Previous Article | Next Article ![]()
Applied and Environmental Microbiology, December 2003, p. 7447-7452, Vol. 69, No. 12
0099-2240/03/$08.00+0 DOI: 10.1128/AEM.69.12.7447-7452.2003
Copyright © 2003, American
Society for
Microbiology. All Rights Reserved.
Department of Environmental Science, Policy and Management, University of California, Berkeley, California 94720
Received 21 April 2003/ Accepted 15 September 2003
|
|
|---|
|
|
|---|
Extensive transmission experiments showed that X. fastidiosa isolates from grapes caused AD and isolates from alfalfa caused PD(12, 30). Davis et al. (6) found that three isolates from grapes and two isolates from almonds inoculated into almonds and grapes caused disease in both hosts. This suggested that PD, ALS, and AD are caused by the same strains. Strains of X. fastidiosa from oleander did not cause PD or ALS but were recovered from mechanically inoculated almond plants in the laboratory (28); neither the tested isolates from grapes nor those from almonds caused disease in or were recovered from oleander (29).
The major leafhopper vectors of X. fastidiosa for grapevines in the Central Valley (Xyphon [Carneocephala] fulgida and Draeculacephalaminerva) (26) are found in almond orchards and transmit the bacterium from almond and grape source plants to healthy almonds and grapes in all four combinations (24). But generally there is no ALS near PD-infected vineyards and vice versa (24). Apparently, another factor besides the presence of vectors may affect disease spread in such situations. An alternative hypothesis to explain the lack of ALS near PD hotspots is that most inoculations of PD strains of X. fastidiosa into almonds do not survive the winter; thus, the number of (symptoms) infections detected would be low in relation to the total number of infections. Data obtained by Davis et al. (6) suggested that this might be the case, since only 2 of 48 successful inoculations of grape strains into almonds in the field survived one winter. Unfortunately, those authors did not use isolates from natural ALS infections in the test because of concerns that the experimental area was located in a region free of ALS. Cold winters have been shown to cure PD in potted grapevines, and higher rates of recovery from infection occurred in vines exposed to winter conditions in areas in which temperatures were lower (25).
Adding to the complexity of the problem, Hendson et al. (10) found (using various genetic approaches) that most isolates from almonds are part of a distinct grouping whose members putatively form two clusters (groups I and II), are distinct from all isolates from grapes and include a minority of isolates from almonds. Thus, the questions about host specificity and whether grape and almond strains have overlapping host ranges are unresolved. We addressed questions raised by previous genetic analyses of X. fastidiosa strains (10) by determining whether various X. fastidiosa isolates from grapes and almonds in California could cause disease and attain similar population levels in both hosts. We also compared the abilities of grape and almond strains to overwinter in almonds and grapes.
(This research was conducted by Rodrigo P. P. Almeida in partial fulfillment of the requirements for a Ph.D. from the University of California, Berkeley.)
|
|
|---|
25°C ±
1°C) greenhouse throughout the experiments unless noted
otherwise. |
View this table: [in a new window] |
TABLE 1. Isolates
of X. fastidiosa used in this study and their growth on PD3
solid medium; pathogenicity, bacterial populations, and overwinter
survival in both host plants
|
Cross-inoculation.
X. fastidiosa suspensions
were prepared by suspending isolates in 1 ml of succinate-citrate
buffer (18) after growth
on PWG solid medium for 1 to 2 weeks, depending on the growth rate of
each isolate. Suspensions were turbid, with estimates of cell
concentrations of 108 to 109 CFU/ml. A
5-µl drop of the suspension was placed on a young stem of the
test plant, and the tissue was pricked through the drop 5 times with a
no. 0 entomological pin
(14). Two sites per plant
were inoculated; each isolate was inoculated into five almond and five
grape plants. Mock inoculations were done with buffer only. After an
incubation period of 2 to 4 months in the greenhouse, one symptomatic
leaf from each plant with PD or ALS symptoms was collected and
processed for X. fastidiosa identification and bacterial
quantification by dilution plating
(13). In cases in which
plants had no symptoms, a mature leaf near one of the inoculation sites
was collected. Samples testing negative by culture assays were tested
twice. All plants were sampled up to 4 months after inoculation. After
this period, plants which were nonsymptomatic and from which X.
fastidiosa was not grown were considered not infected. We
determined that a plant was symptomatic when it had two or more leaves
with typical PD or ALS symptoms. After the 4-month period, all test
plants were moved to an insect-proof screened cage kept outside the
greenhouse and were watered as necessary (November 2002). After leaf
abscission the plants were pruned and were taken in January 2003 to a
cold box (4°C). In February 2003, plants were brought back into
the greenhouse and kept for up to 6 months for symptom development.
X. fastidiosa was again isolated from these plants by the
culture method.
