Abstract
The incidence of Pierce’s disease (PD) within an American hybrid vineyard in the Texas Gulf Coast was evaluated. Multiple viticultural parameters and Xylella fastidiosa levels were assessed for the varieties Black Spanish, Blanc du Bois, and Cynthiana in this region of intense disease pressure. Potential PD symptoms were observed uniformly in Cynthiana and sporadically in Blanc du Bois and Black Spanish postharvest. Consistent with these symptoms, Cynthiana showed significantly lower crop yield than either Blanc du Bois or Black Spanish and had significantly higher positive ELISA values for X. fastidiosa in summer petioles. Individual Cynthiana vines with higher bacterial levels as determined by ELISA exhibited correspondingly lower yields (p = 0.06). Indirect immunofluorescence microscopy of petiole xylem sap and PCR of petiole DNA confirmed positive X. fastidiosa ELISA readings in all three varieties. Cynthiana petioles tested by ELISA showed a dramatic increase in bacterial levels over the growing season. In contrast, bacterial levels in Black Spanish and Blanc du Bois remained lower all season despite high disease pressure. An evaluation of potential local plant reservoirs for X. fastidiosa demonstrated that at least six asymptomatic plant species tested positive by both ELISA and PCR: Ilex vomitoria, Vitis mustangensis, Rubus trivialis, Platanus occidentalis, Lippia nodiflora, and Lonicera japonica. In addition, June insect traps contained a significant number of X. fastidiosa-positive glassy-winged sharpshooters, Homalodisca coagulata, and several positive sharpshooters identified as Oncometopia orbona, suggesting the presence of multiple species of insect vector in this region.
Pierce’s disease (PD) has been a chronic problem in many of the viticultural areas of California. The establishment of the glassy-winged sharpshooter as a vector has aggravated the spread of PD beyond its more typical range in warmer environments and riparian microclimates. Outside California, PD has historically been problematic along the Gulf Coast region where the disease is thought to be endemic (Hewitt 1958, Lipe 1980). However, during the last several years there have been sporadic outbreaks of PD within the drier Texas foothills, causing major damage to the burgeoning Texas wine industry (Kamas 2002). Recently, the Texas Pierce’s Disease Research and Education Program initiated a statewide investigation of the ecology and epidemiology of PD’s causative bacterium, Xylella fastidiosa. A major thrust of the program is to significantly increase understanding of the disease along the Gulf Coast. Because of the prevalence of PD within many vineyards along the Gulf Coast, this area has been labeled a “PD hot zone.”
The Gulf Coast has warmer temperatures, a much higher humidity, and subsequently higher disease pressure than most grapegrowing regions. In addition to many fungal maladies, PD has been a persistent problem. Historically, attempts to grow Vitis vinifera in the southeastern United States from the late 1800s have failed because of PD (Purcell and Feil 2001). Most grapegrowers in the Gulf Coast area now plant disease-resistant American hybrids. Hybrids of V. vinifera and native species such as V. aestivalis display either resistance or tolerance to X. fastidiosa (Galet 1979). Genetic resistance to PD is prevalent in wild Vitis species found within the Gulf Coast (Mortensen et al. 1977). Possible PD-resistance genes within populations of wild vines and their hybrids are the subject of intense investigation (Krivanek and Walker 2005, Ruel and Walker 2006). Eventually, V. vinifera varieties may either be bred for resistance or transformed with specific resistance genes. Until then, growers have two options for reducing the impact of PD. First, they can use standard viticultural techniques (reducing weeds, removing surrounding host plants, and spraying for sharpshooter insect vectors) to reduce the likelihood of PD infection and the rate of spread. Second, they may plant American hybrids such as Black Spanish, Blanc du Bois, and Cynthiana. American hybrids typically have fruit characteristics less suited for winemaking than V. vinifera, but these hybrids are a solution for vineyards that have been severely affected by PD or vineyards within PD hot zones. American hybrids typically are used for winemaking from Florida to Missouri, and cold fermentation of Blanc du Bois from Texas has recently produced some promising wines. Despite the use of these hybrids and their existence for over 50 years, there is little published on the use of American hybrids in vineyards.
Blanc du Bois is an American hybrid with a complex parentage including Cardinal, Golden Muscat, Pixiola, and V. aestivalis (Mortensen 1987). Black Spanish (syn. Jacques or Lenoir) is considered a natural V. aestivalis hybrid, and grape taxonomists generally agree that Black Spanish includes the parentage of three species, V. aestivalis, V. cinerea, and V. vinifera (Galet 1979, Pongracz 1983, Viala and Ravaz 1903). Cynthiana (syn. Norton) is often considered a cultivar of pure V. aestivalis, but it has much larger clusters and more palatable fruit than wild V. aestivalis, suggesting it may actually be a natural hybrid, probably of V. aestivalis and V. labrusca (Pongracz 1983).
