ARTHROPODS IN RELATION TO PLANT DISEASE
Transmission Efficiency of Xylella fastidiosa by Sharpshooters (Hemiptera: Cicadellidae) in Coffee and Citrus ROSANGELA C. MARUCCI,1 JOA˜O R. S. LOPES,2
AND
RODNEY R. CAVICHIOLI3
J. Econ. Entomol. 101(4): 1114Ð1121 (2008)
ABSTRACT Xylella fastidiosa (Wells, Raju, Hung, Weisburg, Mandelco-Paul, and Brenner) is a bacterial pathogen transmitted by several sharpshooters in two tribes of Cicadellinae (Proconiini and Cicadellini). Here, we compared the transmission efÞciency of X. fastidiosa in coffee (Coffea arabica L.) and citrus [Citrus sinensis (L.) Osbeck] by Cicadellini [Bucephalogonia xanthophis (Berg) and Dilobopterus costalimai Young] and Proconiini [Homalodisca ignorata Melichar and Oncometopia facialis (Signoret)] sharpshooters that occur in both crops. At different seasons, healthy adults of each species were submitted to a 48-h acquisition access period on citrus or coffee source plants infected with X. fastidiosa isolates that cause Citrus variegated chlorosis (CVC) and Coffee leaf scorch (CLS), respectively, and then conÞned on healthy seedlings of the corresponding host plant for a 48-h inoculation access period. No signiÞcant effect of inoculation season was observed when comparing infection rates of citrus or coffee plants inoculated by vectors at different times of the year. In citrus, the transmission rate by single insects was signiÞcantly higher for H. ignorata (30%) in relation to B. xanthophis (5%) and O. facialis (1.1%), but there was no difference among vector species in coffee, whose transmission rates ranged from 1.2 to 7.2%. Comparing host plants, H. ignorata was more effective in transmitting X. fastidiosa to citrus (30%) in relation to coffee (2.2%), whereas the other vectors transmitted the bacterium to both hosts with similar efÞciencies. Despite these variations, vector efÞciency in coffee and citrus is lower than that reported in other hosts. RESUMEN Xylella fastidiosa (Wells, Raju, Hung, Weisburg, Mandelco-Paul, and Brenner) e´ uma bacte´ ria transmitida por cigarrinhas de duas tribos de Cicadellinae (Proconiini e Cicadellini). Neste estudo, comparou-se a eÞcieˆ ncia de transmissa˜o de X. fastidiosa em citros [Citrus sinensis (L.) Osbeck] e cafeeiro (Coffea arabica L.) por espe´ cies de Cicadellini [Bucephalogonia xanthophis (Berg) e Dilobopterus costalimai Young] e Proconiini [Homalodisca ignorata Melichar e Oncometopia facialis (Signoret)]. Em diferentes e´ pocas, adultos sadios de cada espe´ cie foram submetidos a um perõ´odo de aquisic¸ a˜o de 48 h em plantas de citros ou de cafe´ infectadas com isolados de X. fastidiosa que causam Clorose variegada dos citros (CVC) ou AtroÞa dos ramos do cafeeiro (ARC), respectivamente. Em seguida os insetos foram conÞnados em “seedlings” sadios do hospedeiro correspondente para um perõ´odo de inoculac¸ a˜o de 48 h. Nenhum efeito da e´ poca de inoculac¸ a˜o foi observado nas taxas de infecc¸ a˜o. Em citros, a taxa de transmissa˜o por indivõ´duo foi maior para H. ignorata (30%) em relac¸ a˜o a` B. xanthophis (5%) e O. facialis (1,1%), mas na˜o houve diferenc¸ a para as quatro espe´ cies em cafeeiro, cujas taxas de transmissa˜o variaram de 1,2 a 7,2%. H. ignorata, foi mais eÞciente na transmissa˜o de X. fastidiosa para citros (30%) em relac¸ a˜o ao cafeeiro (2,2%), enquanto que os demais vetores mostraram eÞcieˆ ncia semelhante nos dois hospedeiros. A despeito dessas variac¸ o˜ es, a eÞcieˆ ncia de vetores em cafeeiro e citros e´ mais baixa do que a relatada em outros hospedeiros. KEY WORDS citrus variegated chlorosis, coffee leaf scorch, leafhopper vectors, Cicadellinae
Xylella fastidiosa (Wells, Raju, Hung, Weisburg, Mandelco-Paul, and Brenner) has been detected in sweet 1 Recursos Humanos no Agronego ´ cio, ReHAgro, Av. Uruguai, 620, conj. 507, Belo Horizonte, MG, 30310-300, Brazil. 2 Corresponding author: Departamento de Entomologia, Fitopatologia e Zoologia Agrõ´cola, ESALQ/Universidade de Sa˜o Paulo, CP. 9, Piracicaba, SP 13418-900, Brazil (e-mail:
[email protected]). 3 Departamento de Zoologia, Universidade Federal do Parana ´, CP. 19020, Curitiba, PR, 81531-990, Brazil.
