Spread of Citrus Tristeza Virus in a Heavily Infested Citrus Area in Spain

P. Moreno, J. Piquer, J. A. Pina, J. Juarez and M. Cambra

ABSTRACT. Spread of Citrus Tristeza Virus (CTV) in a heavily infested citrus area in Southern Valencia (Spain) has been monitored since 1981. Two adjacent plots with 400 trees each were selected and tested yearly by ELISA (enzyme-linked immunosorbent assay). One of them was planted to 4-yr-old Newhall navel orange on Troyer citrange and the other to 8-yr-old Marsh seedless grapefruit on the same rootstock. Both had been established using virus-free budwood. In 1981, 98.7% of the Newhall navel plants indexed CTV-positive and by 1984 all of them were infected, whereas only 17.8% of the Marsh grapefruit indexed CTV-positive in 1981, and 42.5% were infected in 1986. This is an indication that grapefruit is less susceptible than navel orange to tristeza infection under the Spanish field conditions. Wild plants of 66 species collected in the same heavily tristeza-infested area were also tested by ELISA to find a possible alternate non-citrus host. CTV was not found in any of the more than 450 plants analyzed. Index words. virus spread, ELISA, noncitrus hosts.

Tristeza was first detected in virus diffusion under various environ- Spain in 1957 and since then has mental conditions. In this paper we caused the death of about 10 million present data on CTV spread in a trees grafted on sour orange and the heavily infested area. A survey progressive decline of an additional among wild plants growing in the several thousand hectares of citrus on same citrus area was also undertaken this rootstock. The disease was ini- to search for some noncitrus natural tially observed at the Ribera Alta hosts of CTV. area (Southern Valencia), but new foci, probably originating through un- MATERIALS AND METHODS controlled movement of infected bud- wood, successively appeared in other In 1981, two adjacent plots with spots. At present, tristeza can be 400 trees each were selected at El found with variable incidence in most Realengo, a citrus farm located at the of the citrus areas (2). Ribera Alta (Southern Valencia). This Three aphid species (Aphis gos- farm is within the area where tristeza sypii Glover, Aphis citricola Van der was first detected. One of them was Goot and Toxoptera aurantii (Boyer planted to 4-yr-old Newhall navel de Fonscolombe)) have enabled trans- orange on Troyer citrange and the mission of CTV in Spain, with vari- other to &yr-old Marsh seedless able efficiency depending on virus iso- grapefruit on the same rootstock. late and vector species (4). These Both of them had been established aphids, particularly A. gossypii and using virus-free budwood. A. citricola, are thought to be respon- Tristeza spread was monitored by sible for natural spread of tristeza (4) yearly indexing all trees by ELISA. from the new foci established through Each tree was sampled taking two infected budwood. young shoots of the last flush from op- Decline of trees in tristeza-in- posite sides of the tree. Shoots were fested orchards has been occurring at trimmed, the clippings placed into a variable rate depending on years plastic tubes and homogenized in ca. and citrus areas (6). Nevertheless, no 10 volumes of an extraction buffer data on the actual rate of virus spread using a Polytron Kinematics homog- under field conditions in Spain are enizer. Extracts were tested by a available. standard ELISA double antibody In 1981 several plots were selected sandwich procedure (1,ll) using CTV in different citrus areas to monitor antisera (879 or R4, kindly provided Tenth ZOCV Conference by Dr. S. M. Garnsey and Dr. D. J. plots were also sampled and tested by Gumpf, respectively) or monoclonal ELISA in the same conditions as cit- antibodies specific to CTV obtained as rus trees. In some cases, extracts of described elsewhere (11, 12). wild plants were also examined by im- The rate of CTV increase was munoelectron microscopy (IEM) as analyzed using the expressions Ln (XI described by Garnsey et al. (3). 1-x) or Ln (111-x) (lo), where x is the per unit incidence of the disease at a RESULTS given time. Tristeza incidence around the ex- The amount and distribution of in- perimental plots was estimated by oculum around the monitored plots is random sampling and testing by shown in figure 1. At the beginning of ELISA 2.5% of the citrus trees grown the experiment, the proportion of within a band 150 m wide around the CTV-infected trees in the 150 m band plots (120 trees in each case). CTV surrounding the grapefruit plot was incidence in citrus plantings of differ- 69 * 9% (P 5 0.05). Corresponding ent age or scion-rootstock combina- values for the navel plot were 70 r tion was also estimated separately. 9% respectively. Wild plants of a number of species Variation of CTV incidence with growing around the experimental time in both plots is summarized in

