Abstract
Vineyard damage due to grape phylloxera, Daktulosphaira vitifoliae Fitch, has been controlled by resistant rootstocks for over 100 years. There are now a wide range of rootstocks used in California vineyards since the collapse of AXR#1. To study the effect of this rootstock diversity on phylloxera genetics and possible host adaptation, a set of microsatellite primers were developed to augment the four produced by Corrie and colleagues (2002). In order to develop more microsatellite loci to improve the sensitivity and effectiveness of these markers for use in genetic diversity and rootstock adaptation studies, a subtractive-based hybridization strategy was used to construct microsatellite enriched genomic libraries from grape phylloxera DNA. Fifty loci were identified for primer design. Nineteen produced good PCR products, seven of which reliably detected polymorphisms across the 32 grape phylloxera populations tested. These seven SSR loci were used to distinguish genetic diversity in California and European grape phylloxera populations. Results confirm the utility of these loci for analyzing genetic diversity, “finger-printing” strains, and studying host associations. A significant deviation from the Hardy-Weinberg equilibrium for the tested California populations suggests that parthenogenesis is perhaps the primary, if not only, reproductive system in California.
Grape phylloxera, Daktulosphaira vitifoliae Fitch (Homoptera: Phylloxeridae), are found throughout the Americas where they appear to have coevolved with the endemic Vitis species (Wapshere and Helm 1987). In the 1850s, this insect was inadvertently imported to Europe where it quickly devastated the highly susceptible V. vinifera vineyards. By the end of the 19th century, phylloxera had been introduced into most of the world’s grape-growing regions. Phylloxera have a complex life cycle that includes parthenogenetic generations on the roots or leaves and the possible occurrence of a sexual phase that may link the asexual root and leaf forms. Environmental stress related to availability of roots to feed on probably triggers the production of alate (winged) individuals, although the nature of this trigger is not fully understood. These alate individuals are capable of laying male and female eggs, which hatch into sexual forms that lack wings and mouthparts. After fertilization, these females lay a single diapausing egg, which hatches the following year and begins the leaf-galling asexual cycle. However, this two-stage, bridged life cycle has rarely been observed (Granett et al. 2001). A survey in the southwestern United States (Downie and Granett 1998) found that only the leaf-galling portion of the life cycle exists on V. arizonica. Alate forms were not found in these galls and the observed sexual forms were produced by nonwinged parthenogenetic females. Mating was presumed to occur near the hatch site. In California, a similar pattern was seen on native vines in the Death Valley area; however, only the asexual root-feeding portion of the life cycle has been reported in grapegrowing regions (Davidson and Nougaret 1921, Granett et al. 2001).
Grape phylloxera are most damaging on the mature roots of susceptible grape species where they create galls (tuberosities) that swell and crack, allowing entry of soil-borne fungi, which decay large portions of the root system and lead to eventual vine death (Omer et al. 1999). Resistant rootstocks were developed from American grape species over 100 years ago that prevent tuberosity development (Granett et al. 2001), although rootstocks often allow galling on feeder roots (nodosities). The use of resistant rootstocks has proven to be a durable solution to phylloxera for more than 100 years. In California, information regarding grape phylloxera genetic diversity within and among fields and regions is not clear. Advances in the understanding of phylloxera genetic diversity will aid in our understanding of the potential for genetic selection within agricultural systems and provide insights into why rootstocks have resisted phylloxera for so long.
Corrie et al. (2002) identified four microsatellite loci and used them to study the genetic structure of phylloxera (Corrie and Hoffmann 2004) and host-associated clones (Corrie et al. 2003). Microsatellite markers, also known as simple sequence repeat (SSR) markers, have advantages over other marker systems because of their polymorphic and highly reproducible nature and the ease of comparing data across laboratories. In this study, we identified 19 additional microsatellite loci, seven of which proved to be useful in detecting polymorphisms among phylloxera populations, to augment those identified by Corrie et al. (2002). These additional SSR markers will add precision and sensitivity to studies on the adaptation of phylloxera strains to rootstock hosts and the genetic diversity among populations and strains. These new markers were tested by evaluating the genetic diversity of a set of grape phylloxera isolates from California and Europe and estimating genetic distances among them. Observed and expected heterozygosity were calculated for the California phylloxera populations. We also tested the assumption that parthenogenetic reproduction predominates in California grape phylloxera populations.