orange [Citrus sinensis (L.) Osbeck] causing Citrus variegated chlorosis (CVC) (Chang et al. 1993) and in coffee (Coffea arabica L.) as the Coffee leaf scorch (CLS) causal agent (Lima et al. 1998). These two diseases are important in Brazil and have been associated with closely related strains of X. fastidiosa (Rosato et al. 1998, Qin et al. 2001). CVC is widespread and has reached a high incidence in the State of Sa˜o Paulo, where 43% of ⬇170 million sweet orange trees were
0022-0493/08/1114Ð1121$04.00/0 䉷 2008 Entomological Society of America
August 2008
MARUCCI ET AL.: X. fastidiosa TRANSMISSION BY VECTORS IN COFFEE AND CITRUS
estimated to be symptomatic in 2003, resulting in annual losses of 100 million dollars (Lopes et al. 2004). The impact and distribution of CLS are not as well understood, but the disease has been found in different Brazilian regions (Li et al. 2001) and caused a yield reduction of 30% in some coffee plantations of northern Sa˜o Paulo (Prato 2000). It has been reported that the CVC strain is transmitted in citrus by various xylem-feeding leafhoppers (Hemiptera: Cicadellidae) in the subfamily Cicadellinae, commonly known as sharpshooters, with low efÞciency (0.3Ð20.3%) (Kru¨ gner et al. 2000, Brlansky et al. 2002, Yamamoto et al. 2002). In addition, it has been suggested that vector species in the tribe Cicadellini transmit X. fastidiosa in citrus with greater efÞciency than those in the tribe Proconiini (Kru¨ gner et al. 2000, Yamamoto et al. 2002). Severin (1949) also observed differences in transmission efÞciency among vectors of both tribes for the causal agent of PierceÕs disease (PD) in grapes, which was later shown to be X. fastidiosa (Davis et al. 1978, Wells et al. 1987). In general, the Proconiini species are larger than those from the tribe Cicadellini and differ in behavioral aspects (Young 1968), showing a tendency to feed on more ligniÞed tissues (Turner and Pollard 1959a). Because X. fastidiosa has an irregular distribution in some host plants (Hopkins 1981, Queiroz-Voltan and Paradela Filho 1999), vector feeding sites may inßuence the probability of acquisition or inoculation of this pathogen (Marucci et al. 2004). In coffee, sharpshooter transmission of X. fastidiosa has been reported (Marucci et al. 2001). However, data on vector range and transmission efÞciency with respect to vector groups in Cicadellinae are not available for CLS. Comparative data on vector transmission of CVC and CLS strains of X. fastidiosa are important to understand the epidemiology of both diseases, because in some Brazilian regions, citrus- and coffeeproducing areas are adjacent, with chance of cross inoculation of these strains. Li et al. (2001) showed that a CVC isolate was able to infect and cause CLS symptoms in coffee after mechanical inoculation. In addition, there are Proconiini and Cicadellini species of sharpshooters that are common to both crops, such as Acrogonia citrina Marucci & Cavichioli, Bucephalogonia xanthophis (Berg), Oncometopia facialis (Signoret), and Dilobopterus costalimai Young (Paradela Filho et al. 1997, Marucci et al. 2002). Thus, it is important to determine the competence of shared vector species for transmission of X. fastidiosa in these two crops. The goal our research was to evaluate the transmission efÞciency of X. fastidiosa to citrus and coffee plants by sharpshooter species representative of tribes Cicadellini (B. xanthophis and D. costalimai) and Proconiini (O. facialis and Homalodisca ignorata Melichar), which occur in both citrus and coffee orchards. In addition to the epidemiological relevance, the transmission data contributes to the biological characterization of CLS and CVC isolates of X. fastidiosa and may be useful to understand their taxonomic relationships. A high degree of genetic similarity has
1115
been found between CLS and CVC isolates (Rosato et al. 1998, Qin et al. 2001, Rodrigues et al. 2003), but it is still not clear whether they represent a same or distinct X. fastidiosa strains. Materials and Methods Vector Rearing. Healthy adult vectors of each species were obtained by rearing the nymphs on a nonhost plant, which does not allow reacquisition of X. fastidiosa (foregut-borne pathogen) after it is lost during molting. Batches of Þeld-collected adults of the four leafhopper species were caged for oviposition (1Ð2 wk) on healthy nursery trees of C. sinensis or on Vernonia condensata Baker (Asteraceae), which is not a host of citrus and coffee strains of X. fastidiosa (Marucci et al. 2003). V. condensata was used as oviposition host for B. xanthophis, whereas C. sinensis was used for D. costalimai, H. ignorata, and O. facialis. When citrus was used as oviposition host, egg-bearing leaves were detached from the plants and placed inside petri dishes with adequate moisture, as described by Marucci et al. (2003). Soon after eclosion, Þrst instars were transferred to potted plants of V. condensata, inside rectangular wooden-framed cages (base of 50 by 60 cm and height of 70 cm), covered with transparent glass on the top and two sides, and with a Þne-mesh nylon screen on the other sides. Nymphs of B. xanthophis were allowed to develop up to the adult stage on V. condensata, which proved to be a good feeding and developmental host for this particular species (Marucci et al. 2003). For the other sharpshooter species, healthy nursery trees of C. sinensis were later added to the cage for development of third to Þfth instars and adult emergence. Test Plants. Healthy seedlings grown inside screenhouses from seeds of sweet orange (C. sinensis) ÔWestinÕ and C. arabica ÔCatuaõ´ VermelhoÕ, clone 99 were used as test plants in the experiments. After reaching a height of 30 Ð50 cm, the sweet orange seedlings were pruned to a height of 8 Ð10 cm above the ground (leaving ⬇4 basal leaves) and fertilized with calcium ammonium nitrate (3 g/10 litter), so that after ⬇25 d they were sprouting uniformly and were at an ideal stage to be used in the transmission tests. X. fastidiosa Isolates and Source Plants. Two X. fastidiosa isolates, CCT6570 and CCT6756 (both deposited at Colec¸ a˜o de Culturas Tropical, Fundac¸ a˜o Andre´ Tosello, Campinas, SP), which cause CVC and CLS, respectively, were used for inoculation of the two plant species. To produce source plants, cultures of these isolates were obtained after two passages on solid periwinkle wilt gelrite (PWG) medium (Hill and Purcell 1995) from puriÞed inoculum preserved at ⫺80⬚C. Bacterial colonies were scrapped from PWG with platinum loop and homogenized in phosphatebuffered saline (PBS) until the resulting suspension became turbid, with cell concentrations of 108Ð109 colony-forming units (CFU)/ml. The suspensions of the CVC and CLS isolates were mechanically inoculated in three different points of the stem of healthy sweet orange and coffee seedlings, respectively; each
1116
JOURNAL OF ECONOMIC ENTOMOLOGY
Vol. 101, no. 4
Table 1. Number of citrus and coffee test plants inoculated by vectors or mechanically (pin-pricking) in different seasons and X. fastidiosa population on citrus and coffee source plants Inoculated plants Seasona
Months
1 2 3 4
MayÐJuly 2000 Aug.ÐOct. 2000 Dec. 2000ÐMarch 2001 Aug.ÐNov. 2001
a
Vectors
X. fastidiosa titer in source plants (CFU/g tissue)
Pin-pricking
Citrus
Coffee
Citrus
Coffee
Citrus
Coffee
25 47 64 43
29 45 64 41
8 8 16 9
9 9 17 9
2.0 ⫻ 106 2.4 ⫻ 106 1.5 ⫻ 106 1.6 ⫻ 106
2.4 ⫻ 106 2.0 ⫻ 106 1.8 ⫻ 106 1.8 ⫻ 106
Inoculation 1, fallÐwinter 2000; 2, spring 2000; 3, summer 2000 Ð2001; and 4, spring 2001.
