vendors that may also be suitable. Tenth IOCV Conference

T4 causes strong stunting, stem pit- tember, and February. Similar re- ting and vein clearing in Mexican petitive assays were made from four lime, but no decline of sweet orange sweet orange trees. Some trees in- on sour orange and no SY. Isolates fected with naturally-occurring CTV T26, T32, and T55 produce mild strains were sampled only one time symptoms and little stunting on Mex- by collecting the indicated number of ican lime, and cause no decline of flushes from throughout the tree sweet orange on sour orange, or SY. canopy. Root sprouts were forced by Isolate Ti33 is an uncharacterized exposing and cutting some of the CTV field strain which produces mod- roots in an 80-year-old grove of Marsh erate vein clearing and stunting on grapefruit on rough lemon rootstock Mexican lime. located in central Florida. About 8 The Nartia mild, Bolton severe, weeks later, new flush tissue was col- and Nkwalini mild CTV isolates from lected from the forced root sprouts South African have been previously and also from new flushes on the described (6, 10). The CSFRI CTV grapefruit scions. strains from South Africa were natur- Tissue sources-South Africa. ally-occurring field selections which Twenty-five flushes were collected did not cause decline of sweet orange from individual trees from ll-year-old on sour orange rootstock. Isolate Nartia grapefruit in the Nkwalini GFSS-1 is a severe strain of CTV Valley, Natal Province, and from 10- which severely stunts grapefruit, and year-old Rose grapefruit propagated GFMS-10 produces moderate stem from the Bolton budwood source. pitting and mild stunting on grape- Samples were also collected from 5- fruit (12). year-old seedling Marsh grapefruit Sample collection and enzyme trees on an experimental plot near linked immunosorbent assay Malelane, Transvaal Province. (ELISA). Samples of young flush Twenty-five samples were collected which had the basal leaves at or near from each of five trees infected with full expansion were collected from three different CTV isolates in the field trees and stored in plastic bags latter location. at 4 C until processed for ELISA, A greenhouse experiment was set usually within 1-7 days. A 0.25-gm up in South Africa to determine the aliquot of chopped bark tissue was uniformity of CTV infection within homogenized in 5.0 ml of 0.05 M Tris- the inoculated plants 4 weeks after HCl buffer, pH 8.0. The double anti- graft inoculation. Two large Marsh body sandwich ELISA procedure (2) grapefruit seedlings and two Mexican was used with antisera prepared lime plants were inoculated using leaf against unfixed CTV (4). These anti- pieces infected with Florida isolates sera have reacted to all CTV isolates T55 and T32 and South African iso- tested. Healthy tissue gave OD405 lates Nartia, GFMS-10 and GFSS-1. readings of 0.02-0.03, and values After 4 weeks, the young growth twice this were considered as the from four different branches of each threshold for a CTV-positive reac- plant was collected for assay. tion. Most OD values were 0.4 or greater; only rarely were OD values RESULTS less than 0.4 in the samples declared CTV positive. Distribution of CTV within Tissue sources-Florida. Thirty grapefruit trees in Florida. In a pre- individual branches were labeled on liminary test, 30 flushes were col- Marsh grapefruit trees systemically lected from each of two 12-year-old infected with CTV isolates T3, T4, Marsh grapefruit trees on rough T26, and Ti33. Samples of new flush lemon rootstock during the spring and tissue were collected from the same were assayed by ELISA. Only eight individual branches in July, Sep- of the flushes from one tree and 12 Tristexa and Related Diseases

