Genetic diversity of isolates of Leptosphaeria maculans from a canola (Brassica napus) paddock in Australia

Share Embed


Descrição do Produto

Australasian Plant Pathology, 2002, 31, 129–135

Genetic diversity of isolates of Leptosphaeria maculans from a canola (Brassica napus) paddock in Australia J. M. BarrinsAE, A. PurwantaraAD, P. K. AdesB, P. A. SalisburyBC and B. J. HowlettA A

School of Botany, The University of Melbourne, Vic. 3010, Australia. Department of Forestry, The University of Melbourne, Vic. 3010, Australia. C Victorian Institute for Dryland Agriculture, Horsham, Vic. 3400, Australia. D Biotechnology Research Unit for Estate Crops, PO Box 179, Bogor 16151, Indonesia. E Corresponding author; email: [email protected] B

AP0201 .JGeM.talnBaBteaicrdinevs,rAi.tyPuofrwatnar, PK. AdesP, . A. SalisburyandB. JHo. w let

Abstract. Amplified fragment length polymorphism analysis was used to differentiate isolates of Leptosphaeria maculans (Desm.) Ces. et de Not. which were collected, using a hierarchical sampling method, from a commercial canola (Brassica napus L.) paddock at Lake Bolac, Victoria, Australia. Many polymorphisms were found between individual isolates, but these did not allow differentiation into groups corresponding to the sampling hierarchy. Six isolates from a historical blackleg collection were included for comparison with the hierarchical sample. Although these isolates were similar, findings based on non-metric multi-dimensional scaling (NMDS) and unweighted pair group with arithmetic means (UPGMA) show that they do not group exactly within the hierarchically sampled isolates. Diversity indices and AMOVA give a conflicting representation of the diversity within the paddock. AMOVA suggests that there is only very weak hierarchical structure of diversity and there is as much diversity within any of the one metre sampling sites as there is between the six historical isolates which came from hundreds of kilometres apart. However, comparison of diversity estimates suggests that there are significant differences in between-isolate diversity between the sites, although these differences are small. It would appear that in order to gain an effective representation of the amount of diversity in any one location as part of a large-scale study of diversity across the country, a complex hierarchical sampling strategy such as the one employed for this study is not required. Additional keywords: ascomycete, blackleg disease, fungal pathogen, population genetics, pycnidiospore, amplified fragment length polymorphism analysis. Introduction Leptosphaeria maculans (Desm.) Ces. et de Not. (anamorph: Phoma lingam (Tode:Fr.) Des.) is the causative agent of blackleg disease of Brassica crops, including canola (Brassica napus L. and B. rapa L.) (for reviews see Howlett et al. 2001; West et al. 2001). Severe blackleg epidemics in the early 1970s limited the expansion of the canola industry in Australia, preventing its resurgence until the recent development of more resistant crops and better disease management strategies (Salisbury et al. 1999). Isolates of L. maculans have been classified into two pathotypes, A and B, based on their ability to form stem cankers on canola and also to produce the phytotoxin sirodesmin PL. Both A and B group isolates are abundant in Europe and Canada, and much of the previous research has focused on differentiating these isolates (for review see Williams and Fitt 1999). Australian blackleg isolates tend to be more virulent than those from other countries and are © Australasian Plant Pathology Society 2002

generally considered to be of the A pathotype (Purwantara et al. 2000), although two studies have described the presence of L. maculans isolates with reduced virulence on B. napus and which did not appear to be of the A pathotype (Plummer et al. 1994; Sosnowski et al. 2001). Previously, we have shown that on the basis of their AFLP banding patterns, A pathotype isolates collected from across Australia in the late 1980s cluster separately from A pathotype isolates from Europe and North America, demonstrating that these populations of the pathogen are genetically differentiated (Purwantara et al. 2000). Little is known about fine-scale geographic structure of L. maculans populations within Australia or how these populations should be efficiently sampled, either to obtain reliable diversity measures or to identify a representative sample of genotypes for pathogenicity testing. The present study uses molecular analysis of hierarchically sampled isolates within a single canola paddock to gain further insight into these issues.