Greenhouse studies have demonstrated variable PD resistance within American hybrids (Fry and Milholland 1990, Krivanek and Walker 2005). Many hybrids have been selected for survival in field trials where disease pressure is high (e.g., Mortensen 1987). While hybrids survive in a PD hot zone, there is still much to learn about the X. fastidiosa levels these vines host and what effect, if any, this bacterium has on growth and yield. Previous evaluations of infection levels in hybrid petioles from the Gulf Coast suggested that X. fastidiosa is not detectable in Black Spanish (J. Kamas, 2004, personal communication).
The sources of X. fastidiosa inoculum within Texas are also unknown at this time. Recent work has implicated wild grape and blackberry as potential plant reservoirs in California (Baumgartner and Warren 2005), and there are many wild vine species within Texas. Recent work has shown that X. fastidiosa can infect a broad host range of plants (Wistrom and Purcell 2005). Although the fate of X. fastidiosa within most inoculated plants is unknown, the disease pressure along the Gulf Coast suggests multiple host plants are likely to be found in this region. Likewise, little is known about the vectors for PD spread in Texas, except that more than a dozen distinct sharpshooter species have been identified.
Our goal was to investigate PD tolerance and resistance in the hybrids Blanc du Bois, Cynthiana, and Black Spanish, specifically to (1) report average crop yield and pruning weights for these varieties in an established vineyard; (2) report yield data (collected by the grower) to evaluate long-term health of these varieties in a high PD area; (3) determine whether these PD-resistant hybrid varieties carried detectable levels of X. fastidiosa within a PD-endemic region and quantify the bacterial content within the petioles temporally and between hybrids; (4) evaluate X. fastidiosa content in the native and ornamental plants in and around the vineyard to determine potential sources of inoculum; and (5) evaluate and survey the endemic sharpshooter species for X. fastidiosa levels during spring and early summer to determine the potential disease pressure the different insect species may bring to this vineyard.
Materials and Methods
Vineyard.
The study was conducted at Austin County Vineyards, a 4.5-acre vineyard located in Cat Spring, Texas, 70 miles west of Houston, long.: 96.342W; lat.: 29.854N. The vineyard consisted of three varieties planted in parallel rows to each other on a south-facing hill. Blanc du Bois vines were 15 years old and Cynthiana and Black Spanish vines were five years old. An additional original small planting of Cynthiana in the vineyard allowed for comparable historical crop data for both Blanc du Bois and Cynthiana dating to 1993. Although the Blanc du Bois data vines were older than those of Cynthiana or Black Spanish, it was decided that data would be taken on all three varieties and conclusions carefully interpreted given age differences. Two insect collections were made with yellow sticky traps in the vineyard over two weeks in April to May 2004 and over two weeks in late May and early June 2004. This preliminary trapping was undertaken to provide initial insight into potential sharpshooter species that may be harboring X. fastidiosa at this vineyard.
Viticultural data.
Crop weights and pruning weights were taken from the same 27 vines over 3 rows in a randomized block design for each of the three varieties (81 vines total). The rows were blocked to determine if there was any effect of row or hill position on vigor. All three hybrids in this vineyard appeared healthy throughout most of the growing season. After harvest, mild PD symptoms including slight leaf scorching, green islands (nonuniform periderm hardening), and matchstick petioles (petioles remain on shoot as leaves senesce) were observed sporadically on a few vines for Blanc du Bois and Black Spanish and more uniformly in Cynthiana (a few symptoms on almost every vine). The 27 data vines of each variety were tagged during winter using a randomized block design and were not specifically tagged based on presence or absence of disease. Despite slight PD symptoms observed late in the season, vines were not suffering from PD infection symptoms typical of a V. vinifera vineyard. PD was not apparent in hybrid vines the following spring, and no vine loss could be attributed to PD after multiple years.
Blanc du Bois was harvested in July 2003 and the red varieties in August 2003. Pruning weights were taken in February 2004. The same 81 vines above were used in analysis of X. fastidiosa levels using enzyme-linked immunosorbent assay (ELISA) and a subset of positive ELISA vines verified with polymerase chain reaction (PCR) and indirect immunofluorescence.
Detection of X. fastidiosa.
ELISA.