Materials and Methods
Sample collection.
Thirty-two phylloxera samples were obtained for this study (Table 1⇓). Nine of these samples were from Europe: three from France and two from Germany (supplied by A. Forneck, Institute of Horticulture, Fruit-growing and Viticulture, Vienna, Austria) and four from Hungary (supplied by L. Kocsis, University of Veszprém, Keszthely, Hungary). The European samples were obtained by bulking phylloxera eggs extracted from leaf galls on one leaf. Twenty-three samples were collected from California vineyards by the authors. These samples were obtained by collecting eggs from young field-collected roots of various rootstocks.
Thirty-two grape phylloxera samples, collection locations, and grape hosts used in this study.
SSR-enriched genomic library construction.
Microsatellite-enriched phylloxera genomic libraries were constructed in order to develop SSR markers. The genomic DNA was isolated from bulked phylloxera eggs according to the method described by Lin and Walker (1996). Eight μg of genomic DNA was equally divided into two tubes. One was digested with 20 U of Dpn II and the other with Rsa-I. Digested DNA samples were separated on a 1.2% agarose gel and fragments ranging from 400 to 600 bp in size were recovered from the gels. Each restricted DNA fragment was pooled and its concentration was determined with a fluorometer using PicoGreen dye; 1 μg of genomic DNA (average size of 500 bp) contains 6.15 pmoles of ends of genomic DNA. For the Dpn II restricted fragments, a 1:2 ratio of double-stranded adaptor was used (Oligo A, 5′-GCGGTACCC GGGAAG CTT GG and Oligo B, 5′-GATCCC AAGCTTCCC GGGTACCGC). For the blunt-end Rsa-I restricted fragments, a 1:10 ratio of double-stranded adaptor was used (Oligo A, 5′-GCGGTACCCGGGAAGCTTGG and Oligo C, 5′-CCAAGCTTCCCGGGTACCGC). The ligation reaction was performed overnight at 16°C. Limited cycles of PCR amplification were performed to amplify the ligation mixture (5 μL of ligation mixture with 95 μL of reaction mixture containing 1X reaction buffer, 2 mM MgCl2, 0.2 mM dNTP, 0.5 U Taq Gold polymerase [Applied Biosystems, Foster City, CA], with 10 pmole Oligo A, 5′-GCGGTACCCGGG AAGCTTGG). The thermal cycler profile was 94°C for 6 min, followed by 10 cycles at 94°C for 30 sec, 55°C for 1 min, and 72°C for 1 min. Two PCR products were then pooled and used to construct SSR-enriched genomic libraries. Eight different types of synthesized SSR oligo probes with 5′-biotin labeled (CA)15, (CT)15, (CTT)10, (GCA)10, (ATT)10, (ATTT)8, (GATA)8, and (AAAG)8 were used to enrich the libraries. Each 5′-biotin labeled probe was hybridized with pooled PCR products for SSR motifs of interest following the protocol of Tozaki et al. (2000) and Li and Kijima (2002). After subtractive hybridization, enriched fragments were ligated with pGEM-T Easy vector (Promega Biosciences, San Luis Obispo, CA) and transformed into Escherichia coli. Results of colonies from eight libraries were screened using a three-primer colony PCR method following the PCR conditions described above. For each library, 10 pmole of M13 forward/reverse primers and 20 pmole of a corresponding SSR oligo primer were used and set at 55°C annealing temperature. Reactions were conducted in an ABI 9700 thermal cycler (Applied Biosystems). PCR products that contained two bands were considered as potentially containing a SSR and were selected for sequencing.
SSR primer design and PCR testing.
Sequence files containing at least five or more repeat units with a 100 to 200 bp flanking sequence were selected for primer design. The Primer Premier 5 software (Premier Biosoft International, Palo Alto, CA) was used with the parameters: GC = 50%, Tm = 58°C, primer length ≈20 bp, and self dimer/cross dimmer ΔG = −5 kcal/mol and amplicon sizes ranging from 150 to 350 bp for every primer design.