10.1071/AP02001

0815-3191/02/020129

130

J. M. Barrins et al.

Methods

Statistical analysis

Fungal isolates

AFLP profiles were examined visually and poorly resolved or lowintensity bands were disregarded. All bands within a size range of approximately 0.15 kb and 1 kb were analysed. In our previous study (Purwantara et al. 2000) we scored only a predefined subset of AFLP bands. However, to show a more complete overview of the genetic variation within the hierarchically sampled population, all bands, monomorphic and polymorphic, were scored. Bands were assumed to be independent and those of identical sizes were assumed to have identical sequence. The AFLP profiles of isolates were scored for presence or absence of bands and entered into a binary matrix that was used to construct a similarity matrix between all pairs of isolates based on simple matching co-efficients. Construction of the similarity matrix, cluster analysis based on unweighted pair group with arithmetic means (UPGMA) and ordination using non-metric multidimensional scaling (NMDS) were carried out by using the software NTSYS-pc 2.0 (Rohlf 1989). Standard measures of genetic variability between isolates within sites [effective number of alleles; gene diversity, Shannon index (Nei 1973; Shannon and Weaver 1949] were calculated with the software POPGENE version 1.21 (Yeh et al. 1997). Differences in diversity indices between sites were tested by Friedman’s procedure using loci as replicates (Zar 1996). The degree of population subdivision between isolates was tested by analysis of molecular variance (AMOVA), in which four levels of hierarchy were defined: differentiation between sites, leaves within a site, lesions within a leaf and isolates within a lesion. The AMOVA was carried out with the software Arlequin version 2.0 (Schneider et al. 2000) in two overlapping stages in which the four hierarchical levels were defined using a combination of the results of both stages.

Leaves from plants grown in a single commercial paddock of canola (cv. Dunkeld) at Lake Bolac, Victoria, were collected in September 1997 using a hierarchical sampling method based on that described by McDonald et al. (1999). Canola plants were 12 weeks old (12–14 leaf stage) when sampled. Each leaf contained between five to eight lesions. Six sites (A–F), 1 m in diameter, in two transects 20 m long and 10 m apart in a rectangular grid, were sampled. Leaves towards the top of the plants (since the oldest leaves were senescing or absent, and thus difficult to culture) were randomly sampled from within the 1 m diameter used to define each site. At site A, one leaf was chosen from each of three plants. On each leaf, three pycnidia were isolated from each of five distinct lesions, giving a total of 45 isolates. For example, at site A, isolates were labeled as follows: Site (A), Leaf (1–3), Lesion (1–5), Pycnidium (1–3). Thus an isolate collected from site A, which was the third pycnidium from the second lesion on the first leaf would be denoted as A123. At sites B–F, one leaf was chosen from each of three plants. From each of these leaves, one pycnidium was sampled from each of three distinct lesions, giving 45 isolates. For sites B–F, isolates were labeled as follows: Site (B–F), Leaf (1–3), Lesion (1–3). It was not necessary to number each pycnidium as only one pycnidium was chosen from each lesion. Thus an isolate collected from site C, which was from the third lesion on the second leaf would be denoted as C23. Accordingly, a total of 90 isolates was cultured. To isolate pycnidia, infected leaves were allowed to dry naturally, then surface-sterilised for 10 s in 70% ethanol followed by 90 s in 0.5% sodium hypochlorite. Leaves were then placed in a Petri dish in a laminar flow cabinet for about 30 min to evaporate excess water, and then placed on moistened cellulose filter paper in a Petri dish. Pycnidia oozed pycnidiospores overnight, and ooze from a single pycnidium was transferred to 10% Campbells V8-juice agar plates containing penicillin G (100 U/mL), streptomycin sulphate (100 μg/mL) and rose bengal (40 μg/mL). After 1–2 days, a hyphal tip of the growing colony was cut and transferred to a fresh V8-juice agar plate which included the above antibiotics. Isolates were then grown at 26°C under white light for 1 week. In addition to the hierarchically sampled isolates, six L. maculans isolates from a previous Australia-wide collection were included for comparison. These were: C13 (from Millicent, South Australia), GA2 (Galong, NSW), M1 (Penshurst, Victoria), NC13 (Unknown), P10 (Penshurst, Victoria) and V4 (Numurkah, Victoria). C13 will be referred to as HC13 (Historical) to separate it from isolate C13 (site C, leaf 1, lesion 3) of this study. These isolates were collected as single ascospores in 1988 (except P10) and are described by Purwantara et al. (2000). DNA isolation and AFLP analysis Fungal mycelia were harvested following growth at 26°C in 10% V8 liquid media for 2 weeks. Mycelia were freeze-dried and DNA was extracted as described by Sexton and Howlett (2000). AFLP analysis was performed according to the method of Purwantara et al. (2000) using two primer combinations (below) which gave a high number of polymorphisms between isolates. (a) 5′-GACTGCGTACCAATTCGA-3′ and 5′-ATGAGTCCTGAGTAAAA-3′ (b) 5′-GACTGCGTACCAATTCGA-3′ and 5′-ATGAGTCCTGAGTAACA-3′. The two primer combinations yielded 50 and 42 markers that could be scored visually. The same DNA preparations were analysed by AFLP at least twice, and a subset of the collection (about 20 isolates) was reanalysed using independently extracted DNA samples. Duplicate experiments with different independent amplifications of individual DNA samples always yielded identical AFLP profiles.