Enzyme-linked immunosorbent assays were performed on three petioles from each of 27 vines for each of the three varieties (81 vines for every ELISA evaluation). A total of 0.2 g from three petioles was macerated using mortar and pestle with 2 mL of Agdia extraction buffer and assay performed as per manufacturer’s instructions (Agdia Inc., Elkhart, IN). Digital ELISA readings at 490 nm were taken using a microplate reader (Bio-Rad, Hercules, CA). If mass of petiole material was less than 0.2 g (during the first-year study), material from the petiolar sinus was added to the sample. ELISA was also used as the initial screen of 96 wild and ornamental plants found in and around the vineyard during spring 2004. Either petiole or stem material was used depending on plant morphology. ELISA was used to screen for PD-positive insects trapped in both spring and summer. Large insects (Homalodisca coagulata) were dissected and their head and digestive canal added to 400 μL of extraction buffer, whereas smaller insects were crushed whole in the same volume of extraction buffer. ELISA values two-fold or greater than background (leaves from greenhouse Black Spanish vines, guaranteed free of X. fastidiosa) were considered positive. Guaranteed PD-free H. coagulata insects (APHIS–USDA) were used as the negative control for insect ELISA studies.
Indirect immunofluorescence.
The immunofluorescence procedure on sap is a novel technique designed by our lab to visualize X. fastidiosa directly from xylem fluid (Carbajal et al. 2004). The positive X. fastidiosa control was a pure culture isolated from a groundsel-tree plant (Baccharis). The antibody used was X. fastidiosa specific (Agdia Inc.). A common laboratory strain of Escherichia coli was used as a control to verify lack of nonspecific association with gram-negative bacteria (Carbajal et al. 2004).
PCR.
After initial ELISA readings were performed on grapevine petioles, wild plants and insects, a subset of ELISA positives were also evaluated using PCR. Blanc du Bois, Cynthiana, and Black Spanish samples with the highest ELISA values were confirmed PCR positive in July 2004 (3 to 4 PCR reactions per each positive vine). PCR was also run on three samples (one from each variety) with negative ELISA values. Additionally, PCR was run on a subset of 20 of the insects collected in June 2004 and on wild plant species from around the vineyard to ensure positive ELISA readings were not false positives.
A sample of 200 μL of the ELISA extraction buffer containing 50 mg/mL polyvinylpolypyrrolidone (Sigma Chemical Co., St. Louis, MO) was added to 200 mg ground petiole tissue. DNA extraction was then performed on the slurry of petiole, ELISA extraction buffer, and polyvinyl-polypyrrolidone using a microbial DNA isolation kit (Mo Bio, Inc., Carlsbad, CA). Total plant and bacterial DNA isolated from petiole tissue was estimated as ~10 μg per 50 μL extraction by comparative ethidium bromide staining. Xylella-specific PCR primers RST31 and RST33 and reaction conditions were used for PCR as described by (Minsavage et al. 1994). Five μL of extracted DNA, two μL of each primer, and 16 μL of water were added to puReTaq Ready-to-Go beads (Amersham Biosciences, Piscataway, NJ) for amplification. DNA extracted from stationary phase X. fastidiosa and E. coli cultures was diluted 1:10,000 for PCR reactions as positive and negative controls. PCR products were electrophoresed in 1.5% agarose gels and stained with ethidium bromide.
Results
Viticultural performance of hybrids.
Although Blanc du Bois vines were older than Cynthiana or Black Spanish vines analyzed in 2003 and 2004, valuable field information was obtained at this site, including 10-year historical data from Blanc du Bois and an original planting of Cynthiana (Figure 1⇓). Crop yields collected by the grower demonstrated stable yields from Blanc du Bois vines for a decade. In contrast, Cynthiana crop yields decreased dramatically over the same period. New Cynthiana and Black Spanish vines were planted in 1999. In summer 2003 extensive crop yield, pruning weights, and ELISA readings were made from 27 vines of each of the three hybrids. These vines included the new Cynthiana and Black Spanish (both 5 years old) and the Blanc du Bois (10 years old) (Figure 1⇓). During 2003, data vines of Blanc du Bois and Black Spanish had average yields of 5.7 and 7.4 kg/vine, respectively, whereas Cynthiana had significantly poorer yields of 1.0 kg/vine (Table 1⇓).
Yield and pruning weights for each of three hybrid varieties at Austin County Vineyard, Texas. Numbers are the means of 27 data vines selected in a randomized block design for each variety within the vineyard.
Ten years of grower-collected crop data for Austin County Vineyard, Texas.
X. fastidiosa levels.
In August 2003 and 2004 all Cynthiana vines showed mild symptoms of PD. Several Black Spanish and Blanc du Bois vines also showed PD symptoms, including scorched leaves localized on specific shoots. Late-summer PD symptoms were very mild and did not result in vine death the following year, demonstrating that despite detectable infection, long-term deleterious effects were not immediately apparent. In summer 2003, crop yields of specific vines in these three varieties were compared (Figure 2A⇓), and in August 2003, X. fastidiosa levels of the same vines were determined by ELISA (Figure 2B⇓). ELISA results suggested that many sampled vines were positive for X. fastidiosa.