Designed primers were tested in PCR reactions, consisting of 20 μL volumes containing 1.5 mM MgCl2, 0.2 mM dNTP, 0.2 U AmpliTaq Gold polymerase, 1X reaction buffer (Applied Biosystems), and 10 pmole SSR primer with 2 μL of phylloxera genomic DNA (10 ng/μL). PCR reactions were conducted in a ABI 9700 thermal cycler (Applied Biosystems) with temperature profile as follows: initial denaturation step at 95°C for 6 min followed by 30 cycles of 95°C for 30 sec, 55°C for 30 sec, and 72°C for 1 min, and a final extension at 72°C for 7 min. The SSR products were then mixed with sample-loading dye (10 mM NaOH, 95% formamide, 0.05% bromophenol blue, and 0.05% xylene cyanol) at a 1:2 ratio. A 2 μL aliquot of this mixture was resolved on 5% of polyacrylamide gel. The gel was run at a constant 100 watts for 2.5 to 3.5 hr depending on the amplicon size together with a known vector’s sequencing products (PUC19) as a molecular sizing marker. Silver staining (Promega Biosciences) was used to visualize the gels and evaluate and record polymorphisms among grape phylloxera samples.
Genetic analysis.
Allele sizes for each locus were assigned from the data obtained from the sequencing gels. The changes in allele size at a specific locus reflect sequential additions or deletions of simple repeat units. Therefore, the number of variable units can be calculated for each sample by measuring the amplicon size divided by the repeat size. The allele size, allelic frequencies, number of alleles per locus, observed heterozygosity (Ho), expected heterozygosity (He), and Nei’s expected heterozygosity were estimated using Microsatellite Analyzer, version 3.12 software (Dieringer and Schlotterer 2003). To evaluate genetic distances among phylloxera samples, allelic sizes were converted into a binary data set, which was then used to generate a pair-wise distance matrix. The UPGMA (unweighted pair group method with arithmetic mean) cluster-analysis technique with simple matching coefficient of resemblance was used to evaluate genetic distances. Results were presented as a graphical cluster using NTSYSpc, version 2.01 (Exeter Software, Setauket, NY).
Results
SSR markers.
In this study, we used a subtractive-based hybridization strategy to construct microsatellite-enriched libraries. Of the 1,528 colonies screened, 178 colonies (11.6%) containing SSR motifs were identified with the following repeat motifs listed with their frequencies: CA (27.3%), CT (10.2%), CTT (12.5%), GCA (16.5%), ATT (20.5%), and ATTT (13%). No GATA and AAAG repeat motifs were detected. The results from DNA sequencing confirmed that 151 (85%) of the identified colonies did contain various lengths of expected repeat motifs. We also found other types of repeat motifs from sequencing files and included them in the primer design. Among these 151 colonies, 72 (48%) were not useful either because the repeat number was too short (less than 5) or because of the lack of sufficient flanking sequences on either end for primer design. Thirteen (9%) of the motifs were not useful because their sequencing quality was poor. After the 151 colonies were sequenced, 66 of the sequence files were selected. Among these, 16 sequence files were removed because they contained duplicate loci after the sequence alignment check. Finally, 50 loci were identified for primer design. Based on the PCR test, 19 primers produced clear PCR products (Table 2⇓). When these 19 SSR primers were tested against 32 grape phylloxera samples, 7 of them reliably detected polymorphisms with 2 to 8 alleles per locus. Thus, these additional primers should expand our ability to genetically analyze phylloxera populations.
Details of SSR primers and loci developed for grape phylloxera.
Data analysis.
Genetic diversity analysis was performed on the 32 grape phylloxera populations (Table 1⇑), 9 from Europe and 23 from California. The seven SSR primers (Table 2⇑) detected various levels of the polymorphism. The average number of alleles per polymorphic locus was 3.4, ranging from 2 to 8 alleles per locus. The grape phylloxera genotypes produced across seven SSR loci for the 32 populations are presented in Table 3⇓. The expected and observed heterozygosity for European and California phylloxera at each locus are presented in Table 4⇓. Because of the small sample size of the European group, genetic analysis was only performed on California populations. A chi-square test (χ2 = ∑2 (Ho − He)/He, where Ho (is observed heterozygosity and He is expected heterozygosity) found that a significant deviation from the Hardy-Weinberg equilibrium existed in California grape phylloxera populations (p < 0.05). When the allele sizes were compared among the seven loci, the European and California groups shared 77% of the total alleles. However, there were specific alleles that characterized these groups. For example, the allele 253 bp at the DVSSR4 locus and alleles 273 and 286 bp at the DVSSR3 locus were only detected in California populations, while alleles 246 and 278 bp at the DVSSR3 locus were only found in European grape phylloxera (Table 4⇓).