RAPD.
Random amplified polymorphic DNA
(RAPD) assays were performed following the protocols used by Hendson et
al. (10) with minor
modifications. Because over half of the isolates used here have already
been typed with various techniques
(10), we tried to
determine how the nontyped isolates would be grouped in relation to the
typed isolates. Therefore, we only used 4 RAPD primers (OPAA-01,
OPAA-02, OPAA-03, and OPAA-04; Operon Technologies, Inc., Alameda,
Calif.) of the 10 previously determined to allow differentiation of
X. fastidiosa strains
(10). All isolates were
tested at least twice with each primer. We did not score faint bands,
and profiles were scored in a binary format. The FreeTree program
(9) was used for genetic
analysis, and TreeView
(http://taxonomy.zoology.gla.ac.uk/rod/treeview.html)
was used for construction of the diagram. We used the Jaccard
similarity coefficient
(32) to compare pairs of
strains and the neighbor-joining tree-construction method to build the
phenogram, with 250 bootstrap replicates to estimate branch support.
Isolate Ann1 from oleander was used as an outgroup. All isolates were
identified (using the specific RST31-33 primer set)
(22) as X.
fastidiosa; all isolates generated the expected
730-bp
fragment (data not shown).
Overwinter
survival of X. fastidiosa.
Experiments were conducted in the
Armstrong orchard of the University of California at Davis and at the
University of California Kearney Agricultural Center (Parlier, Calif.).
The nine isolates used for this experiment are listed in Fig.
1. Cells were grown for 1 to 2 weeks on PWG solid medium before mechanical
inoculation (14) and
suspended on phosphate-buffered saline to slight turbidity
(108 to 109 CFU/ml) immediately before
inoculations were made. Suspensions were needle inoculated into new
shoots of mature almonds (Non-pareil variety) in April 1997 by pin
pricking through a 5-µl drop of the cell suspension. A total of
8 trees were inoculated in multiple branches at Davis, and 15 trees
were inoculated at Parlier; 20 to 25 stems were inoculated per isolate
in each location. Samples of inoculated plants were collected in the
fall months of 1997, 1998, and 1999 in Davis but only in 1997 and 1998
in Parlier (the orchard was eliminated in 1999), and bacteria were
quantified by dilution plating
(13). Positive samples
were tested only once a year, but stems with negative results were
tested at least twice and usually three times every year. Negative
controls were inoculated with phosphate-buffered saline only
(n = 20). No mock inoculations caused ALS, and no
X. fastidiosa was recovered at any period from these branches.
Efficiency of mechanical inoculation was analyzed by the chi-square
test using the Yates correction for continuity
(1).
![]() View larger version (36K): [in a new window] |
FIG. 1. Rates
of infection by needle inoculation for various X. fastidiosa
isolates in almonds at Davis (solid bars) and Parlier (empty bars) in
California. Percentages represent the proportion of leaves from which
X. fastidiosa was recovered (August to October) after needle
inoculation on stems earlier (April) in the same year. The almond
strains used were Contra Costa (CC), Dixon (DI), ALS4 (AL), and Manteca
(MA); the grape strains used were Tulare (TU), Traver (TR), UCLA (UC),
Conn (CO), and Medeiros (ME). The strain grouping for the almond
clusters follows the results shown in Fig.
2.
|
|
|
|---|
Virulence in grape and
almond.
Because we used the
same bacterial suspension to inoculate grape and almond plants with
each isolate, we assumed that similar numbers of cells were introduced
into each plant. Thus, the differences in the frequency of infection by
the same isolate on different plants probably reflect host specificity
rather than variability in inoculation procedures (Table
1). Isolates from grapes
infected all grapevines inoculated; only 3 (1 plant without symptoms
and 2 plants with only initial foliar symptoms of reddening on leaf
edges) of 50 plants did not show typical PD symptoms. As determined on
the basis of pathogenicity results in grapes and growth on PD3 medium,
three isolates (Fresno-ALS, Stanislaus, and Tulare) from almonds
mimicked strains from grapes. Interestingly, Fresno-ALS caused little
disease in almonds (only one plant showed typical ALS symptoms); in
fact, this isolate was the only one (besides the almond-avirulent Ann1
isolate from oleanders) (29) with low virulence in
almonds.
With the exception of isolates Fresno-ALS, Stanislaus, and Tulare, X. fastidiosa isolated from almonds did not cause PD (Fig. 1). Some of the isolates caused mild foliar symptoms (usually in only one of the five inoculated grapes). The Contra Costa isolate from almond caused mild PD symptoms. Because we kept the plants in a greenhouse with natural light, the experiment had to be completed within 4 months after inoculation due to a shortening photoperiod, which reduces plant growth and slows symptom development. All almond strains caused ALS, but isolates ALS4 and Contra Costa were notably more virulent to almond than any other tested almond or grape strains, causing advanced symptoms of marginal leaf necrosis 2 months after inoculation, at which time leaf scorch caused by all other strains were barely visible. We also observed that symptom development varied for both grape and almond strains in both host plants. For example, isolates UCLA and Baja consistently took 1 month longer to develop PD symptoms than other isolates from grapes.