(A) Crop yield of three varieties for summer 2003. Means represent the average of 27 vines. (B) Absorbance readings for ELISA assays of X. fastidiosa run on grape petioles collected 18 Aug 2003 on the same 27 vines. Tissue from 3 petioles/vine used in each assay. Higher numbers represent greater bacteria within leaves.
Cynthiana had high bacterial titer and produced less crop compared to Black Spanish or to Blanc du Bois (Figure 2⇑). A regression of crop yield and ELISA levels from the 27 Cynthiana vines was not significantly different from zero (p = 0.06). Although not significant at the p = 0.05 level, numerous vines exhibited high bacterial levels and low yields (Figure 3⇓).
Linear regression of yield versus ELISA numbers for 27 Cynthiana data vines. All vines had similar mild PD symptoms at the end of the growing season. Although the slope does not vary significantly from 0, the graph suggests low-yielding vines have higher bacterial levels (p = 0.06). Only three vines with low yields did not have high ELISA levels. Dashed lines are 90 and 95% prediction bands, suggesting a negative correlation between bacterial level and yield in Cynthiana at this vineyard.
The maximum reliable sensitivity of the commercially available ELISA kit (Agdia) was ~104 cells/mL, in agreement with a previous report (Minsavage et al. 1994). The theoretical limit of PCR is 10 bacteria per PCR reaction (Pooler and Hartung 1995). However, we routinely detected the equivalent of only 104 cells/mL by PCR, equivalent to the lower detection limit of our ELISA (data not shown). All three varieties exhibited positive ELISA readings in 2003, suggesting they were capable of maintaining an average X. fastidiosa load at or above 104 cells. PCR and immunofluorescence were used as secondary confirmation of ELISA analysis, particularly for Black Spanish and Blanc du Bois, which appeared to have lower titers.
Immunofluorescence of sap samples from ELISA-positive grape petioles showed rod-shaped cells, consistent with the morphology of X. fastidiosa. Samples from ELISA-negative petioles displayed no detectable fluorescent cells. Immunofluorescence of undiluted sap from ELISA-positive petioles collected during summer 2004 routinely displayed >100 X. fastidiosa per field of view in Cynthiana petioles, whereas sap from Black Spanish and Blanc du Bois showed only a few scattered cells (Figure 4⇓). Samples depicted in Figure 4⇓ were from ELISA-positive petiole tests and ELISA values obtained from Cynthiana corresponded to an order of magnitude higher X. fastidiosa cell concentration, consistent with the dramatically higher cell counts observed by immunofluorescence.
Verification of X. fastidiosa within the petiole sap of all three grape varieties by indirect immunofluorescence. Sap from Cynthiana shows large population of X. fastidiosa, whereas sap from Black Spanish and Blanc du Bois contain very few cells. Fields are representative from multiple observations of ELISA positive samples.
ELISA of petioles from 2004 was followed by DNA extraction from the same petiole sample and PCR using Xylella-specific primers RST31 and RST33. Petiole samples testing positive by ELISA resulted in a 733 bp amplicon (Figure 5⇓). PCR from ELISA negative samples did not show bands at the same time that the highest ELISA plant was PCR positive (data not shown).
Gel showing positive PCR results (733 bp) from petiole samples of all three varieties previously shown to be positive by ELISA. Lanes: Neg., PCR of E. coli DNA; 1, Blanc du Bois; 2, Black Spanish; 3, Cynthiana; L, 100 bp ladder; Pos., DNA extraction and PCR of X. fastidiosa lab culture. DNA extractions of colonies were diluted 10,000x.
Temporal changes in Xylella.
It was speculated that higher X. fastidiosa levels detected in Cynthiana petioles in August 2003 might translate into higher spring titers. Assuming xylem bacterial levels for all vines are severely reduced in spring because of pruning of infected tissues, high bacterial levels in very early spring could suggest bacteria were overwintering in woody tissue. Correspondingly, ELISA data from small spring shoots (~20 to 30 cm) showed very low X. fastidiosa levels in all varieties, only 6 of the 27 Cynthiana vines having positive ELISA values (Figure 6⇓). Although their average was negative, some Cynthiana vines appeared to have bacterial concentrations >104 cells/mL in earliest spring shoots.
X. fastidiosa levels over the course of the growing season as determined by the average of 27 vines of each of three varieties. Approximate concentration of bacterial cells per mL of plant tissue shown on the right y-axis (log). Concentrations are based on average ELISA readings from known dilutions of X. fastidiosa lab cultures.