Allele sizes generated at seven SSR loci for grape phylloxera collections.
Genetic analysis of California and European grape phylloxera at seven microsatellite loci. Alleles labeled in bold indicate exclusive alleles detected in either California or European collections.
A dendrogram based on the UPGMA method was constructed for the 32 grape phylloxera collections. European phylloxera grouped separately from the California grape phylloxera (Figure 1⇓). Relationships among the nine European samples reflected the origins of the samples. Among California collections, samples collected from the Oakville Station, Napa County, were grouped into a single cluster. The cluster was further divided into three subclusters, each distinctly linked to a host cultivar (101-14 Mgt: Napa 4, 5, 6, 13, and 14; 1103P: Napa 7, 8, and 9; Teleki 5C: Napa 10, 11, and 12) from which the samples were collected. Similarly, samples collected from AXR#1 rootstock and own-rooted Chardonnay were grouped. However, four samples collected from own-rooted Chardonnay at the UC Davis vineyards were split into two subclusters, two of them loosely clustered with Napa County samples that were collected from AXR#1 rootstock, while the other Davis samples loosely clustered with phylloxera collected from AXR#1 rootstocks in Mendocino County and Sonoma County. There were no differences detected among four populations collected from AXR#1 roots in Mendocino and Sonoma County vineyards.
Genetic similarity among 32 grape phylloxera collections based on UPGMA cluster analysis from seven microsatellite loci.
Discussion
The failure of the grape rootstock AXR#1 to adapted grape phylloxera strains, designated Biotype B, stimulated molecular genetic studies of California grape phylloxera populations. The initial studies used random amplified polymorphic DNA (RAPD) markers to analyze the genetic diversity of California (Fong et al. 1995) and United States populations (Lin et al. 1999, Downie et al. 2000). These studies found that the genetic diversity of grape phylloxera populations in California was higher than expected given the parthenogenetic life cycle of grape phylloxera in California (Davidson and Nougaret 1921) and that spontaneous mutations are not expected to contribute to high degrees of genetic diversity. Granett et al. (1996) concluded that multiple introductions were likely to be responsible for the degree of genetic diversity observed in California phylloxera.
Forneck et al. (2000) used amplified fragment length polymorphism (AFLP) markers to study European phylloxera and found relatively high levels of polymorphism within and among phylloxera populations from northern and southern Europe. Variation in sequences of mitochondrial markers have also been used to compare genetic relationships among agricultural growing regions (California, Oregon, and Washington) with native populations of V. vulpina from the Atlantic Coast and V. riparia from the northeastern United States. The authors concluded that various levels of genetic variation exist in native and agriculture populations (Downie et al. 2001).
The use of DNA markers to examine the patterns and degrees of genetic variation in agricultural system can provide insights into pest population dynamics over time and space. However, previous studies have relied on dominant markers, which provide no information about allelic recombination and population heterozygosity. Consequently, the deviation of populations from Hardy-Weinberg equilibrium cannot be estimated with dominant markers. Such information is crucial to understanding phylloxera population genetics because of the complicated reproduction modes in phylloxera and unknown factors that trigger the switch between holocyclic and anholocyclic reproduction. This induction of the sexual cycle would help explain the high levels of variation seen in California phylloxera populations if recombination is prevalent in California. Codominant markers are able to record the recombination of alleles and therefore quantify the contribution of meiotic events to observed genetic diversity. Corrie et al. (2002) developed four SSR markers and used them to assess phylloxera populations in Australia. These codominant markers are capable of distinguishing heterozygosity in populations and therefore provide informative details of population structures. Corrie et al. (2002) observed a limited number of genotypes and significant departures from Hardy-Weinberg equilibrium, which led to the conclusion that the predominant reproductive mode of both roots and leaf-galling grape phylloxera was parthenogenesis.
In order to draw more complete conclusions about the role the sexual cycle plays in grape phylloxera diversity and to better identify and characterize grape phylloxera strains, more SSR loci are needed. Seven new SSR loci were developed in this study; this marker system combines an accurate genotyping system with a powerful means of detecting polymorphism throughout the genome. However, development of microsatellite markers is difficult and costly when genomic sequence information is not available. Corrie et al. (2002) used the conventional genomic library cloning method to identify 10 putative repeat loci out of 30,000 screened colonies (0.03%). Four of these loci were developed into PCR-based primers. In the study presented here, a microsatellite-enriched technique was used to enhance SSR locus identification. The efficiency of identifying SSR-containing colonies was about 11%, indicating the method is relatively efficient for the discovery of SSR loci in phylloxera.