from the other reacted positively for In another study, 22 Pineapple CTV. In a test of the scions and the sweet orange trees on sour orange rootsprouts from an 80-year-old grove rootstock were sampled at three in- of Marsh grapefruit on rough lemon tervals. The first collection was Janu- rootstock, CTV was present in both ary 1985, after a severe freeze had de- the scion and rootsprouts of 21 trees, foliated the trees and the new flush in only the scion of seven trees, and was about 2-4 cm long. Only eight of in only the rootsprout of six trees. It the 22 trees were found CTV positive was not detected in either scion or using a sampling of 25 flushes from rootstock of nine trees. each tree. Subsequent collections of The apparent erratic distribution five flushes from each tree were made of CTV within grapefruit trees was in April 1985 and August 1986, and examined more closely by individually all flushes from all trees tested posi- labeling 30 separate branches on four tively for CTV infection. 6-year-old Marsh grapefruit trees, Distribution of CTV in grape- each inoculated with a different CTV fruit and sweet orange in South Af- strain. Young flush tissue was col- rica. Several grapefruit and sweet lected for ELISA from each tagged orange trees were tested for the pres- branch in summer (July), late summer ence of CTV by collecting 25 flushes (September), and during spring flush from around the canopy of each tree. (February) (table 1). Isolate T26 was A list of the trees tested and the sev- well distributed throughout the tree erity of the CTV symptoms is given with 90% of the branches positive on in table 3. All individual flushes from all three sampling dates. Isolate Ti33 all trees tested in South Africa as- was very unevenly distributed sayed positively for CTV. throughout the tree. Only 53% of the The movement of Florida and branches indexed CTV positive in South African CTV isolates into July, while 37% were positive in Sep- freshly inoculated plants was deter- tember, and 83% were positive in mined by testing the new flush tissue February. Only 13% of the tagged from plants 4 weeks after graft inocu- branches were CTV positive on all lation. All South African isolates were three sampling dates, whereas 10% of detected in all flushes from grapefruit the branches were CTV negative on and Mexican lime seedlings. The all three sampling dates. Isolates T4 Florida isolates T55 and T32 were and T3 were intermediate in their dis- present in all Mexican lime flushes 4 tribution. weeks after inoculation. However, Distribution of CTV in sweet isolate T55 was detected in only three orange trees in Florida. Samples of four flushes from each grapefruit were collected from 30 separate new plant, and T32 was detected in two of flushes on four naturally CTV-in- four and one of four flushes, respec- fected Hamlin sweet orange budwood tively, from each of two grapefruit source trees on rough lemon plants tested. rootstock. Branches were tagged and samples collected in late spring DISCUSSION - (April), in summer (July), and fall (September) (table 2). Citrus tristeza The distribution of the mild strain virus was well distributed within within the plant is important when these trees on all sampling dates. considering mild strain cross protec- From 90 to 100% of the individual tion as a control strategy for CTV (9). branches assayed positively for CTV The virus must distribute itself in all infection on all three sampling dates. parts of the plant and have the ability In a separate test, 30 flushes from to rapidly invade new growth flushes. two 12-year-old Valencia trees on Any part of the plant which is virus- rough lemon rootstock were sampled free, even temporarily, provides an in June. All 30 flushes from both trees opportunity for an aphid to infect the tested positively for CTV infection. plant with a severe CTV strain which, 36 Tenth ZOCV Conference

TABLE 1 DISTRIBUTION OF DIFFERENT CITRUS TRISTEZA VIRUS (CTV) ISOLATES IN GRAPEFRUIT TREES DURING DIFFERENT TIMES OF THE YEAR IN FLORIDA

Percent of branchesY Flushes infected with CTV CTV Always Always isolate July Sep. Feb. positive negative

Ti33 16/30" 11130 25130 13 10 T3 20130 18/30 27/30 43 3 T4 23130 25130 30130 60 0 T26 30130 27130 30130 90 0

"Number of flushes infected with CTVItotal number of flushes samples. Presence of CTV in tissue was determined by ELISA. YNew flush was collected at each period from the same labeled branch.

given the advantage of starting in The severity and titer of CTV is virus-free tissue, could ultimately re- influenced by temperature. More se- sult in the breakdown of cross protec- vere symptoms are produced in cooler tion. temperatures, and extremely hot Most CTV strains tested from growing conditions reduces CTV titer Florida were unevenly distributed and can result in temporary ther- within grapefruit trees (table I), but motherapy (8, Lee, unpublished well distributed within sweet orange data). This is consistent with our find- trees (table 2). While most Florida ing that the distribution of CTV CTV strains were unevenly distri- within grapefruit trees was poorest in buted within grapefruit, isolate T26 September (table 1). At this sampling was present in 90% of the same date, the mean daily temperature had branches on three different sampling been 30 C for 3-4 months and the dates which included summer and overnight low has not been below 21 winter conditions in Florida (table 1). C for the same period. Isolate T26 appears to have cross-pro- Finding flush from 14 of 22 tecting potential in Mexican lime Pineapple sweet orange trees in plants in field tests in Hawaii, where Florida CTV negative just after de- severe stem-pitting strains of CTV foliation by a freeze was unexpected. and the efficient CTV aphid vector, Although these trees had not been Toxoptera dtricida Kirk are present sampled prior to the freeze, all were (Garnsey & Yokomi, unpublished presumed CTV infected and this was data). confirmed by positive tests for all trees in later sampling. While severe TABLE 2 freezes have been rare in Florida, DISTRIBUTION OF CITRUS TRISTEZA there may be a correlation between VIRUS (CTV) WITHIN SWEET ORANGE defoliation of field trees and periodic TREES IN FLORIDA DURING DIFFER- ENT TIMES OF THE YEAR epidemics of CTV-induced decline 1-2 years later. More research needs to Flushes infected with CTV be done on virus distribution within Tree code April July Sep. recently defoliated trees. The erratic distribution of CTV BA7 30130" 30130 29130 within grapefruit trees in Florida is BA9 29130 30130 29130 in contrast to the uniform distribution BA12 30130 30130 30130 BA25 30130 29/30 27130 of CTV in grapefruit observed in South Africa. This uniform distribu- "Number of flushes infected with CTVItotal tion of South African CTV isolates number of flushes sampled. Presence of CTV was also apparent in the greenhouse in tissue was determined by ELISA. Flushes were collected from the same labeled branches tests where Mexican lime and grape- at each period. fruit seedlings, inoculated with differ- Tristeza and Related Diseases 37