Data from June petiole samples showed average bacterial levels in Blanc du Bois and Black Spanish to be below the positive threshold, with average Cynthiana vines testing at the positive threshold. By August the levels of ELISA readings in Cynthiana and Black Spanish were significantly above the designated positive threshold (Figure 6⇑). By comparing the ELISA petiole sample to known dilutions of X. fastidiosa cultures, the average bacterial levels within petioles were estimated to be an order of magnitude higher in August than the following June. Bacterial concentration approached an average of 105 cells/mL for Cynthiana, with some samples scoring >107 cells/mL and average levels of bacteria within Black Spanish were also significantly >104 cells/mL. Interestingly, the average Blanc du Bois ELISA levels were significantly higher than the positive threshold in August 2003, but not in August 2004 (Figure 6⇑). Despite the fact that the average Blanc du Bois petiole in August 2004 was not above the positive threshold, several of the 27 individual vines were positive.
Given the faster increase in bacterial levels within Cynthiana vines and higher bacterial levels after harvest, we speculated that detectable X. fastidiosa populations may overwinter in Cynthiana wood. If Cynthiana were able to host viable X. fastidiosa cells within its cordons and trunks, then it could explain faster colonization of tissues the following season. In April 2004, 0.2 g of wood was collected from cross sections of cordons, trunks, or canes of 27 Cynthiana buffer vines (not part of the 27 data vines). The wood was soaked in water for 24 hr, and ELISA tests performed on this solution. ELISA values were positive (twice the negative control reading) for either cordon or trunk wood of three Cynthiana vines. Random sampling of cordon and trunk wood of several Blanc du Bois and Black Spanish vines never resulted in a positive ELISA for X. fastidiosa (data not shown). Although these data do not unequivocally allow us to state that X. fastidiosa can overwinter in the wood of this hybrid, the finding that occasional Cynthiana vines, but not Blanc du Bois and Black Spanish, tested positive suggest that it is a distinct possibility requiring additional investigation.
Evaluation of potential plant reservoirs.
An important component to understanding disease pressure in this Gulf Coast vineyard was to determine if any local plants were serving as sources of inoculum. To that end, 96 native and ornamental plants from within or around the vineyard were surveyed between spring and early summer 2004. Xylella fastidiosa-negative, greenhouse-grown Black Spanish leaves routinely scored an average of 0.05 via ELISA. It was therefore decided that any native or ornamentals with ELISA values of 0.10 or greater absorbance would be further evaluated using immunofluorescence and PCR.
From the original 96 plants collected, 14 different species were considered for further study (Table 2⇓), including several questionable plants that approached the positive ELISA number but that were not greater than the 0.10 absorbance threshold. All species testing below the ELISA threshold were likewise confirmed negative by PCR or indirect immunofluorescence, including dayflower (Commelina erecta), peppervine (Ampelopsis aborea), little quaking grass (Briza minor), and Chinese tallow tree (Sapium sebiferum). Plants that tested positive by ELISA (>0.10) and confirmed by PCR included Yaupon holly (Ilex vomitoria), Mustang grape (Vitis mustangensis), American sycamore (Platanus occidentalis), southern dewberry (Rubus trivialis), frogfruit (Lippia nodiflora), and Japanese honeysuckle (Lonicera japonica). Yellow puff (Neptunia lutea) and crape-myrtle (Lagerstroemia indica) were confirmed positive by ELISA and the visualization of X. fastidiosa in sap by immunofluorescence, but PCR was not performed on these two species. Two species scored positive by ELISA but appeared to be negative by PCR and indirect immunofluorescence, including periwinkle (Vinca minor) and Spanish moss (Tillandsia usneoides) (Table 2⇓). In sum, eight of the 96 plants tested positive for X. fastidiosa in two out of three detection assays. Importantly, Yaupon holly, Mustang grape, American sycamore, and southern dewberry together represented a high percentage of the total plant biomass around this vineyard.
Native and ornamental plants from Austin County vineyard that were surveyed for the presence of X. fastidiosa using ELISA. Of the 96 plants tested, those close to or above the positive cut off (>0.10) were further evaluated with PCR or immunofluorescence (IF) or both. If ELISA and PCR were negative, then IF was not performed (NP). In several cases there was not enough sap or plant material to perform both IF and PCR of ELISA-positive material.
Evaluation of potential insect vectors.