point received 5 l of suspension, which was placed on the stem surface and pin-pricked Þve times with a no. 0 entomological pin. These source plants were kept in insect-proof screenhouse and used in the transmission experiments at 6 Ð12 mo after inoculation, when X. fastidiosa could be detected by primary isolation on PWG medium (Almeida et al. 2001) and reached a concentration of ⬇106 CFU/g tissue. Transmission Experiments. At four different times during the period of May 2000 to November 2001, groups of 20 Ð50 healthy adult leafhoppers of each species were submitted to a 48-h acquisition access period (AAP) on citrus and coffee source plants infected with the CVC and CLS isolates, respectively. The insects were conÞned on the source plants by using sleeve cages made of Þne-mesh fabric (“voile”). After the AAP, the insects were transferred to the corresponding test plants (citrus seedlings for CVC and coffee seedlings for CLS isolate) for an inoculation access period (IAP) of 48 h. The leafhoppers were conÞned to the young shoots of test plants inside a rectangular Styrofoam-framed cage (12 by 16 cm) with two ventilation openings (7 by 11 cm) on the sides covered with voile. The cage consisted of two symmetrical frames that were held together and fastened to the plants with rubber bands, and were supported by a thin bamboo stake. A strip of plastic foam was used on the area where both frames connected to provide a snug Þt, and also to accommodate the plant stem. Three leafhopper individuals were used per test plant for B. xanthophis, D. costalimai, and O. facialis. Only a single individual of H. ignorata was used per test plant; this species is less abundant in the Þeld and fewer adults were collected for egg laying and production of healthy individuals. Both AAP and IAP were carried out in a greenhouse with temperatures ranging from 20 to 30⬚C. Simultaneously to the vector inoculations, groups of 8 Ð17 citrus and coffee test plants were inoculated with suspensions of the CVC and CLS isolates, respectively, by the pin-pricking method (Almeida et al. 2001). These mechanically inoculated plants were used as positive controls for the transmission experiments, which involved vector inoculations in different seasons (Table 1). The bacterial concentrations of the inoculated suspensions were determined by serial dilution and plating on solid PWG medium, ranging from 107 to 109 CFU/ml. Similar numbers of citrus and coffee test plants were inoculated by leafhoppers or
mechanically in each inoculation season (Table 1). As a negative control, 10 Ð20 noninoculated citrus and coffee seedlings were used in each season. All citrus and coffee test plants were transplanted to 5-dm3 pots containing a mixture of soil, manure, and sand at a rate of 3:2:1, and they were maintained in a vector-proof screenhouse at Campus Luiz de Queiroz (Universidade de Sa˜ o Paulo), in Piracicaba, SP. The plants were fertilized with 2.5 g/liter of Chelal RD (B, 2%; Cu, 0.5%; Fe, 3.25%; Mn, 4.0%; and Zn, 5.1%) (BMS Micro-nutrients N.V., Bornem, Belgium) every 15 d, and every 4 mo with 3 g per pot of Vip Enduro (N ⫽ 30%; P2O5, 11%; Mg, 1.3%; B, 0.05%; and Zn, 1.0%) (Nutriplant Indu´ stria e Come´ rcio, Paulõ´nia, SP, Brazil). Test plants also were sprayed with pesticides against mites and leafminers when needed. Detection of X. fastidiosa in Citrus and Coffee Plants. The test plants were evaluated for infection by X. fastidiosa starting 6 mo after inoculation, by primary isolation on solid PWG medium (Hill and Purcell 1995, Almeida et al. 2001) and by polymerase chain reaction (PCR). PCR tests were repeated at 12 mo, 18 mo, or both. Only the lot inoculated between August and October 2001 (season 4) was not evaluated at 18 mo. A plant was conÞrmed to be infected by X. fastidiosa based on at least one of the following criteria: 1) primary isolation of the bacterium from a leaf sample and isolate identity conÞrmation by PCR; and 2) positive detection of X. fastidiosa in a leaf sample by the PCR test in two or more attempts. PCR Assays. For extracting bacterial DNA from plant tissue, we used the protocol described by Minsavage et al. (1994), which was adapted to coffee plants by optimizing ascorbic acid concentration to 0.1 M and diluting the extract by 1:100 (Pinto and Leite Ju´ nior 1999). Universal primers (RST31 and RST33) that amplify a 733-bp fragment of X. fastidiosa (Minsavage et al. 1994) were used in the PCR. SpeciÞc primers (CVC1 and 272Ð2 int) for detection of CVC (Pooler and Hartung 1995) and CLS (Coletta Filho and Machado 2001) strains, which amplify a 500-bp fragment, also were used in the same reaction (multiplex PCR). The following reagent mixture was used for DNA ampliÞcation: buffer 1⫻ (20 mM Tris-HCl, 100 mM KCl, pH 8.4; Promega, Madison, WI); 2 mM MgCl2; 200 M of each dNTP; 1 U of TaqDNA polymerase; 0.4 M of each oligonucleotide (RST31, RST33, CVC1, 272-2 int); and 3 l of DNA sample for
August 2008
MARUCCI ET AL.: X. fastidiosa TRANSMISSION BY VECTORS IN COFFEE AND CITRUS
1117
Table 2. Sharpshooter mortality (percentage) during the AAP and IAP of X. fastidiosa on source (infected) and test (healthy) plants, respectively 48-h AAP
48-h IAP
Sharpshooter species
Citrus
Coffee
Citrus
Coffee
B. xanthophis D. costalimai H. ignorata O. facialis
14.3a (175)a 6.3 (160)a 3.8 (52)a 8.0 (98)a
7.4 (162)b 5.7 (159)a 3.8 (52)a 7.2 (97)a
1.4 (150)a 3.4 (150)a 0 (50)b 1.1 (90)a
2.0 (150)a 6.0 (150)a 10.0 (50)a 1.1 (90)a
a Mortality rates of each sharpshooter species on C. sinensis vs. C. arabica followed by the same letter, during the AAP or IAP, do not differ statistically (P ⬍ 0.05; two-sample Z test). Numbers in parentheses represent the total number of individuals evaluated.
