Abstract

Intraspecific genetic diversity of Drechslera tritici-repentis as detected by random amplified polymorphic DNA analysis

Ana Maria Pujol Vieira dos Santos¹, Aida T. Santos Matsumura¹ and Sueli Teresinha Van Der Sand²

¹Departamento de Fitossanidade, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.

²Departamento de Microbiologia do Instituto de Ciências Básicas da Saúde, Universidade Federal do Rio Grande do Sul, Porto Alegre, RS, Brazil.

Send correspondence to Sueli Teresinha Van Der Sand. E-mail svands@vortex.ufrgs.br.

ABSTRACT

The phytopathogenic fungus Drechslera tritici-repentis causes tan spot, an important disease of wheat in the southern Brazilian state of Rio Grande do Sul. Twelve D. tritici-repentis isolates were obtained from wheat seeds from different locations in the state. Their colony morphology on potato dextrose agar and polymorphisms in genomic DNA by the random amplified polymorphic DNA (RAPD) method were investigated. For the RAPD method, 23 primers were tested of which nine were selected for use in the study of D. tritici-repentis polymorphisms. The degree of similarity between isolates was calculated using a simple matching coefficient and dendrograms constructed by the unweighted pair-group method with arithmetical averages (UPGMA). The morphological and RAPD analyses showed intraspecific polymorphisms within the isolates, but it was not possible to establish a relationship between these polymorphisms and the geographical regions from where the host seeds were collected.

Key words:Drechslera tritici-repentis, RAPD, DNA polymorphism, genetic diversity.

Received: July 19, 1999; accepted: January 24, 2000.

INTRODUCTION

Tan spot of wheat leaves is caused by Drechslera tritici-repentis (Died.) Shoem. (anamorphic phase, the teleomorphic phase being Pyrenophora tritici-repentis (Died.) Drechs). The nomenclature of this fungus has been discussed by several authors, with Pyrenophora trichostoma, Helminthosporium tritici-vulgaris and Helminthosporium tritici-repentis having been considered as being the same fungus (Mehta, 1975).

Tan spot is prevalent worldwide (Hosford, 1972; Mehta, 1979; Wiese, 1991), and in Brazil it has been found in the states of Rio Grande do Sul, Paraná (Luz, 1982), Santa Catarina and Mato Grosso do Sul (Linhares and Luz, 1996). In Australia in 1982, losses due to a severe epidemic of this pathogen reached about 49% wheat-grain production (Rees et al., 1982).

It is recognized that D. tritici-repentis presents enormous variability in its morphology, genome and pathogenicity. Since traditional morphological methods of identification can be slow and tedious, other methods for the detection and differentiation of strains of D. tritici-repentis have been investigated by various workers. Methods based on the analysis of genomic DNA have the potential to allow the direct examination of fungal samples, eliminating the need for culturing. Random amplified polymorphic DNA (RAPD) has been extensively used for the characterization of biological material and in this work this technique was used along with colony morphology to identify polymorphisms and investigate the genetic similarity between different isolates of D. tritici-repentis.

MATERIAL AND METHODS

Drechslera tritici-repentis isolates

Twelve D. tritici-repentis isolates recovered from wheat seeds (supplied by CNPT-EMBRAPA, Passo Fundo-RS, Brazil) of different cultivars growing in different regions in the southern Brazilian state of Rio Grande do Sul were used (Table I). The wheat seeds were immersed in sodium hypochlorite 2% for 2 min, rinsed three times in sterile distilled water, and subsequently transferred to Petri plates containing potato dextrose agar (PDA). The samples were then incubated at 24 ± 2 °C for approximately 9 days at a 12 h photoperiod. After confirmation of the vegetative structure of the fungi, conidia were transferred to PDA slants and incubated as above described. All strains were stored as conidia and hyphae at 4 °C.

Morphological analysis

The morphologic analysis for each isolate followed the method described by Frazzon, et al. (2002, this issue).

Total DNA extraction

Genomic DNA was extracted from each isolate by a modification of the method of Ashktorab and Cohen (1992). The isolates were incubated in potato dextrose broth [20% potato, 2% dextrose (w/v)] for 7 days at 24 ± 2 °C with a 12 h photoperiod. The DNA extraction procedures were those described by Frazzon, et al. (2002, this issue).