As little is known about potential PD insect vectors along the Texas Gulf Coast, it was important to identify which xylem-feeding insect species were present and what fraction might be vectoring X. fastidiosa. Traps collected in April to May contained only two Homalodisca coagulata (glassy-winged sharpshooters) and 14 Oncometopia orbona sharpshooters; all 16 tested negative by ELISA. A second batch of traps collected after two weeks of exposure in late May and early June captured 100 H. coagulata and 40 O. orbona (Figure 7⇓). At this time the vast majority of H. coagulata tested positive, with many scoring as two or three times over the positive threshold. A smaller fraction of O. orbona sharpshooters also exhibited ELISA values several-fold over the positive threshold (Figure 7⇓). PCR tests were inconsistent at detecting ELISA-positive insects. However, we were unable to accurately retest PCR given the small volume of ELISA-positive insect material remaining. Additionally, indirect immunofluorescence was unsuccessful when performed on insect extracts.
Sharpshooter species collected in the vineyard during two weeks in early summer 2004 (last week May, first week June) at increasing ELISA readings. Guaranteed clean lab insects routinely gave ELISA absorbance readings of 0.05; all insects >0.10 absorbance were considered potential positives. Many insects scored 2 to 3x over the positive threshold.
Discussion
Low yield and high bacterial levels in Cynthiana.
Cynthiana vines in this vineyard had lower crop yields and higher X. fastidiosa titers compared to Black Spanish of the same age and Blanc du Bois vines three times their age. One explanation may be differential resistance to PD among hybrids: Cynthiana does not survive as long as other American hybrids in greenhouse inoculation studies (A. Walker, 2002, personal communication). The linear regression of ELISA values by yield in Cynthiana at this vineyard indicates a negative correlation between yield and bacterial levels (p = 0.06). Only three low-yielding vines did not have high bacterial levels (Figure 3⇑). However, given only three petioles from an entire vine were sampled, there was a possibility of false negative if “clean” leaves were randomly selected and the vine was not uniformly infected. It is tempting to generalize that Cynthiana has lower PD resistance compared to Black Spanish or Blanc du Bois and that higher bacterial levels negatively impact yield. However, whether this is a causal or correlative relationship cannot be demonstrated in a descriptive vineyard evaluation such as this one. Alternatively, Cynthiana may be poorly suited for this location and additional unknown environmental factors negatively impacting yield may also be increasing Cynthiana’s susceptibility to increased X. fastidiosa populations.
Bacterial level comparison among hybrids.
ELISA, PCR, and immunofluorescence analysis demonstrated that X. fastidiosa cells are detectable in petioles of Black Spanish, Blanc du Bois, and Cynthiana within the PD hot zone. Immunofluorescence with a X. fastidiosa-specific antibody allows for visualization of these rod-shaped xylem endophytes directly from sap. PCR confirmed the presence of X. fastidiosa DNA from ELISA-positive petioles. These two tests confirmed that the occasional high ELISA values from Black Spanish and Blanc du Bois were likely not false positives. Collectively, visualization of X. fastidiosa cells from sap, positive ELISA, and positive PCR confirm that all three varieties may harbor X. fastidiosa, but ELISA data suggest Cynthiana does so at a consistently higher level than either Blanc du Bois or Black Spanish. Reliance on ELISA data for predictions of bacterial levels in this vineyard is supported by recent work showing that X. fastidiosa ELISA values are highly predictive of bacterial populations (Krivanek and Walker 2005).
In August 2003 all three hybrids had high ELISA levels, but Blanc du Bois and Black Spanish had fewer positive vines and lower average ELISA values (Figure 6⇑). In August 2004 both Black Spanish and Cynthiana had positive mean ELISA levels, but Cynthiana again had a much higher mean because of more positive vines. In 2004 Blanc du Bois had a mean ELISA value similar to the positive threshold (Figure 6⇑), perhaps because white-fruited Blanc du Bois is harvested a month earlier than Cynthiana and Black Spanish. Following the July harvest and several summer rainstorms, a new flush of growth in Blanc du Bois may have resulted in lower than expected August X. fastidiosa levels. Despite low mean ELISA levels in Blanc du Bois, several samples were positive and these were confirmed positive by immunofluorescence and PCR.
Distinct increase in Cynthiana bacterial levels.
Although bacterial levels were positive by the end of the 2003 season for all three varieties, the levels measured in new petioles the following spring were negative (Figure 6⇑). It was expected that spring X. fastidiosa populations would be low, since much of the infected leaf and shoot material from the previous growing season would senesce or be pruned away. Additionally, there were few sharpshooters in the vineyard in early spring, and it is unlikely that vines would be reinfected until vector numbers increased in early summer.