a Þnal reaction volume of 12 l. AmpliÞcation was performed in a PTC-100 thermocycler, programmed for the following conditions: one cycle at 95⬚C for 1 min, 40 cycles involving denaturation at 95⬚C for 30 s, annealing at 55⬚C for 30 s, extension at 72⬚C for 45 s; and one extension cycle at 72⬚C for 5 min, with stabilization at 4⬚C for indeterminate time. After ampliÞcation, 2 l of dye (0.25% bromophenol blue, 40% sucrose) was added per sample, and the PCR reaction product was visualized by gel electrophoresis consisting of agarose at 1.5% in Tris borate-EDTA buffer (89 mM Tris; 89 mM boric acid, and 2 mM EDTA, pH 8.0), supplemented with 0.5 g/ml ethidium bromide. The ampliÞed fragments were visualized under UV light and documented in an Eagle Eye II Still Video System (Stratagene, La Jolla, CA). Statistical Analysis. The mortality percentage data during AAP and IAP for the four vector species were compared between hosts (citrus and coffee plants) by a test of difference between proportions from two independent samples (two-sample Z test) (␣ ⫽ 0.05; Z critical value ⫽ 1.96) (Magalha˜es and de Lima 2002). The analyses of the transmission experiments in citrus and coffee plants were performed separately. Variables 0 and 1 were used to represent negative and positive plants for the presence of X. fastidiosa, respectively, considering that the data followed BernoulliÕs distribution. For each host plant species, the variables were compared among vector species and inoculation seasons by using a completely randomized design in a factorial arrangement, considering four vector species and four inoculation seasons. The GENMOD procedure of the SAS program (SAS Institute 1996) with binomial distribution and logit link function was used for the analyses. Whenever the likelihood ratio statistic (2) revealed an effect of one of the factors (species and seasons) or of the species ⫻ season interaction, factors were compared by means of contrasts. Probabilities of transmission by single insects for each leafhopper species were estimated according to Swallow (1985). The comparisons of transmission probabilities for each vector species in citrus versus coffee and between each two vector species on the same host plant were performed by two-sample Z test (␣ ⫽ 0.05; Z critical value ⫽ 1.96) (Magalha˜es and de Lima 2002).
Results Sharpshooter Mortality during AAP and IAP. Low mortality (0 Ð14.3%) was observed during the 48-h periods in which the insects were maintained on citrus or coffee source plants (AAP) and test plants (IAP) in this transmission study. When the mortalities of individual sharpshooter species on citrus versus coffee plants were compared, a higher percentage of B. xanthophis died during the AAP on citrus (14.3%) (Z ⫽ 2.0, P ⬍ 0.05), whereas a higher mortality of H. ignorata during the IAP was observed on coffee (10%) (Z ⫽ 2.29, P ⬍ 0.05). The mortalities of the other species did not differ between the two hosts (Table 2). X. fastidiosa Detection in Test Plants. Most infected plants were conÞrmed by PCR and culture assays. Some infected plants that were not detected by culture at 6 mo after inoculations, were later conÞrmed as positives by at least two consecutive PCR assays (12 and 18 mo). Diagnostic CVC symptoms were observed in most infected citrus plants, but clear CLS symptoms were not observed in the infected coffee plants during the course of this study, suggesting that incubation period of CLS after vector inoculation is longer than that of CVC. X. fastidiosa was not detected by culture or by PCR in any of the noninoculated plants maintained as negative controls. Despite the observed oscillations in transmission data (Table 3), the chisquare test did not detect signiÞcant differences in the percentages of positive plants inoculated by sharpshooters in citrus (2 ⫽ 1.79, P ⫽ 0.62) and coffee (2 ⫽ 3.44, P ⫽ 0.33) among inoculation seasons. Also, there was no interaction between