Results AFLP analysis DNA from six isolates from the hierarchical sampling failed to yield bands with either AFLP primer combination, probably because the DNA was of poor quality. As every leaf was represented by at least two isolates, these six were omitted from further analysis. AFLP patterns of 90 isolates of L. maculans were analysed (including the six historical isolates). The two AFLP primer combinations yielded a total of 92 scorable markers, 82 of which were polymorphic (Fig. 1). Although the number of polymorphic markers was high, most bands were either extremely abundant or extremely rare, with only 11% of bands (10/92) present in the population at frequencies of between 16% and 85% (Fig. 2). Seventy individual genotypes (distinct AFLP patterns) were seen among the 84 isolates from the canola paddock. Banding patterns of each of the six historical isolates were distinct from each other and also from the 84 hierarchically sampled isolates. Cluster analysis of the AFLP profiles placed most of the isolates present within one major group, with a few outliers and eight isolates (C12, A152, A153, A343, A342, A352, A353, A351) forming distinct groups (Fig. 3) This pattern was also observed in the NMDS scatterplot (Fig. 4) which gives a three dimensional representation of the variation seen within the paddock. A stress 1 value of 0.109 suggests that the NMDS scatter plot is a reliable representation of the relationships between isolates. Of the six historical isolates,

Genetic diversity of Leptosphaeria maculans

131

five clustered within the major group in the phenogram; however, the ordination suggests that they are slightly separated from the major group as they are positioned slightly above the main plane of the other isolates. The sixth historical isolate, P10, clustered as an outlier, as did C23 (site C, leaf 2, lesion 3). Although neither method revealed clusters pertaining to particular sampling sites, at a finer scale, 28 isolates shared the same AFLP pattern with at least one other isolate and were indistinguishable within the resolution of this study. However, only eight pairs of ‘clonal’ isolates came from the same sampling site, suggesting that the pycnidial spread is not limited to within the original lesion. The estimates of all three diversity indices are similar for the historical isolates and for three of the six hierarchically sampled sites (A, C and F), despite the large geographic distribution of the historical isolates (Table 1). Friedman’s test indicates that there are significant differences in diversity indices between these sites and the other collection sites, sites B, D and E. AMOVA showed that 60% of the total variation was attributable to differences between individual isolates within lesions, and a further 32% variation was attributable to differences between lesions on a leaf. Only 7% of the variation was distributed among leaves in a site or between sites in the paddock, a result that is not consistent with that of Friedman’s test. The six historical isolates were not included in this analysis as they were not part of the hierarchical collection and would have altered the results of the AMOVA.

Fig. 1. AFLP analysis of DNA from a random selection of Leptosphaeria maculans isolates using primer combinations EcoRI (+GA)/MseI (+AA). Bands range in size from 0.15 kb to 1 kb. All clearly reproducible bands were scored for analysis. Some polymorphic bands are marked (arrows). 1

0.9

0.8

Frequency of marker

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0 1

4

7

10

13

16

19

22

25

28

31

34

37

40

43

46

49

52

55

58

61

64

67

70

73

76

79

82

AFLP marker

Fig. 2. Frequency of AFLP markers amongst 84 Leptosphaeria maculans isolates collected by hierarchical sampling. Eighty-two markers were polymorphic, 10 were monomorphic (not shown).