Analysis of amplified DNA

The genomic DNA of the D. tritici-repentis isolates was amplified by the RAPD technique (Williams et al., 1990). The primers tested were the series A (A1-10) and B (B1-10) from BIODYNAMICS SRL (Buenos Aires, Argentina) and OPB3, OPB17 and OPC13 from OPERON Technologies-CA ( Alameda, CA. US). Each 25 mL of reaction mixture contained 40 ng of DNA, 2.5 mM of each dNTP (Pharmacia, Sweden), 45 ng of primer, 3 mM of MgCl₂, 0.01 mg of BSA, 2.5 mL reaction buffer (10x) and 1.5U Taq polymerase (CENBIOT, Porto Alegre, Brasil), the surface of the reaction mixture being overlaid with two drops of mineral oil. RAPD was performed in a Minicycler MJ Research thermocycler for 46 cycles, one cycle of 1 min at 94 °C, 5 min at 30 °C and 2 min at 72 °C;. 44 cycles of 1 min at 94 °C, 1 min at 30 °C, 2 min at 72 °C, and a final extension of 10 min at 72 °C. Amplification products were resolved on 1.4% agarose gel and visualized after staining with ethidium bromide.

Data analysis

The RAPD data and morphologic characteristics of the colonies were analyzed using the Statistical Package for the Social Sciences (SPSS) software, 2^nd edition, thus calculating similarity coefficients and constructing the dendrogram for genetic distances. The similarity was evaluated through simple association and the genetic distance as the Euclidean Distance. The binary matrix was built pairwise, and the presence or absence of a determined RAPD band scored 1 and 0, respectively. The hierarchical groupings were based on the Unweighted Pair Group using Arithmetical Averages - UPGMA (Sneath and Sokal, 1973).

RESULTS

Inspection of the colony morphology of the isolates showed that there was variation in the color, borders, texture and sectors of the colonies (Figure 1).

Table II shows the nine primers selected for the analyses on the basis of the patterns obtained, after being tested twice in different experiments with each D. tritici-repentis isolate. For cluster analysis 45 fragments were used, of which 73% were polymorphic. Each primer generated a different amplification pattern with a variable number of fragments (Table II).

Primer B06 amplified four fragments, one of which was monomorphic (Figure 2), while primer B07 generated more fragments than the others, seven of which were polymorphic (Figure 3). Primer OPC13 generated six fragments, and a 0.75 kb fragment amplified only from isolates BR23-S.Rosa and E15-L.Ver (Figure 4).

Figure 5 shows the dendrogram of morphological and RAPD data combined and contains five groups. One group is formed by isolates PF90120-PF, E16-Vacaria, E16-NMTI, E16-NMTII and E24-Butiá, with isolates PF90120-PF and E16-Vacaria showing the highest similarity coefficient (0.8704). Another group is made up of isolates BR23-S.Rosa, E15-L.Ver, E16-R.Alta and C19-C.Alta. The remaining three groups contain only one isolate each, E16-Coxilha, E16-Selbach and C24-Selbach respectively. The lowest similarity coefficient (0.5556) was observed between isolates E24-Butiá and C24–Selbach (Table III).

The dendrogram with the morphological characters (Figure 6) consists of three groups. One group contains isolates E16-NMTI, E16-Vacaria, C19-C.Alta, E16-NMTII and E16-Coxilha. Another group is formed by isolates E16-R.Alta, PF90120-PF, E24-Butiá, E15-L.Verm, BR23-S.Rosa and E16-Selbach, all with different degrees of similarity. Isolate C24-Selbach stands alone as a separate group and has the lowest similarity coefficient (0.2222) with isolates BR23-S.Rosa, E24-Butiá, E15-l.Ver and E16-Selbach (Table IV).

Figure 7 presents the dendrogram of the RAPD data and is made up of three groups. One group is formed by isolates E16-Vacaria, PF90120-PF, E24-Butiá, E16-NMTII, E16-Coxilha, E16-NMTI, BR23-S.Rosa, E15-L.Verm and E16-R.Alta, with isolates E16-Vacaria and PF90120-PF having a similarity coefficient of 0.9318 (Table V), the highest among the isolates. Another group contains isolates C19-C.Alta and C24-Selbach. Again, isolate E16-Selbach stands alone in its own group and showed the lowest genetic similarity with the other isolates. The lowest similarity coefficient (0.5682) was that between E16-Selbach and E16-R.Alta (Table V).