Bacterial
populations within grapes and almonds.
The numbers of live X.
fastidiosa cells that we recovered from grapes were approximately
10- to 100-fold higher for the strains from grapes than for strains
from almonds (Table 1).
Similar population levels occurred in both hosts for isolates
Fresno-ALS and Contra Costa, but Fresno-ALS caused PD in similarity to
other strains from grapes, whereas Contra Costa was the
sole strain from almonds that consistently caused mild PD symptoms of
reddening and occasional necrosis limited to leaf margins under
greenhouse conditions. We recovered X. fastidiosa from at
least one of the five grapevines inoculated with each of the various
almond strains but not from those inoculated with the oleander strain.
From almonds, we recovered all strains with approximately identical
population sizes and frequencies. The exception was the oleander
strain, which we recovered only in low numbers (103 to
104 CFU/g of tissue) and which never caused ALS symptoms in
three of the five isolate Ann1-inoculated almond plants from leaves
that had grown after inoculation, indicating that systemic movement of
the bacterium had occurred. Grape strains overwintered more
successfully in grapes (57 of 65 plants) than in almonds (9 of 59
plants by chi-square test [P < 0.001; 1
df]), as did almond strains within almonds (22 of 39) compared to
almond strains in grapes (2 of 18 plants by chi-square test
[P < 0.001; 1
df]).
RAPD and enzyme
restriction.
We confirmed the
trends observed by Hendson et al.
(10): all isolates from
grapes were grouped together genetically, but most almond isolates were
distinctive (Fig.
2). Three isolates from almonds, Tulare, Stanislaus, and Fresno-ALS, were
grouped with grape strains and behaved on media and within plants
similarly to grape strains. Even though we used only four RAPD primers,
the two coastal California isolates from grapes (Conn Creek and STL)
were genetically distinct from Central Valley and southern California
isolates from grapes. The two almond groups identified previously
(10) were also separated
in our analysis, with the exception of the Manteca isolate, which
behaved as a strain from almonds in biological assays and in our RAPD
assay but was previously grouped with strains from grapes; therefore,
we believe that the same isolate was not used in both studies. When the
RST31-33 PCR-amplified product was digested with RsaI, all
almond strains generated two fragments (
550 and
200
bp) and no grape strain was cut (data not shown), as previously
reported
(10).
![]() View larger version (17K): [in a new window] |
FIG. 2. Tree
determined on the basis of distance values obtained by RAPD-PCR
analysis using the Jaccard similarity coefficient. Bootstrap values
(percent of 250 repetitions) are shown for each branch point. Vertical
lines and groupings indicate whether associated strains in the diagram
(i) grew on PD3 solid medium (PD3+) or not (PD3-), (ii)
caused PD (PD+) or not (PD-), or (iii) caused ALS
(ALS+).
|
At Parlier, a total of 22 of 23 infections for all almond group strains combined survived the 1997-to-1998 winter. A total of 45 of 51 plants infected with strains from the grape cluster survived the winter. For the Davis location we had data for 1997, 1998, and 1999; therefore, overwinter survival was estimated according to the levels of X. fastidiosa being recovered from inoculated branches in 1998 or 1999. We also used only those branches that had tested positive in 1997 and that we had evaluated at least once again in one of the successive years. At Davis, 5 of 6 infections with X. fastidiosa almond strains survived and 24 of 57 infections by grape cluster strains tested positive after one or two winters (P > 0.05 [comparing survival of pooled grape and almond strains and using chi-square testing with a Yates correction]). We did not observe differences of symptom severity among strains or localities, although symptoms (consisting of the presence of 2 to 7 leaves with marginal leaf scorch) were mild during the first year (1997). Such first-year symptoms were so light that it is likely that few of these would have been noticed in orchard surveys. Symptomatic leaves tended to more readily fall from the stem, increasing the difficulty of seeing disease symptoms.
|
|
|---|
We examined the overwintering survival rates of X. fastidiosa in almonds to determine whether grape strains survived winters more poorly than almond strains. In a previous study that examined only grape strains inoculated into almond trees in two southern Central Valley locations, overwintering survival was very poor (4.2%) (6). The hypothesis that strains from almonds survived winters better than grape strains might explain the lack of ALS in orchards adjacent to PD hotspots in the southern Central Valley. Our data from field-grown trees did not support this hypothesis, but our data on overwintering potted plants did. Both strain groups survived winter dormancy for 2 years at about the same rates in central California. Differences in winter severity may have contributed to the higher rates of survival of X. fastidiosa recorded in our study compared to those determined in the earlier study (6). In addition, our use of mechanical inoculation may have resulted in faster or wider spread of X. fastidiosa within plants than that caused by vector inoculation. It was surprising to find that some infections of almonds that survived the first winter (1997 to 1998) failed to survive the second winter. Our laboratory experiments showed results for overwintering survival of strains in grapes and almonds which were especially surprising because of the milder temperatures at Berkeley compared to those in the Central Valley. Future research with more field sites distributed throughout California is needed to further clarify this epidemiological aspect of both strains.