Multiple studies have shown that the concentration of X. fastidiosa increases over the growing season and that susceptible varieties become infected sooner (Hopkins and Thompson 1984, Fry and Milholland 1990). Rapid shoot growth in all three American hybrids from April through June likely keeps bacterial levels in petioles low until August when shoot growth slows. Given the slow growth of X. fastidiosa, it is possible that rapid spring shoot growth is exceeding bacterial growth rates. However, once shoot growth stops, bacterial levels increase faster in Cynthiana than in either Black Spanish or Blanc du Bois.
Why were bacteria present at higher concentrations and in more vines of Cynthiana than Black Spanish or Blanc du Bois? Based on preliminary data of woody tissue, X. fastidiosa may be present in the woody xylem of Cynthiana. If X. fastidiosa is able to survive in trunks or cordons of Cynthiana preferentially over Black Spanish or Blanc du Bois, then it would be an early inoculum source for Cynthiana the following season. Recent inoculation studies in potted susceptible Chardonnay vines have shown that X. fastidiosa is able to move down from the point of inoculation on a shoot into the trunk and up into a new noninoculated shoot (Stevenson et al. 2004). Mild winters along the Gulf Coast could ensure survival of X. fastidiosa in the wood and these populations could then be transported into growing shoots the following spring. Again, symptoms are not likely until shoot growth slows in the summer. This hypothesis is consistent with the mild PD symptoms seen consistently in Cynthiana and only sporadic symptoms in Blanc du Bois or Black Spanish. If the bacteria remain more localized in Black Spanish and Blanc du Bois, then positive readings in this vineyard are due to frequent new inoculations by summer insects with high bacterial titers. Likewise, if Black Spanish and Blanc du Bois have higher resistance to PD, limiting bacterial growth within the plant, most of the localized X. fastidiosa in Black Spanish and Blanc du Bois would be removed with pruning and infections would be based primarily on new vector transmissions.
Recent work shows that high stem bacterial level is correlated with PD susceptibility (Krivanek and Walker 2005), which supports our observation that Cynthiana may not be as PD resistant as Black Spanish and Blanc du Bois at this location. If bacteria were not detected in the winter wood of Blanc du Bois or Black Spanish, but were found in Cynthiana, then this variety is likely harboring bacteria within its stems at least at a level higher than that found in the other two hybrids.
There are other plausible explanations for the higher rate of bacterial accumulation in Cynthiana. It may be a more attractive host for sharpshooters and therefore inoculated more frequently. Additionally, sharpshooter inoculation rates may be the same in all hybrids, but the bacteria may spread faster throughout the vascular system of Cynthiana. Recent work suggests a model for X. fastidiosa movement in plant xylem (Newman et al. 2004). Colonies grow on the xylem wall until xylem flow is impeded, which results in the colonies becoming nutrient starved and producing a diffusible signaling factor. This factor induces production of degradative enzymes that break down border pit membranes, allowing movement of X. fastidiosa into neighboring xylem vessels. These investigators further suggest that in some cases X. fastidiosa may not be able to enter neighboring xylem vessels if the membranes are more resistant to digestion. American hybrids differ in parentage and therefore may differ in this potentially crucial aspect of xylem physiology.
Potential plant reservoirs.
The presence of X. fastidiosa in many Gulf Coast plants is supported by research showing that although the fate of bacterial levels within many plants is unknown, X. fastidiosa has a wide host range (Wistrom and Purcell 2005). A screen of 96 potential plant reservoirs in and around this hybrid vineyard resulted in eight plants testing positive by at least two assay methods. Several of these have been reported to host X. fastidiosa, including sycamore in Florida (Hopkins 1988) and South Carolina (Blake 1993) and species of the genus Rubus (blackberry/dewberry) from California to Florida (Hill and Purcell 1995, Hopkins 1988, Raju et al. 1983). X. fastidiosa was reported in spring samples of blackberry in two different grapegrowing counties in California (Baumgartner and Warren 2005), suggesting that this plant may serve as a long-term reservoir for X. fastidiosa.
Yaupon holly, a common Gulf Coast plant, tested positive by ELISA and was confirmed positive by both PCR and immunofluorescence. Yaupon holly is evergreen, and in the warmer Gulf climate it may represent a continuous source of inoculum, although we did not observe any sharpshooters on this plant during spring or summer.
Other potential reservoirs that tested positive by ELISA and PCR included Mustang grape, American sycamore, southern dewberry, frogfruit, and Japanese honeysuckle. Xylella fastidiosa levels, above the threshold needed to infect sharpshooters, have been reported in late-summer samples of wild grapevine and blackberry (Baumgartner and Warren 2005). Given that Mustang grape and southern dewberry are particularly abundant in Gulf plant communities and that they tested positive in our study makes them a strong potential X. fastidiosa reservoir. Finally, crape-myrtle tested positive by ELISA and immunofluorescence, and, although not confirmed by PCR, should be considered a potential reservoir worthy of further investigation given that it is a common plant along the Gulf Coast.