species ⫻ inoculation season for both host plants. Therefore, inoculation season did not have an inßuence on the results obtained. In the case of mechanically inoculated plants, there was a signiÞcant difference in the percentage of infected plants among inoculation seasons for coffee (2 ⫽ 15.32, P ⬍ 0.01) but not for citrus (2 ⫽ 2.10, P ⫽ 0.55); the percentage of positive coffee plants was lower during the summer than in the other three inoculation seasons (Table 3). Transmission Efficiency by Vectors. Mean transmission rate of X. fastidiosa per individual sharpshooter varied with host plant and was signiÞcantly higher in citrus than in coffee plants for H. ignorata (Z ⫽ 3.71, P ⬍ 0.01) (Table 4). For B. xanthophis (Z ⫽ 0.4, P ⬎ 0.05), D. costalimai (Z ⫽ 1.0, P ⬎ 0.05), and O. facialis (Z ⫽ 0.04, P ⬎ 0.05), there were no signiÞcant differ-
1118
JOURNAL OF ECONOMIC ENTOMOLOGY
Vol. 101, no. 4
Table 3. Detection of X. fastidiosa by PCR and/or culture in C. sinensis and C. arabica plants inoculated by vectors or mechanically (pin-pricking) in different seasons Seasona
Inoculation method B. xanthophis
D. costalimai
H. ignorata
O. facialis
Noninoculated plants (negative control)
Mechanically inoculated plants (positive control)
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
Pb
Rate of infected plants (%) Citrus 0/4 (0)c 2/12 (16.7) 3/16 (18.6) 2/17 (11.8) 5/6 (83.3) 4/11 (36.4) 3/17 (17.6) 4/14 (28.6) 1/5 (20) 5/15 (33.3) 4/22 (18.2) 2/5 (40) 0/10 (0) 1/9 (11.1) 0/5 (0) 0/6 (0) 0/10 (0) 0/10 (0) 0/20 (0) 0/10 (0) 5/8 (62.5) 7/8 (87.5) 13/15 (86.7) 7/9 (77.8)
Coffee
Citrus
Coffee
0/3 (0) 0/11 (0) 1/20 (5) 4/16 (25) 2/12 (16.7) 4/12 (33.3) 2/16 (12.5) 2/10 (20) 0/10 (0) 0/9 (0) 1/26 (3.8) 0/5 (0) 0/4 (0) 0/13 (0) 0/2 (0) 1/10 (10) 0/10 (0) 0/10 (0) 0/20 (0) 0/10 (0) 7/9 (77.8) 7/9 (77.8) 6/17 (35.3) 9/9 (100)
0 0.060 0.069 0.041 0.450 0.140 0.063 0.106 0.200 0.333 0.182 0.400 0 0.038 0 0 NAd NA NA NA NA NA NA NA
0 0 0.017 0.091 0.059 0.126 0.043 0.072 0 0 0.038 0 0 0 0 0.035 NA NA NA NA NA NA NA NA
a
Inoculation seasons are described in Table 1. Transmission probability per individual estimated by the formula P ⫽ 1 ⫺ (1 ⫺ I)1/k, where I corresponds to the proportion of positive plants and k corresponds to the number of insects per plant (Swallow 1985). c Number of infected plants over the total number of plants tested; corresponding percentages are shown in parentheses. d NA, not applicable. b
ences in transmission efÞciencies comparing the two host plants. In citrus, H. ignorata transmitted X. fastidiosa more efÞciently than B. xanthophis (Z ⫽ 3.18, P ⬍ 0.01) and O. facialis (Z ⫽ 3.2, P ⬍ 0.01), but it was not signiÞcantly different from D. costalimai (Z ⫽ 1.89, P ⬎ 0.05) (Table 4). When D. costalimai, B. xanthophis, and O. facialis were compared in pairs, they did not differ signiÞcantly between themselves with regard to transmission efÞciency in citrus. In coffee, there were no signiÞcant differences in transmission efÞciency between sharpshooter species (Table 4).
Table 4. Transmission rates (ⴞ SEM) of X. fastidiosa to C. sinensis and C. arabica per individual of different sharpshooter species Sharpshooter species
Citrus
Coffee
Bucephalogonia xanthophis Dilobopterus costalimai Homalodisca ignorata Oncometopia facialis
5.0 ⫾ 1.7aAbbc 13.3 ⫾ 6.4Aab 30 ⫾ 4.2Aa 1.1 ⫾ 0.94Ab
3.5 ⫾ 2.1Aa 7.2 ⫾ 1.1Aa 2.2 ⫾ 1.8Ba 1.2 ⫾ 0.9Aa
a Transmission probabilities per single insects were estimated by the formula P ⫽ 1 ⫺ (1 ⫺ I)1/k, where I corresponds to the proportion of positive plants and k corresponds to the number of insects per plant (Swallow 1985). P values were converted to percentage (⫻100). Rates were obtained by pooling the data for all inoculation seasons. b Rates followed by the same capital letter in the row are statistically similar (P ⬍ 0.05; two-sample Z test). c Rates followed by the same lowercase letter in the column are statistically similar (P ⬍ 0.05; two-sample Z test).