132

J. M. Barrins et al.

A211 A112 A212 A231 A213 A232 A331 D21 D22 A142 A143 F33 B23 D11 D12 F21 F22 D23 A111 A222 A321 A323 A332 E31 D13 A252 A253 A222 C22 C13 A241 A123 A341 B12 C11 E11 B32 A131 A133 F11 A132 B13 E33 E22 C33 C31 A242 B11 A121 D33 A251 A151 A312 A311 F13 B33 A122 B22 D32 C21 E13 GA2* HC13* D31 F12 F23 B31 A221 A333 B21 M1* NC13* A313 A243 E22 A223 A233 V4* E21 A152 A153 A342 A343 A351 A352 A353 C12 F32 P10* C23 0.60

0.70

0.80

0.90

1.00

Similarity coefficient Fig. 3. Dendrogram of 90 Leptosphaeria maculans isolates, 84 collected from six sites within a single paddock and six historical isolates, created using UPGMA analysis based on simple matching coefficients (NTSYS-pc 2.0 F. J. Rohlf, Exeter software, NY, USA). Isolates from the historical collection have an asterisk. See text for further details.

Genetic diversity of Leptosphaeria maculans

133

P10*

V4*

A153 A152 C12

A343 A342

A352 A353 A351

NC13* C23

D31 B21

E21

D33 A122 A312 A121B53B22 A042 A151 C91 A341 A132 A123 A511 A133 F13 F12 A131 F11 B03 E33 E52 B52 C51 B02 E51 A241 A522 D32 C22 C33 A251 A521 A523 A532 E31 A222 D03 D61 A032 A142 B23 D01 F21 A531 D02 A143 F73 B01 F22 A031 A053 D62 A013 E13 A052 GA2* D23

M1*

HC13*

A513 B3 A333

C13 A111

A223

A233 A211

A112

C21 F23 F32

E22 A221

A243

Fig. 4. Non-metric multidimensional (NMDS) scaling of 90 Leptosphaeria maculans isolates, 84 collected from six sites within a single paddock and six historical isolates. This scatterplot was constructed using NT-SYSpc Version 2.0 (F. J. Rohlf, Exeter Software, NY, USA). Stress 1 value is 0.109. See text for further details.

134

J. M. Barrins et al.

Table 1. Diversity indices for collections of isolates of Leptosphaeria maculans from different areas. Sites A–F represent six sites from within the same paddock, and historical refers to six historical isolates collected from south-eastern Australia. Values in brackets represent one standard error Location of isolates Site A Site B Site C Site D Site E Site F Historical

Sample size

Effective no. of alleles

Nei’s gene diversity

Shannon index

43 9 8 9 7 8 6

1.22 (0.05) 1.11 (0.08) 1.24 (0.10) 1.07 (0.06) 1.33 (0.09) 1.15 (0.08) 1.26 (0.15)

0.14 (0.03) 0.07 (0.05) 0.16 (0.06) 0.05 (0.04) 0.22 (0.05) 0.10 (0.05) 0.15 (0.08)

0.22 (0.04) 0.11 (0.07) 0.25 (0.09) 0.07 (0.06) 0.36 (0.07) 0.17 (0.08) 0.22 (0.12)

Discussion To further understand the spread and degree of variation in populations of L. maculans, it is important to develop tools for determining the degree of differentiation of populations of Australian isolates of the fungus. AFLP markers appear to be reliable markers for genetic mapping and studies of genetic variation. Such markers have been used to develop a genetic map of L. maculans and shown to segregate in a Mendelian fashion (Cozijnsen et al. 2000). Previously, we have used AFLP analysis to separate L. maculans populations from Europe, North America and Australia (Purwantara et al. 2000). Our current results are similar to those seen in the sexually reproducing Mycosphaerella graminicola, which causes leaf blotch on wheat, by Boeger et al. (1993) and McDonald and Martinez (1990). These authors used a hierarchical sampling method as a means of partitioning genetic variation. In both cases, fine scale distribution of genetic variability was seen within the populations, and distinct genotypes could be found both within lesions and between lesions on the same leaf. However, we have seen a greater amount of overall variation in the L. maculans population than that in the M. graminicola isolates, with 70 individual AFLP profiles from 84 isolates. Although McDonald and Martinez (1990) showed that there was genetic variation between isolates within the same lesion, they observed only 22 different haplotypes in 93 isolates. This probably reflects the fact that the two primer combinations we used for AFLP analysis generated more scorable markers than the Restriction Fragment Length Polymorphism (RFLP) probes used in the M. graminicola study. Additionally, hierarchical sampling has also been used in Rhynchosporium secalis, cause of barley scald (McDonald et al. 1999). This Australian study suggested that most variation was within the sampling areas of 1 m2. Although R. secalis is thought to reproduce exclusively asexually, only a low degree of clonality was observed, suggesting that the initial infection of the sampled barley crops was caused by a genetically diverse population.