SALUSTIANA/TROYER 100% CTV GRAPEFRUIT/TROYER - 45% CTV

NEWHALL/TROYER 92% CTV Monitored plot NEWHALL/TROYER Grapft. /Troyer 100 % CTV

Monitored plot Newh./Troyer

VEGETABLES

0% CTV

GRAPEFRUIT/SOUR ORANGE 43% CTV

I

Fig. 1. Scion-rootstock combinations and tristeza incidence in citrus plantings around the experimental plots. Tristeza and Related Diseases 73

TABLE 1 in both cases. Thus, natural spread of RATE OF CITRUS TRISTEZA VIRUS tristeza in grapehit, in our condi- SPREAD IN SOUTHERN VALENCIA (SPAIN) tions, may be explained by the expo- nential or compound interest model Newhall Marsh (transformation Ln (xll-x)) as well as navel orangesz grapefruitz by the linear or single interest model Years (% infe~ted)~ (% infe~ted)~ (transformation Ln (Ill-x) (10). CTV incidence in the Newhall navel plot was too high from the be- ginning as to make any mathematical consideration about the rate of dis- ease increase in the following years. The species tested in the survey "Each plot contained 400 trees. of CTV among vegetation in the cit- YAsdetermined by ELISA. rus orchards are listed below with the number of plants sampled in brackets. table 1. All navel plants were infected AMARANTHACEAE: Amaranthus by tristeza seven years after being sp. [2], A. blitoides [21, A. planted in the field and more than gracilis [29], A. hybridus [I]; 98% were already infected in the APOCY NACE AE : Nerium oleander fourth year. In contrast, less than [3]; ARACE AE : Arisamm vulgare 18% of the grapefruit trees were in- [70]; CARIOPHYLLACEAE: Dian- fected in the eighth year and more thus valentinus [4], Stellaria media than a half of the trees were still [I]; CHENOPODIACEAE: Beta sp. healthy 13 yr after planting. [22], Chenopodium album [71, Vibo Figure 2 shows the tristeza in- spinosum [7]; COMPOSITAE: An- crease in the grapefruit plot (data dryala integrifolia [Z], Calendula ar- from table I), using two logarithmic vensis [5], Centaurea aspera [8], transformations of the per unit inci- Chondrilla juncea 121, Conyxa am- dence. Our data can be reasonably bigua [2], Cychorium intibus [Z], well fitted to a straight line with Inula viscosa [6], Phagnalon either of the two transformations mpestre 131, Senecio vulgaris [31, with a correlation coefficient of 0.97 Sonchus oleraceus [4], S. tenerrirnus

Fig. 2. Progress of tristeza in a grapefruit plot in Southern Valencia (Spain). Per unit inci- dence of tristeza (x) plotted as Ln xll-x (*) and as Ln 111-x (+). Tenth ZOCV Conference