DISCUSSION

Using the RAPD technique and morphological polymorphism we detected genetic diversity among D. tritici-repentis isolates recovered from seeds collected in different locations in the southern Brazilian state of Rio Grande do Sul.

The isolates of D. tritici-repentis used in this study were from wheat plantations located in three different regions in the State of Rio Grande do Sul, Brazil (Table I), but it was not possible to relate the polymorphisms found among the isolates with the region from where the host seeds were collected. In Figure 5 (based on morphology and RAPD data combined) isolates from the same region appeared in different groups and most of the isolates which shared the highest similarity coefficients were from different geographic regions. Analysis of morphological data (Figure 6) and RAPD data (Figure 7) separately also showed no obvious relationship with the location from which the seeds were collected so it appears that in this case there was no advantage in analyzing the data separately.

The wheat fields from which the seeds were collected are located in the middle or northern regions of Rio Grande do Sul where the soils are very variable in their chemical characteristics despite the fact that they were formed from the same geological material (Brauner, 1982). It would be interesting to know more about these variations in soil chemistry, and especially whether or not they happen in the same regions. Another aspect that might have influenced the variability of the isolates is the length of time during which the host-pathogen interaction has taken place. The D. tritici-repentis isolates show a great deal of polymorphism, with the similarity coefficient in the genetic analysis ranging from 59% to 93%, suggesting that the population is not in equilibrium.

There are many factors affecting polymorphism analysis e.g. the number of samples selected for analysis, the organism studied, genetic flow between populations, environmental adaptation and adaptation to a new host, selective pressure and migration. Peever and Milgroom (1994) have stated that agricultural pathogens are subject to extensive extinction and re-colonization and are rarely in equilibrium. In studies with D. teres from different parts of the world these authors found a high degree of genetic differentiation between populations when compared with most other fungal populations studied, about 46% of the total genetic variability observed occurring in the D. teres populations.

Although Guthrie et al. (1992) and Assigbetse et al. (1994) were able to detect a relationship between intraspecific polymorphisms and geographic location, most work done in this field has shown no direct relationship and there is no general agreement among researchers as to whether such a relationship exists.

Fabre et al. (1995), in a study of Colletotrichum lindemuthianum from different countries, investigated grouping by DNA polymorphism in relation to the geographical origins of the isolates and found no correlation, because the isolates formed two groups with isolates from Latin America occurring in both groups. The relationship of Trichoderma harzianum strains from various geographical regions and presenting similar RAPD patterns is not yet understood (Zimand et al., 1994). Moller et al. (1995) detected intraspecific diversity not only between isolates of Chaunopycnis alba from different geographic regions or hosts, but also between isolates from a single location. In the present work with D. tritici-repentis, the isolates recovered from seeds collected in the towns of Selbach and Não-Me-Toque (the same geographic origin) showed both morphological and RAPD diversity (Figures 1 and 4).

Using RAPD markers it was possible for Bayman and Cotty (1993) to separate isolates based on the relationship between their genetic and morphological variability. However, this approach was not possible in the present work probably because the number of samples from each region was too small and only nine primers were used for the final analysis. The study of a larger number of primers and isolates (and re-isolating in different years) may show clearer results.

It will be very important to find a molecular marker for D. tritici-repentis, since Cochliobolus sativus and D. tritici-repentis produce very similar symptoms on seedlings and wheat leaves, making it very difficult to distinguish the disease based only on symptoms (Luz, 1982). It is normally necessary to isolate the fungus by plating for identification and confirmation, thus delaying diagnosis (Mehta, 1979). The RAPD technique is quick, effective and produces reliable markers which have been used for the identification of various fungal pathogens. Using this technique to identify pathogens by DNA fingerprinting is proving to be very useful and should be equally beneficial for classifying D. tritici-repentis isolates.

ACKNOWLEDGMENTS

This work was supported by the Brazilian agencies Fundação de Apoio à Pesquisa no Estado do Rio Grande do Sul (FAPERGS) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).

Publication Dates

History