Identification of X. fastidiosa cells in a plant does not immediately implicate that plant as the bacterial reservoir for nearby grape. Numerous studies point to genetic differences between X. fastidiosa strains (Hendson et al. 2001, Rodrigues et al. 2003), and other research has shown different strains do not cause infection among all susceptible plant species (Almeida and Purcell 2003, Purcell et al. 1999). Members of the Texas Pierce’s Disease Research and Education Program have established a collaborative and long-term goal of collecting X. fastidiosa strains from cultivated vines and plant reservoirs from around the state, characterizing the strains and describing their diversity. It may be that some potential plant reservoirs in the Gulf Coast only harbor strains that will not infect grapes (V. vinifera or American hybrids). However, until that can be shown unequivocally by genetic comparison or inoculation studies, all X. fastidiosa positive plants should be considered possible sources of X. fastidiosa in the region. Given the speed at which V. vinifera vines die in the Gulf Coast area (usually in their first year), the strains within some plant reservoirs are likely infectious to grape.
Two potential insect vectors.
Our insect data suggests that there is either a local emergence or a migration of H. coagulata in early summer, and during this time many of these insects become inoculated with X. fastidiosa. Given the number of positive plants in the area, it most likely occurs from feeding on nearby plant reservoirs. By June, total numbers of vineyard-captured H. coagulata greatly increased, with a concomitant increase in ELISA-positive insects, some with readings three-fold or greater than the positive threshold. Of additional interest is the fact that a smaller sharpshooter species, O. orbona, was found in significant numbers in June and that several of these insects were also positive. This smaller sharpshooter may not inoculate plants as efficiently as H. coagulata, but until controlled inoculation experiments can be conducted, it should be considered a potential vector for PD.
Conclusion
Crop yields of Blanc du Bois, Cynthiana, and Black Spanish from an area of intense Pierce’s disease pressure suggest that Blanc du Bois and Black Spanish may be viticulturally more productive than Cynthiana along the Gulf Coast. Successful establishment of Cynthiana (Norton) in other states may be due to cold winters that therapeutically reduce insect vectors or bacterial levels within plants or both. Positive PCR results from petioles and visualization of X. fastidiosa within xylem sap of all three varieties also indicate that under intense disease pressure all of these hybrid varieties are susceptible to colonization by X. fastidiosa. Given that these varieties are hybrids with native vines, xylem invasion by X. fastidiosa and subsequent mild PD symptoms does not necessarily imply lethal susceptibility to PD. Despite testing positive for X. fastidiosa at the end of the growing season, these three hybrids have not succumbed to a PD infection. Cynthiana has high levels of bacteria and low yields at this site and is either more susceptible to X. fastidiosa or under some environmental stress, which is correlated with higher X. fastidiosa levels. The varieties Black Spanish and Blanc du Bois remain healthy despite some positive results at the end of the season, while existing in a hot zone of positive plant reservoirs and in proximity to a neighboring block of Cynthiana vines with very high X. fastidiosa levels. Black Spanish and Blanc du Bois may be very good hybrid varieties along the Gulf Coast or in areas of heavy PD pressure. Xylella fastidiosa was detected in many native and ornamental plants in and around this particular vineyard. Significant bacterial populations were also found in two species of sharpshooters, H. coagulata and O. orbona, implicating these xylem-feeders as possible vectors of X. fastidiosa in the PD hot zone.
Footnotes
-
Acknowledgments: Research was conducted at the University of Houston-Downtown with primary funding from the Texas Pierce’s Disease Research and Education Program, APHIS-USDA. Additional supply funds and faculty/student research stipends were provided by the U.S. Army Research Office, HBCU/MI Infrastructure Support Program (W911NF-04-1-0024), and the National Science Foundation, DUE STEM Talent Expansion Program (0336612).
-
The authors thank Jerry and Cozette Watson of Austin County Vineyard for access to their vines, Dr. Kevin Morano for assistance with indirect immunofluorescence, Kristi Montgomery for help evaluating DNA extraction techniques to maximize PCR results, Dr. Andrew Walker (UC Davis) for providing Xylella fastidiosa-negative greenhouse grape leaves, Dr. Isabelle Lauziere (Texas A&M) for providing X. fastidiosa-negative, lab-reared glassy-winged sharpshooters, and Jim Kamas (Texas Cooperative Extension) for help with insect identification.
- Received January 2005.
- Revision received September 2005.
- Revision received February 2006.
- Copyright © 2006 by the American Society for Enology and Viticulture
Literature Cited
Vol 57 Issue 3