The limitations in cultivating X. fastidiosa in vitro were only overcome in 1978 (5), and in the following years other media were developed to grow this bacterium. In general, little difference has been found among different media in the ability to support growth of X. fastidiosa from grapes or almonds (3, 4) but the number of isolates and media combinations tested for media specificity has been limited. We believe that Davis et al. (6) did not test isolates from almonds avirulent to grapes because, at that time, only PD3-like media were available (Davis et al. used PD2 medium for their work). PW (the base medium for PWG solid medium) was developed a few years later for the growth of more fastidious strains of X. fastidiosa, such as those found in peaches (4). Therefore, although the conclusions of the studies are different, our findings corroborate their results.
Hopkins (16) first quantified X. fastidiosa populations within plants, associating bacterial numbers with isolate virulence. The technique was later used to determine differences in resistance and susceptibility to PD among grape varieties (8). X. fastidiosa populations within plants were also correlated with vector acquisition efficiency (15) and used to identify natural reservoirs of this bacterium (14, 27). Recently, independent results of quantification of X. fastidiosa populations from citrus by culture (2) and quantitative PCR (23) techniques suggest that the number of live cells recovered from culture assays is very similar to the total number of cells estimated by PCR. We found that for grape strains pathogenic to both grapes and almonds, the number of live X. fastidiosa cells within host plants is dependent on the host and not the strain type.
The reason for the lack of virulence of almond strains in grapes has not been addressed, but we suggest that limited colonization (103 to 105 CFU/g) and movement precludes the virulence of almond strains in grapes. Because X. fastidiosa multiplies and causes symptoms significantly faster in the greenhouse than in the field (A. H. Purcell, personal observations), we expect that the almond strains from group II, which caused mild PD symptoms in our experiment, would be virtually avirulent to grapes under field conditions. The recently published complete genome sequence for a grape strain (Temecula) (33) included in our study and a draft sequence for the Dixon isolate (http://www.jgi.doe.gov/JGI_microbial/html/xylella_almond/xyle_almnd_homepage.html) in group I of the almond strains should facilitate searches for the genetic basis of resemblances between these strains that cause differentiated pathology in grapes.
Our results suggest that no PD is found near ALS orchards because ALS strains likely do not cause PD under field conditions. In addition, all ALS isolates that we and Hendson et al. (10) typed were obtained from northern counties in the Central Valley (areas with little evidence of PD). All isolates from almonds identified as grape strains came from southern areas of central California. Because grape strains cause ALS, we hypothesize that the rare cases of infected almond plants in these areas are due to accidental infections by infective vectors. This still leaves unanswered the issue of why ALS has been rare or not detected at all in almond orchards adjacent to vineyards with a high incidence of PD. The principal insect vectors for PD can be found only rarely on grapes (11, 26), as appears to be the case for these insects on almonds (24; A. H. Purcell, unpublished data). It is possible that the seasonality determining when the insects are most likely to feed on almonds differs substantially from that determining when the insects are most likely to feed on grapes, which could lead to lower rates of disease development in almonds. Since RsaI differentially digests the PCR product of the X. fastidiosa primer set RST31-33, which is widely used for detection of X. fastidiosa in plants, it may provide an easy way to type isolates collected from vineyards or almond orchards for their pathogenicity to almonds or grapes. The future impact of the invasive vector Homalodisca coagulata on the spread of these different X. fastidiosa types is unknown, but it may modify the epidemiology of both diseases, because it feeds on grapes and almonds (28).
Despite the biological differences we found between almond strains and those from grapes, we suggest that the pathovar classification approach is not useful for studies involving X. fastidiosa. For example, citrus and coffee strains from Brazil are genetically distant from grape strains and are more distant and nonsympatric than almond strains but cause PD in grapes (20, 21). The relevance of the number of isolates tested, the location of collection, the isolation media, the host specificity, and other unknown factors might be larger for X. fastidiosa diversity studies than previously expected on the basis of comparisons with other plant pathogenic bacteria.
This work was supported by the Almond Board of California and USDA/CSREES Viticultural Consortium grant 2001-34360-10328. R.P.P.A. had a scholarship from CNPq-Brazil.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»