[7]; CONVOLVULACEAE: Convol- p-nitrophenyl phosphate in dieth- vulus althaeoides [I], C. arvensis anolamine buffer, pH 9.6) the solution [141; CRUCIFERAE: Diplotaxis turned yellow in a few minutes, erucoides [18], Lepidium draba 151, suggesting that the positive ELISA L. gramingolium [5], Lobularia reaction observed was probably due mritima [lo], Sinapis nigra [3]; to unspecific binding of enzyme pres- DIPSACACEAE: Cephalaria leuc- ent in Arisarum extracts that con- antha [I], Scabiosa mritima [I]; verted the substrate to nitrophenol SCROPHULARIACEAE: Linaria and caused a false positive reaction. viscosa [I], Verbascum sinnuatum [7], Veronica persica [I]; EUPHOR- DISCUSSION AND BIACEAE: Euphorbia sp. [3], Eu- CONCLUSIONS phorbia segetalis [I]; GERANI- The estimated incidence of ACE AE: Erodium cicutarium [I], E. tristeza in the citrus plantings close malacoides [6], E. moschatum [4]; to the monitored plots is similar to GRAMINEAE: Cynodon dactylon that obtained for this area in a general [I], Poa annua [I], Setaria sp [23]; survey throughout the Valencian HYPERICACEAE: Hypericum per- Community (2). About 95% of citrus foratum [I]; LABIATAE: Sideritis trees (mainly oranges and mandarins) angustgolia [I], Thymus vulgaris were estimated to be infected in the, [I]; LEGUMINOSAE: Glycirhixa Rafelguaraf district, where our ex- glabra [ll], Ononis minutissima perimental plots are located. At El [I], Psoralea bituminosa [36]; Realengo farm almost all orange or LILIACEAE: Allium sp. [2], Allium mandarin trees are CTV infected. In ampeloprasum [6], Asparagus sp. addition, the aphid species vectoring [2]; MALVACEAE: Malva sp. tristeza feed mainly on these species [lo], M. sylvestris [17]; OXALIC- and much less on grapefruit. Thus, ACEAE: Oxalis corniculata [3]; the inoculum potential of CTV around POLYGONACEAE: Polygonum av- the experimental plots was very high. iculare [2], Rumex crispus [9]; POR- The different rates of CTV spread TULACACEAE: Portulaca oleracea measured suggest that navel trees [6]; RANUNCULACEAE : Clematis (and probably any sweet orange) are jlammula [14]; ROSACEAE: Rubus much more susceptible than grape- ulmijlolius [lo], Sanguisorba minor fruit trees to CTV infection, at least [5]; RUBIACEAE: Gallium ap- under our field conditions (virus purine [lo]; RUTACEAE: Ruta strains and vector species present). angustgolia [I]; SOLANACEAE: Additional support for this conclusion Solanum nigrum [lo]; UMBEL- comes from the fact that CTV inci- LIFERAE: Foeniculum vulgare [I]; dence in orange trees around the ex- URTICACEAE: Urtica urens [3]. perimental plots was close to loo%, Only extracts from Arisarum vul- whereas in grapefruit plantings it was gare gave a positive ELISA reaction less than 50% (fig. I), although all when tested with CTV antisera. This grapefruit plantings sampled were positive reaction was obtained with older than the orange plantings. The extracts from all individual plants of planting of grapefruit on sour orange this species sampled in the area heav- was more than 35 yr old. ily infested by tristeza as well as in a Tristeza has been considered as a non-citrus area located more than 25 "compound interest disease" (10) km away from the closest tristeza since its rate of increase depends on focus. Attempts to detect CTV parti- the proportion of trees already in- cles by IEM in different Arisarum fected. We plotted data on the ad- extracts were unsuccessful. vance of tristeza in California and In addition, when a clarified ex- Florida (8) using Ln (xll-x) and they tract from this plant species was fitted straight lines with correlation added to the substrate solution (0.1% coefficients of about 0.98. Tristeza and Related Diseases