We expected that at 12 weeks, the leaves sampled would have been infected mainly through successive pycnidial cycles, rather than from windborne ascospores. Whilst this may have been occurring to some degree, our results show that, based on AFLP profiles, fewer than 25% of the L. maculans isolates can be clonal. It would appear that in the population sampled from Lake Bolac, like that of R. secalis (McDonald et al. 1999), the original infection must have come from a diverse founding population. This is consistent with the wide dispersal distances of ascospores (West et al. 2001). With the Lake Bolac collection, 28 isolates were identical to at least one other isolate. In 16 of these instances, this could be accounted for by pycnidial spread due to rainsplash because the identical isolates were located within the same sampling site. However, the remaining 12 isolates shared the same banding pattern with isolates that were separated by at least 10 m. In these instances, the most likely method of dispersal would appear to be mechanical. It seems that the AMOVA is missing the structure of genetic diversity present in the paddock. The Friedman’s test indicates that there are differences in between-isolate diversity between the sites, but sites contribute only a small and non-significant component of variation in the AMOVA. The AMOVA pools variation at lower levels of the sampling hierarchy to compare the sites. This may be why it does not test differences in variability within each site, as does Friedman’s test. For binary data, all three indices are monotone transformations of each other so a test based on the rank transformation of any of these will produce the same results. For this reason, all three diversity indices effectively provide the same information. Sites A, C and F share the same level of diversity as the historically collected isolates, suggesting that there is an overall background of isolates that disperse evenly over large distances. However, the differentiation of the remaining sites indicates that infection by a small number of similar isolates in a localised area, in the cases of sites B and D, or a large number of distinct isolates in the case of site E, are viable means of blackleg spread. This may result in a distinction in between-isolate diversity. These differences in genetic diversity are not evident in the graphical representation of the data. Although the six historical isolates were positioned slightly above most of the Lake Bolac isolates in the NMDS, five of the six clustered within the major group in the phenogram. Minor inconsistencies occur between the results of these two analyses. Both techniques reduce a multidimensional structure to something that can be visualised in two or three dimensions. That both have produced very similar results suggests that the picture is probably a reliable representation of the real pattern. Isolate P10 from the historical collection was an outlier in both the phenogram and the NMDS, but this was not surprising as in the previous study by Purwantara et al. (2000), P10 was also an outlier. We believe that the use

Genetic diversity of Leptosphaeria maculans

135

of hyphal-tipped isolates from a single host cultivar has not biased our estimates of genetic diversity. Although cv. Dunkeld shows leaf and cotyledon blackleg lesions, it develops fewer stem cankers than other canola cultivars and consequently is one of the most blackleg-resistant cultivars sown in Australia (Salisbury and Wratten 1999). We have tested about 200 pycnidiospore-derived isolates and more than 80 ascospore-derived isolates of L. maculans from a range of different cultivars. All isolates are able to infect a range of canola cultivars (Howlett, unpublished). A study of L. maculans isolates from Canada used Random Amplified Polymorphic DNA markers (RAPDs) to distinguish between 93 isolates collected from two different fields, 20 km apart (Mahuku et al. 1997). The Canadian study, using isolates derived from both ascospores and pycnidiospores, showed that 45.5% of total variation could be attributed to differences between the fields, and suggested that a large proportion of this difference between fields could be attributed to ascospore sources that were unique for each field. This is unlikely to have occurred in the Australian paddock sampled, as all three diversity indices (Table 1) show that there is almost as much diversity between isolates collected within one metre of each other, from sites A, C and F, as there is between the six historical isolates that were collected hundreds of kilometres apart. As with the studies discussed above, the next step in the characterisation of the L. maculans populations in Australia is to test whether isolates collected from geographically more distant populations have the same level of variation as shown here. We are currently surveying isolates from a number of paddocks in Western Australia, New South Wales and Victoria. Given the significant level of variation between the isolates collected from within the same paddock and even within the same lesions, it would appear that we do not require an elaborate hierarchical sampling in order to gain an accurate representation of the diversity at particular localities across the country. References Boeger JM, Chen RS, McDonald BA (1993) Gene flow between geographic populations of Mycosphaerella graminicola (anamorph Septoria tritici) detected with Restriction Fragment Length Polymorphism markers. Phytopathology 83, 1148–1154. Cozijnsen AJ, Popa KM, Purwantara A, Rolls BD, Howlett BJ (2000) Genome analysis of a plant pathogenic ascomycete Leptosphaeria maculans; mapping mating type and host specificity loci. Molecular Plant Pathology 1, 293–302. Howlett BJ, Idnurm AI, Pedras MSC (2001) Review: Leptosphaeria maculans the causal agent of blackleg disease of brassicas. Fungal Genetics and Biology 33, 1–14. Koch E, Song K, Osborn TC, Williams PH (1991) Relationship between pathogenicity and phylogeny based on Restriction Fragment Length Polymorphism in Leptosphaeria maculans. Molecular Plant–Microbe Interactions 4, 341–349.