TABLE 3 DISTRIBUTION OF CITRUS form distribution of South African TRISTEZA VIRUS (CTV) WITHIN CTV isolates may reflect natural GRAPEFRUIT AND SWEET ORANGE IN SOUTH AFRICA selection over many years for highly invasive isolates. In this area, severe IsolateY CTV CTV isolates are endemic, efficient Tree code severity infection aphid vectors are present, and trees are naturally challenged repeatedly Nartia grapefruit Bedlane M8 Mild by different sources. The erratic dis- M1 Mild tribution of CTV isolates from S3 Severe Florida, particularly in grapefruit Rose grapefruit trees, suggests that problems may be Bolton R4T7 Severe encountered, if at some future date, R6T9 Severe the citrus industry in Florida has to R7T6 Mild rely on these existing mild strains for Marsh grapefruit sdlgs. protection against severe stem-pit- Nartia 1 Mild ting strains of CTV such as those 2 Mild 3 Mild present in South Africa. Further 4 Mild selection of protecting strains for 5 Mild Florida conditions is indicated. Nkwalini 1 Mild All of the characteristics of CTV 2 Mild isolates associated with a good cross 3 Mild protecting ability are 'still not known. 4 Mild A mild selection from an invasive iso- 5 Mild late or from a mild isolate obtained Bolton 1 Severe from a mutagenized invasive isolate 2 Severe may be good cross protecting candi- 3 Severe 6 Severe dates. While thorough distribution 7 Severe within the protected plant seems im- Valencia/sour orange portant, it may not be the only consid- CSFRI R8T14 Mild eration. For example, CTV is well R7T12 Mild distributed in sweet orange in Mild Florida, yet decline apparently occurs -- "Number of flushes infected with CTVItotal readily when plants infected with number of flushes tested. Presence of CTV many mild isolates are challenged by was determined by ELISA. severe decline inducing isolates. YIsolateseverity as determined by stem pitting induced on grapefruit. ACKNOWLEDGMENTS ent CTV isolates, assayed uniformly The technical assistance of H. La- for the presence of CTV in flushes 4 sater and M. Washington is gratefully weeks after inoculation. These results acknowledged. This research was also indicated that differences in field supported in part by Grant No. US- assays between Florida and South Af- 874-84 from United States-Israel Bi- rica were not solely due to environ- national Agricultural Research and ment. This inherent rapid and uni- Development Fund.

LITERATURE CITED

1. BarJoseph, M., S. M. Garnsey, and D. Gonsalves 1979. The closteroviruses: a distinct group of elongated plant viruses. Adv. Virus Res. 25: 93-168. 2. BarJoseph, M., S. M. Garnsey, D. Gonsalves, M. Moscovitz, D. E. Purcifull, M. F. Clark and G. Loebenstein 1979.The use of eyzyme-linked immunosorbent assay for detection of citrus tristeza virus. Phytopathology 69: 190-194. 38 Tenth IOCV Conference

BarJoseph, M., C. N. Roistacher, S. M. Garnsey, and D. J. Gumpf 1981. A review on tristeza, an ongoing threat to citriculture. 1981 Proc. Int. Soc. Citricul- ture 1: 419423. Brlansky, R. H., S. M. Garnsey, R. F. Lee, and D. E. Purcifull 1984. Application of citrus tristeza virus antisera in labeled antibody, immuno-electron microscopical, and sodium dodecyl sulfate-immunodiffusiontests. Pages 337-342. In Proc. 9th Conf. IOCV. IOCV, Riverside, CA. Costa, A. S., and G. W. Muller 1980. Tristeza control by cross protection: a U.S.-Brazil cooperative success.Plant Dis. 64: 538-541. da Graca, J. V., L. J. Marais, and L. A. von Broembsen 1984. Severe tristeza stem pitting decline of young grapefruit in South Africa. Pages 62-64. In Proc. 9th Conf. IOCV. IOCV, Riverside, CA. de Lange, J. H., S. P. van Vuuren, and G. S. Bredell 1981. Groei-vunt entine- suiver sitrusklone vir die su~ervlantskema- van virusse. Subtrovica 2(5): 11-16. Garnsey, S. M., R. G. Christie, K. S. Derrick, and M. BarJoseph 1980. Detection of citrus tristeza virus. 11. Light and electron microscopy of inclusions and viral particles. Pages 9-16. In Proc. 8th Conf. IOCV. IOCV, Riverside, CA. Lee, R. F., R. H. Brlansky, S. M. Garnsey, and R. K. Yokomi 1987. Traits of citrus tristeza virus important for mild strain cross protection of citrus: the Florida approach. Phytophylactica 19: 215-218. Marais, L. J., R. F. Lee, J. V. da Graca, and J. M. Kotze 1988. Environmental effect on Marsh grapefruit (Citms paradisi Macf.) infected with different isolates of citrus tristeza virus. Phytophylactica 19: in press. Rosner, A., R. F. Lee, and M. BarJoseph 1986. Differential hybridization with cloned cDNA sequences for detecting a specific isolate of citrus tristeza virus. Phytopathology 76: 820-824. van Vuuren, S. P., and J. N. Moll 1987. Glasshouse evaluation of citrus tristeza virus isolates. Phytophylactica 19: 219-221.