When the DNA from European and California phylloxera populations was analyzed with seven SSR loci, 77% of alleles were shared in both groups and 23% of the alleles were unique to either group. The unique alleles suggest that they were derived from independent evolutionary processes. Genetic studies of European phylloxera have suggested that sexual cycles might occur in vineyards (Forneck et al. 2000). However, there is no evidence from California grape phylloxera populations, suggesting that sexual recombination plays a role in genetic diversity. The significant deviation from Hardy-Weinberg equilibrium among the seven SSR loci analyzed and the limited number of allelic recombination types in the studied populations imply that parthenogenesis is the dominant, if not the only, reproductive mode in California populations. These results are similar to studies of Australian phylloxera populations (Corrie et al. 2002).
These results question the sources of genetic diversity in California phylloxera populations. Mutation is one of the possible sources. Increases or decreases in population genetic diversity depend on the rate of mutation and the fitness of the new genotypes in a population. Mutation can also be neutral if mutated loci are not subject to selection pressure. Currently, the mutation rate of grape phylloxera and its contribution to population diversity in California vineyards is not clear (Downie 2003).
Multiple introductions could also explain the relatively high rates of California phylloxera genetic diversity (Granett et al. 1996). Host specialization had been accounted for through the maintenance of high intrapopulation diversity in aphid studies (De Barro et al. 1995). Exotic strains of grape phylloxera are subject to evolutionary pressure from their new environment and hosts. Any new clonally based genetic diversity will be maintained if the new genotype successfully adapts to the environment. A possible example of this phenomenon is the identification of three different genotypes on AXR#1 rootstock in northern California vineyards.
Host plant species have been reported to be an important factor influencing adaptation of races or demes in aphids (Kimberling and Price 1996). Corrie et al. (2002, 2004) used SSR markers to identify strong associations between asexual lineages and host types in vineyards. Grape phylloxera performance on excised root bioassays (Kocsis et al. 1999) and in an aseptic dual culture system (Forneck et al. 2001) demonstrated that phylloxera form host-adapted strains. More recently, Corrie et al. (2003) reported strong associations between a grape host genotype and the asexual lineages identified by microsatellite and mitochondrial markers within a vineyard. In this study, the existence of host-associated genotypes was also suggested by grape phylloxera genotypes associated with the root-stocks 101-14 Mgt, 1103P, and Teleki 5C at the Oakville Station in Napa Valley. These results, however, are inconsistent with an earlier report based on AFLP markers that found no association between phylloxera and a given rootstock host (Forneck et al. 2000). The presence of host-associated genotypes detected at the Oakville Station may suggest that asexual lineages of particular genotypes are the results of selectively adaptive advantages and provide evidence of phylloxera adaptation to rootstock hosts.
Conclusion
The SSR primers developed from this study have proven useful at detecting genetic diversity in California and European grape phylloxera. These multilocus SSR markers can be used to detect host-or region-specific alleles and analysis of these alleles can be used to measure inter- and intrapopulation diversity and structure. This marker system is also well-adapted to the “fingerprinting” of local strains and the monitoring of exotic strains. Understanding genetic diversity and reproductive biology of the pest will help design and develop effective strategies for the pest management. These newly generated SSR markers will now be used to conduct a hierarchical sampling of California grape phylloxera by sampling from multiple vines in multiple vineyards from multiple counties. When coupled with seasonal sampling this survey will detail the spatial and temporal population dynamics of grape phylloxera in California. These preliminary data are being used to support continued study of suspected rootstock host adaptation by phylloxera strains. We also plan to use all available SSR primers to continue studies of phylloxera genetic diversity on national and international levels.
Footnotes
Acknowledgments: The authors gratefully acknowledge funding from the USDA-ARS and the California Grape Rootstock Improvement Commission. They also gratefully acknowledge Dr. Astrid Forneck and Dr. Laszlo Kocsis for the collection and preparation of European phylloxera DNA.
- Received June 2005.
- Revision received October 2005.
- Copyright © 2006 by the American Society for Enology and Viticulture
Literature Cited
Vol 57 Issue 1
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