In our grapefruit plot, the rate of a secondary peak in early autumn (5) increase of CTV incidence with time and inoculated trees have a latent fitted a straight line whether the ex- period of several months before be- ponential or the linear model was coming systemically infected, aphid- used. The key for a biological explana- inoculated trees probably only be- tion of our data is that the exponential come an inoculum source in the year model for disease spread is based on after infection. Since 98.7% of the the assumption that the only inoculum trees in the Newhall plot were al- available is the infected plants pres- ready infected four years after plant- ent at a given time. This assumption ing, we can safely assume that a high might not be true in this case, since proportion of them may have been in- the inoculum infecting new grapefruit fected with from outside inoculum trees might come from orange or and, consequently, that the disease mandarin trees around the plot where increase model may have been similar the inoculum potential was very high. to that in the grapefruit plot. In this In addition, CTV transmission from hypothetical situation, the average grapefruit to grapefruit occurs at number of new infections per year in much lower efficiency than from the navel plot would have been close orange to grapefruit (9). Thus, it to 25% of trees, in contrast to less seems more realistic to assume that than 5% in the grapefruit plot. most of the new grapefruit infection More than 450 plants of 66 species in the monitored plot occurred with sampled from wild vegetation in the inoculum coming from outside rather heavily infested CVT area were found than from previously infected grapef- free of the virus. This confirms that ruit trees within the plot, at least dur- CTV has a very restricted host range. ing the initial years when the propor- Negative results were also obtained tion of infected trees was still rela- by Miiller and Garnsey (7) who tried tively low. to slash-inoculate nonrutaceous plants The linear model of disease spread of 55 different species. The positive assumes that the rate of disease in- ELISA reaction obtained with ex- crease depends on the amount of in- tracts of Arisarum vulgare proved oculum initially available. Consider- not to be due to CTV infection but ing that in our situation most of the rather to the presence of some en- inoculum could come from orange and zyme inducing a yellow reaction in the mandarin plantings around the substrate solution. This emphasizes grapefruit plot, where the inoculum the importance of using more than potential was very high and stable, one method to confirm diagnosis of the linear model might be adequate virus diseases, particularly when as- to explain tristeza increase in the saying new hosts. grapefruit plot for a number of years. Summarizing, tristeza spread in Later, when the proportion of in- heavily infested areas of Spain occurs fected trees grew higher, new infec- at a very high rate in orange trees tions may have taken place with in- and at slower rate in grapefruit and oculum coming from outside as well the virus seems to be restricted to cit- as from inside. This particular situa- rus trees. tion may explain why our data can be fitted to both mathematical models. ACKNOWLEDGEMENTS Data on tristeza spread within the navel plot cannot be fitted to any par- The invaluable technical assist- ticular model since the disease inci- ance of Encarnacion Martinez is dence was already too high the first gratefully acknowledged. We also year of this study. Nevertheless, thank Dr. Carlos Ramos and Nuria since the peak aphid feeding on citrus Duran for their critical reading of the in the area occurs in late spring with manuscript. Tenth IOCV Conference

LITERATURE CITED Cambra, M., P. Moreno, and L. Navarro 1979. Deteccion rapida del virus de la "tristeza" de 10s citricos (CTV), mediante la tkcnica inmunoenzimatica ELISA-Sandwich. An. ININSer. Protec. Veg. No. 12, p. 1-11. Cambra, M., J. Serra, D. Villalba, and P. Moreno 1986. Present situation of the citrus tristeza virus in the Valencian Community, p. 1-7. In Proc. 10th Conf. IOCV. IOCV, Riverside. 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, p. 9-16. In Proc. 8th Conf. IOCV. IOCV, Riverside. Hermoso de Mendoza, A., J. F. Ballester Olmos, and J. A. Pina Lorca 1984. Transmission of citrus tristeza virus by aphids (Homoptera, Aphididae) in Spain, p. 23-27. In Proc. 9th Conf. IOCV. IOCV, Riverside. Hermoso de Mendoza, A., C. Fuertes, and J. Serra 1986. Proporciones relativas y graficas de vuelo de pulgones (Homoptera, Aphidinea) en 10s citricos espaiioles. Inv. Agrar.: Prod. Prot. Veg. 1: 393-408. Marti Fabregat, F. 1973. La enfermedad de la tristeza de 10s agrios en Espafia. Proc. Int. Soc. Citriculture 2: 531-535. Miiller, G. W. and S. M. Garnsey 1983. Susceptibility of citrus varieties, species, citrus relatives, and non-rutaceous plants to slash-cut mechanical inoculation with citrus tristeza virus (CTV), p. 33-40. In Proc. 9th Conf. IOCV. IOCV, Riverside. Roistacher, C. N. 1976. Tristeza in the Central Valley: A warning. Citrograph 62: 15-23. Roistacher, C. N. and M. BarJoseph 1983. Transmission of tristeza and seedling yellows tristeza virus by Aphis gossypii from sweet orange, grapefruit and lemon to Mexican lime, grapefruit and lemon, p. 9-18. In Proc. 9th Conf. IOCV. IOCV, Riverside. Van Der Plank, J. E. 1963. Plant Diseases: Epidemics and Control. Academic Press, New York and London. 349 P. Vela, C., M. Cambra, E. Cortes, P. Moreno, J. G. Miguet, C. Perez de San Roman, and A. Sanz 1986. Production and characterization of monoclonal antibodies specific for citrus tristeza virus and their use for diagnosis. J. Gen. Virol. 67: 91-96. Vela, C., M. Cambra, A. Sanz, and P. Moreno 1986. Use of specific monoclonal antibodies for diagnosis of citrus tristeza virus, p. 55-61. In Proc. 10th Conf. IOCV. IOCV, Riverside.