Mahuku GS, Goodwin PH, Hall R, Hsiang T (1997) Variability in the highly virulent type of Leptosphaeria maculans within and between oilseed rape fields. Canadian Journal of Botany–Revue Canadienne de Botanique 75, 1485–1492. McDonald BA, Martinez JP (1990) DNA Restriction Fragment Length Polymorphisms among Mycosphaerella graminicola (anamorph Septoria tritici) isolates collected from a single wheat field. Phytopathology 80, 1368–1373. McDonald BA, Zhan J, Burdon JJ (1999) Genetic structure of Rhynchosporium secalis in Australia. Phytopathology 89, 639–645. Nei M (1973) Analysis of gene diversity in subdivided populations. Proceedings of the National Academy of Sciences (USA) 70, 3321–3323. Plummer KM, Dunse K, Howlett BJ (1994) Non-aggressive strains of the blackleg fungus, Leptosphaeria maculans, are present in Australia and can be distinguished from aggressive strains by molecular analysis. Australian Journal of Botany 42, 1–8. Purwantara A, Barrins JM, Cozijnsen AJ, Ades PK, Howlett BJ (2000) Genetic diversity of isolates of the Leptosphaeria maculans species complex from Australia, Europe and North America using Amplified Fragment Length Polymorphism analysis. Mycological Research 104, 772–781. Rohlf FJ (1989) ‘NT-SYSpc: numerical taxonomy and multivariate analysis system.’ (Exeter Publishers: Setauket, NY) Salisbury PA, Wratten N (1999) Brassica napus breeding. In ‘Canola in Australia: the first thirty years’. (Eds PA Salisbury, TD Potter, G McDonald, AG Green) pp. 29–36. (Australian Oilseeds Federation) Schneider S, Roessli D, Excoffier L (2000) ‘Arlequin: a software for population genetics data analysis. Version 2.000.’ (Genetics and Biometry Laboratory, Department of Anthropology, University of Geneva: Geneva) Sexton AC, Howlett BJ (2000) Characterisation of a cyanide hydratase gene in the phytopathogenic fungus Leptosphaeria maculans. Molecular and General Genetics 263, 463–470. Shannon CE, Weaver W (1949) ‘The mathematical theory of communication.’ (University of Illinois Press: Urbana) Sosnowski MR, Scott ES, Ramsey MD (2001) Pathogenic variation of South Australian isolates of Leptosphaeria maculans and interactions with cultivars of canola (Brassica napus). Australasian Plant Pathology 30, 45–51. West JS, Kharbanda PD, Barbetti MJ, Fitt BDL (2001) Epidemiology and management of Leptosphaeria maculans (phoma stem canker) on oilseed rape in Australia, Canada and Europe. Plant Pathology 50, 10–27. Williams RH, Fitt BDL (1999) Differentiating A and B groups of Leptosphaeria maculans, causal agent of stem canker (blackleg) of oilseed rape. Plant Pathology 48, 161–175. Yeh FC, Yang R-C, Boyle TBJ, Mao JX (1997) ‘Shareware for population genetic analysis.’ (Molecular Biology and Biotechnology Centre, University of Alberta: Canada) Zar J (1996) ‘Biostatistical analysis.’ (Prentice-Hall: Upper Saddle River, NJ) West JS, Kharbanda PD, Barbetti MJ, Fitt BDL (2001) Epidemiology and management of Leptosphaeria maculans (phoma stem canker) on oilseed rape in Australia, Canada and Europe. Plant Pathology 50, 10–27.

Received 28 June 2001, accepted 19 September 2001

http://www.publish.csiro.au/journals/app

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.