Discussion Several species of sharpshooter leafhoppers (Hemiptera: Cicadellidae: Cicadellinae) have been listed as vectors of X. fastidiosa in different crops affected by this pathogen (Nielson 1985, Redak et al. 2004). There is little, if any, vector speciÞcity for X. fastidiosa within Cicadellinae, but transmission efÞciency may vary among sharpshooter species, bacterial strains, and host plants (Severin 1949, Purcell 1980a, Redak et al. 2004). In this study, we compared transmission efÞciencies of X. fastidiosa in coffee and citrus by Proconini and Cicadellini sharpshooters commonly found on these host plants. Pairwise comparisons for each vector species in the two hosts showed that most sharpshooters (B. xanthophis, D. costalimai, and O. facialis) transmitted the CVC and CLS isolates of X. fastidiosa with similar efÞciencies; the exception was H. ignorata, which transmitted the former isolate with higher efÞciency. By comparing vector species within each host, H. ignorata (Proconiini) and D. costalimai (Cicadellini) were the most efÞcient vectors of the CVC isolate in citrus, whereas O. facialis (Proconiini) was a poor vector. In coffee, transmission efÞciencies of the CLS isolate were low (1.2Ð7.2%) and not statistically different among vector species of both tribes. The highest transmission rates observed for X. fastidiosa in citrus and coffee here (30 and 7.2%, respectively) and in a previous study (20.3% in citrus) (Brlansky et al. 2002) are lower than those reported in
August 2008
MARUCCI ET AL.: X. fastidiosa TRANSMISSION BY VECTORS IN COFFEE AND CITRUS
other host plants. Although transmission efÞciency in other hosts can vary widely depending on the vector species (Redak et al. 2004), rates as high as 48, 75, 83, and ⬎90% were recorded in peach (Turner and Pollard 1959b), almond (Purcell 1980a), oleander (Costa et al. 2000a), and grape (Purcell and Finlay 1979, Hill and Purcell 1995), respectively. Variations in transmission efÞciency may be explained by interactions of vectors and bacterial strains with host plants. Hill and Purcell (1997) showed that acquisition rate of X. fastidiosa by sharpshooters in grape and other host plants is positively related to bacterial concentrations in the source plants; populations lower than 106 to 107 CFU/g plant tissue result in reduced transmission efÞciency. It was previously reported that average concentration of X. fastidiosa cells in citrus (105-106 CFU/g tissue) is 100 Ð1,000 times lower than in grapevines (108 CFU/g) (Almeida et al. 2001). In this study, we also detected relatively low X. fastidiosa populations in coffee source plants, around 106 CFU/g tissue (Table 1). Thus, the low titer of the CVC and CLS isolates probably contributes to the low transmission rates observed in coffee and citrus. The low quality of diseased citrus as a host for the vectors also may affect X. fastidiosa transmission. The sharpshooters vectors D. costalimai and O. facialis reject plants showing CVC symptoms in choice tests between a healthy and a diseased citrus plant; when conÞned on branches of symptomatic plants, D. costalimai showed a very low ingestion rate (Marucci et al. 2005). The low feeding activity of sharpshooters on diseased citrus plants may result in lower acquisition efÞciency of the pathogen, explaining at least in part the relatively low transmission rates reported for the CVC strain. In the current study, the lower transmission efÞciency by H. ignorata in coffee (2.2%) compared with citrus (30%) also may be associated with feeding preferences; its mortality rate during the IAP on coffee was 10%, whereas on citrus there was no mortality, suggesting that citrus plants were more acceptable to this particular sharpshooter. It may be argued that the higher transmission efÞciency observed for H. ignorata in citrus (30%) could be an artifact due to differences in number of insects used per test plant among the sharpshooter species; a single individual per plant was used for this species compared with three individuals per plant for B. xanthophis, O. facialis, and D. costalimai. In some studies with X. fastidiosa, lower estimated probabilities of transmission by single insects were observed when groups of sharpshooters were used per test plant (Hill and Purcell 1997, Costa et al. 2000a, Brlansky et al. 2002). A large number of insects per plant may affect sharpshooter feeding behavior, producing competition for space and excessive dispersal during the inoculation access period. In fact, a previous study in citrus indicated a much lower probability of transmission per individual of H. ignorata (0.5%) by using 10 insects per test plant (Yamamoto et al. 2002). However, in our study the transmission rate by H. ignorata in coffee was very low (2.2%) despite the fact that a single insect was used per test plant, suggesting that
1119
the host plant (or X. fastidiosa isolate) had a greater impact on vector efÞciency than the number of insects used to inoculate each plant. Although we carried out transmission trials at different times of the year for all vector species, no signiÞcant seasonal effect was detected on rates of infected citrus or coffee plants after vector inoculations. Low winter temperatures are known to affect X. fastidiosa infection and survival in grapevines (Purcell 1980b, Hopkins and Thompson 1984). Apparently, the range of temperatures observed in the screenhouse (10 Ð35⬚C) during our experiment was not wide enough to affect infection of citrus and coffee plants inoculated at different times of the year. The only observed effect was a reduction in the proportion of infected coffee plants after mechanical inoculation during the summer 2000 Ð2001, for which we do not know the reason. Determining key vectors and their ability to transmit X. fastidiosa is important to better understand the epidemiology of diseases associated with this pathogen. The results of this study show that X. fastidiosa is transmitted in coffee and citrus with similar, but low efÞciencies by three sharpshooter species (B. xanthophis, D. costalimai, and O. facialis) commonly found both in citrus and coffee orchards. Because of the low transmission efÞciency, long feeding periods are probably required for successful acquisition and inoculation of the pathogen by sharpshooters in coffee or citrus, thus favoring the establishment of control measures directed to the vectors in these crops. The similar transmission efÞciency of X. fastidiosa in coffee and citrus by most vector species is consistent with genetic similarities found among citrus and coffee isolates of X. fastidiosa. Genetic studies have indicated that X. fastidiosa isolates from these two crops are closely related in comparisons with isolates from other plants (e.g., grape, plum, and almond) in California, although they cluster separately from each other based on random ampliÞed polymorphic DNA or arbitrarily primed PCR analyses (Rosato et al. 1998, Costa et al. 2000b, Qin et al. 2001). Li et al. (2001) observed that a CVC isolate colonized and caused CLS-like symptoms in coffee plants. In contrast, Prado et al. (2008) observed a lower infection rate and multiplication in coffee by another CVC isolate, and no colonization of citrus by a CLS isolate. Apparently, more information on pathogenicity and biological characteristics of various isolates from coffee and citrus is needed to establish whether CLS and CVC are caused by the same or by distinct X. fastidiosa strains. Acknowledgments We thank Jose´ Eduardo Corrente (Departamento de Bioestatõ´stica do Instituto de Biocieˆ ncias [UNESP/Botucatu]) for help with the statistical analyses. We also acknowledge Coordenadoria de Aperfeic¸ oamento de Pessoal de Nõ´vel for granting a scholarship to R.C.M., Conselho Nacional de Desenvolvimento Cientõ´Þco e Tecnolo´ gico for providing scholarships to J.R.S.L. and R.R.C., and FUNDECITRUS and
1120
JOURNAL OF ECONOMIC ENTOMOLOGY
CBP&D-Cafe´ /EMBRAPA for Þnancial support for conducting the experiments. This work has been extracted from R.C.M.Õs doctoral dissertation, developed at Departamento de Entomologia, Fitopatologia e Zoologia Agrõ´cola of ESALQ/USP, Piracicaba, SP.
References Cited Almeida, R.P.P., E. F. Pereira, A. H. Purcell, and J.R.S. Lopes. 2001. Multiplication and movement of a citrus strain of Xylella fastidiosa within sweet orange. Plant Dis. 85: 382Ð 386. Brlansky, R. H., V. D. Damsteegt, and J. S. Hartung. 2002. Transmission of the citrus variegated chlorosis bacterium Xylella fastidiosa with the sharpshooter Oncometopia nigricans. Plant Dis. 86: 1237Ð1239. Chang, C. J., M. Garnier, L. Zreik, V. Rossetti, and J. M. Bove´. 1993. Culture and serological detection of the xylemlimited bacterium causing citrus variegated chlorosis and its identiÞcation as a strain of Xylella fastidiosa. Curr. Microbiol. 27: 137Ð142. Coletta Filho, H. D., and M. A. Machado. 2001. Hospedeiros, transmissa˜o e te´ cnicas de diagno´ stico da bacte´ ria Xylella fastidiosa. Laranja 22: 121Ð132. Costa, H. S., M. S. Blua, J. A. Bethke, and R. A. Redak. 2000a. Transmission of Xylella fastidiosa to oleander by the glassywinged sharpshooter, Homalodisca coagulata. HortScience 35: 1265Ð1267. Costa, P. I. da, C. F. Franco, V. S. Miranda, D. C. Teixeira, and J. S. Hartung. 2000b. Strains of Xylella fastidiosa rapidly distinguished by arbitrarily primed PCR. Curr. Microbiol. 40: 279 Ð282. Davis, M. J., A. H. Purcell, and S. V. Thompson. 1978. PierceÕs disease of grapevines: isolation of the causal organism. Science (Wash., D.C.) 199: 75Ð77. Hill, B. L., and A. H. Purcell. 1995. Acquisition and retention of Xylella fastidiosa by an efficient vector, Graphocephala atropunctata. Phytopathology 85: 209 Ð212. Hill, B. L., and A. H. Purcell. 1997. Populations of Xylella fastidiosa in plants required for transmission by efÞcient vector. Phytopathology 87: 1197Ð1201. Hopkins, D. L. 1981. Seasonal concentration of the PierceÕs disease bacterium in grapevine stems, petioles, and leaf veins. Phytopathology 71: 415Ð 418. Hopkins, D. L., and C. M. Thompson. 1984. Seasonal concentration of the PierceÕs disease bacterium in ÔCarlosÕ and ÔWelderÕ muscadine grapes compared with ÔSchuylerÕ Õbunch grape. HortScience 19: 419 Ð 420. Kru¨ gner, R., M.T.V. de C. Lopes, J. S. Santos, M.J.G. Beretta, and J.R.S. Lopes. 2000. Transmission efÞciency of Xylella fastidiosa to citrus by sharpshooters and identiÞcation of two new vector species, p. 81. In Proceedings, 14th Conference of International Organization of Citrus Virologists, 13Ð18 September 1998, Campinas, Brazil. IAC, Campinas, Brazil. Li, W. B., W. D. Pria, Jr., D. C. Teixeira, V. S. Miranda, A. J. Ayres, C. F. Franco, M. G. Costa, C.-X. He, P. I. Costa, and J. S. Hartung. 2001. Coffee leaf scorch caused by a strain of Xylella fastidiosa from citrus. Plant Dis. 85: 501Ð505. Lima, J.E.O., V. S. Miranda, J. S. Hartung, R. H. Brlansky, A. Coutinho, S. R. Roberto, and E. F. Carlos. 1998. Coffee leaf scorch bacterium: axenic culture, pathogenicity, and comparison with Xylella fastidiosa of citrus. Plant Dis. 82: 94 Ð97. Lopes, S. A., F. F. Laranjeira, L. Amorim, and A. Bergamin Filho. 2004. Clorose variegada: perdas anuais de US$100 milho˜ es. Visa˜o Agrõ´c. 1: 20 Ð23.
Vol. 101, no. 4
Magalha˜ es, M. N., and A.C.P. de Lima. 2002. Noc¸ o˜ es de probabilidade e Estatõ´stica. EDUSP, Sa˜o Paulo, SP, Brazil. Marucci, R. C., T. A. Giustolin, M. P. Miranda, P. C. Ferraz, and J.R.S. Lopes. 2001. Sharpshooter transmission of a coffee strain of Xylella fastidiosa to coffee seedlings. Fitopatol. Bras. 26: 277. Marucci, R. C., R. R. Cavichioli, and R. A. Zucchi. 2002. Espe´ cies de cigarrinhas (Hemiptera, Cicadellidae, Cicadellinae) em pomares de citros da regia˜o de Bebedouro, SP, com descric¸ a˜o de uma nova espe´ cie de Acrogonia Stål. Rev. Bras. de Entomol. 28: 149 Ð164. Marucci, R. C., T. A. Giustolin, M. P. Miranda, H. Miquelote, R.P.P. Almeida, and J.R.S. Lopes. 2003. IdentiÞcation of a non-host plant of Xylella fastidiosa to rear healthy sharpshooter vectors. Sci. Agric. 60: 669 Ð 675. Marucci, R. C., J.R.S. Lopes, J. D. Vendramim, and J. E. Corrente. 2004. Feeding site preference of Dilobopterus costalimai Young and Oncometopia facialis (Signoret) (Hemiptera: Cicadellidae) on citrus plants. Neotrop. Entomol. 33: 759 Ð768. Marucci, R. C., J.R.S. Lopes, J. D. Vendramim, and J. E. Corrente. 2005. Inßuence of Xylella fastidiosa infection of citrus on the feeding preference of sharpshooter (Hemiptera: Cicadellidae) vectors. Entomol. Exp. Appl. 117: 95Ð103. Minsavage, G. V., C. M. Thompson, D. L. Hopkins, R.M.V.B. Leite, and R. E. Stall. 1994. Development of a polymerase chain reaction protocol for detection of Xylella fastidiosa in plant tissue. Phytopathology 84: 446 Ð 461. Nielson, M. W. 1985. Leafhoppers systematics, pp. 11Ð39. In L. R. Nault and J. G. Rodriguez [eds.], The leafhoppers and planthoppers. John Wiley, New York. Paradela Filho, O., M. H. Sugimori, I.J.A. Ribeiro, A. Garcia Jr., M.J.G. Beretta, R. Harakawa, M. A. Machado, F. F. Laranjeira, J. Rodrigues Neto, and L.O.S. Berian. 1997. Constatac¸ a˜o de Xylella fastidiosa em cafeeiro no Brasil. Summa Phytopathologica 23: 46 Ð 49. Pinto, F.G.S., and R. P. Leite Ju´ nior. 1999. Detecc¸ a˜o de Xylella fastidiosa em Coffea spp. atrave´ s da te´ cnica de PCR. Fitopatol. Bras. 24: 254. Pooler, M. R., and J. S. Hartung. 1995. SpeciÞc PCR detection and identiÞcation of Xylella fastidiosa strains causing citrus variegated chlorosis. Curr. Microbiol. 31: 377Ð381. Prado, S. S., J.R.S. Lopes, C.G.B. Deme´trio, A. F. Borgatto, and R.P.P. Almeida. 2008. Host colonization differences between citrus and coffee isolates of Xylella fastidiosa in reciprocal inoculations. Sci. Agric. 65: 251Ð258. Prato, J.R.A. 2000. Quem diria, a Xylella fastidiosa sempre esteve no cafe´ . A Granja 614: 40 Ð 43. Purcell, A. H. 1980a. Almond leaf scorch: leafhopper and spittlebug vectors. J. Econ. Entomol. 73: 834 Ð38. Purcell, A. H. 1980b. Environmental therapy for PierceÕs disease of grapevines. Plant Dis. 64: 388 Ð390. Purcell, A. H., and A. H. Finlay. 1979. Evidence for noncirculative transmission of PierceÕs disease bacterium by sharpshooter leafhoppers. Phytopathology 69: 393Ð 395. Qin, X., V. S. Miranda, M. Machado, E.G.M. Lemos, and J. S. Hartung. 2001. An evaluation of the genetic diversity of Xylella fastidiosa isolated from diseased citrus and coffee in Sa˜o Paulo, Brazil. Phytopathology 91: 599 Ð 605. Queiroz-Voltan, R. B., and O. Paradela Filho. 1999. Caracterizac¸ a˜o de estruturas anatoˆ micas de citros infectados com Xylella fastidiosa. Laranja 20: 55Ð76. Redak, R. A., A. H. Purcell, J.R.S. Lopes, M. J. Blua, R. F.
August 2008
MARUCCI ET AL.: X. fastidiosa TRANSMISSION BY VECTORS IN COFFEE AND CITRUS
Mizell, and P. C. Andersen. 2004. The biology of xylem ßuid-feeding insect vectors of Xylella fastidiosa and their relation to disease epidemiology. Annu. Rev. Entomol. 49: 243Ð270. Rodrigues, J.L.M., M. E. Silva-Stenico, J. E. Gomes, J.R.S. Lopes, and S. M. Tsai. 2003. Detection and diversity assessment of Xylella fastidiosa in Þeld-collected plant and insect samples by using 16S rRNA and gyrB sequences. Appl. Environ. Microbiol. 69: 4249 Ð 4255. Rosato, Y. E., J. R. Neto, V. S. Miranda, E. F. Carlos, and G. P. Manfio. 1998. Diversity of a Xylella fastidiosa population isolated from Citrus sinensis affected by citrus variegated chlorosis in Brazil. System. Appl. Microbiol. 21: 593Ð598. SAS Institute. 1996. Guia do usua´rio SAS/STAT, versa˜o 6 ed. SAS Institute, Cary, NC. Severin, H.H.P. 1949. Transmission of the virus of PierceÕs disease by leafhoppers. Hilgardia 19: 190 Ð202. Swallow, W. H. 1985. Group testing for estimating infection rates and probabilities of disease transmission. Phytopathology 75: 882Ð 889.
1121
Turner, W. F., and H. N. Pollard. 1959a. Life histories and behavior of Þve insect vectors of phony peach disease. U.S. Dep. Agric. Tech. Bull. 1188: 1Ð32. Turner, W. F., and H. N. Pollard. 1959b. Insect transmission of phony peach disease. U.S. Dep. Agric. Tech. Bull. 1193: 1Ð27. Wells, J. M., B. C. Raju, H.-Y. Hung, W. G. Weisburg, L. Mandelco-Paul, and D. J. Brenner. 1987. Xylella fastidiosa gen. nov., sp. nov: gram-negative, xylem-limited, fastidious plant bacteria related to Xanthomonas spp. Int. J. Syst. Bacteriol. 37: 136 Ð143. Yamamoto, P. T., S. R. Roberto, W. D. Pria, Jr., M. R. Felippe, V. S. Miranda, D. C. Teixeira, and J.R.S. Lopes. 2002. Transmissa˜o de Xylella fastidiosa por cigarrinhas Acrogonia virescens e Homalodisca ignorata (Hemiptera: Cicadellidae) em plantas cõ´tricas. Summa Phytopathol. 28: 178 Ð181. Young, D. A. 1968. Taxonomic study of the Cicadellinae (Homptera: Cicadellidae) Part 1. Proconiini. Smithson. Inst. Bull. Received 25 September 2007; accepted 11 March 2008.
