CHARACTERIZATION OF RUST, EARLY AND LATE LEAF SPOT RESISTANCE IN WILD AND CULTIVATED PEANUT GERMPLASM

Groundnut (Arachis hypogaea) has an AB genome and is one of the most important oil crops in the world. The main constraints of crop management in Brazil are fungal diseases. Several species of the genus Arachis are resistant to pests and diseases. The objective of our experiments was to identify wild species belonging to the taxonomic section Arachis with either A or B (or “non-A”) genomes that are resistant to early leaf spot (Cercospora arachidicola), late leaf spot (Cercosporidium personatum) and rust (Puccinia arachidis). For the identification of genotypes resistant to fungal diseases, bioassays with detached leaves were done in laboratory conditions, with artificial inoculation, a controlled temperature of 25°C and a photoperiod of 10 h light/14 h dark, for 20–42 days, depending on the fungi species. Most of the accessions of wild species were more resistant than accessions of A. hypogaea for one, two or all three fungi species studied. Arachis monticola, considered to be a possible tetraploid ancestor or a derivative of A. hypogaea, was also more susceptible to Cercosporidium personatum and Puccinia arachidis, as compared to most of the wild species. Therefore, wild germplasm accessions of both genome types are available to be used for the introgression of resistance genes against three fungal diseases of peanut.


INTRODUCTION
Peanut (Arachis hypogaea L.) is the fourth most important oleaginous plant in the world. It is used mainly for oil and candy production, or for consumption in natura. World production is thought to be more than 30 million tons per year (CONAB, 2003). Brazil has about 90,000 ha planted with peanut, with a production of about 220,000 tons in 2006. The main constraints of the peanut production in Brazil, and indeed in the world, are fungal diseases, such as web blotch (Phoma arachidicola Marasas, Pauer & Boerema), early leaf spot (Cercospora arachidicola Hori), late leaf spot (Cercosporidium personatum (Berk & Curt.) Deighton), rust (Puccinia arachidis Speg.) and scab (Sphaceloma arachidis Bitancourt & Jenkins) (Godoy et al., 1999).
Species in the genus Arachis have potential for peanut improvement (Fávero et al., 2006). Several species have higher resistance levels to diseases when compared to A. hypogaea germplasm accessions (Pande & Rao, 2001;Stalker & Moss, 1987). The genus has nine taxonomic sections, and A. hypogaea is in the section Arachis, along with 30 wild species (Krapovickas & Gregory, 1994;Valls & Simpson, 2005).
The objective of the present study was to test diploid and tetraploid wild species within the section Arachis, with A and/or B or "non A" genomes, as well as several A. hypogaea varieties, for resistance against three fungal diseases -early and late leaf spot and rust, with the aim of future introgression of disease resistance genes into a breeding program of the cultivated peanut. Different accessions of a particular species may have different rates of resistance to fungal diseases. Although several previous publications have demonstrated the resistance of assorted wild Arachis germplasm against fungal diseases, old and newly collected accessions of many species were put together in the present work, and were tested with fungi isolates from São Paulo State, where 80% of the total Brazilian peanut acreage is concentrated. Isolates from Brazil may be different from those from other countries, so, this work was necessary as a basic step to implement a Peanut Pre-breeding project in Brazil.

MATERIAL AND METHODS
Accessions used in this study are shown in Table 2 and were obtained from the Arachis germplasm bank of Embrapa Recursos Genéticos e Biotecnologia, Brasília, Brazil. Only accessions of the section Arachis were selected because crossability with the cultivated peanut has not been accomplished so far with species of other taxonomic sections. The three diploid species of section Arachis with 2n = 18 chromosomes Arachis praecox, A. decora and A. palustris (Peñaloza & Valls, 1997;Lavia, 1998) were not included. Diploid species with 2n = 20 were assigned the A or B (non-A) genome according to the available documentation of the presence or absence of the small "A" chromosome pair in at least one accession of each in the literature (Fernández & Krapovickas, 1994;Lavia, 1998;Lavia, 1999;Peñaloza & Valls, 2005). The tetraploid wild species Arachis monticola Krapov. & Rigoni (2n = 40) is considered to share the same AB genome of A. hypogaea. Seeds from 102 accessions were treated with Carboxim and Thiram (0.5 g L -1 ) fungicide and germinated at 25ºC in germitest paper immersed in Ethrel solution (6 mL L -1 ) to break dormancy. Plants were maintained under screenhouse conditions, with four replications per each accession.
Cercospora arachidicola and Cercosporidium personatum spores were collected from infected plants at the Experimental Station of Ribeirão Preto (21º10' S, 47º48' W), from the Agronomic Institute of São Paulo State, Brazil (IAC), and Puccinia arachidis spores were collected from infected plants at the Experimental Station of Pindorama, (21º11' S, 48º54' W), from the Agronomic Institute of São Paulo State, Brazil (IAC).
For Cercosporidium personatum and Cercopora arachidicola, fungi cultures were grown in oat-agar medium (Moraes & Salgado, 1979). In addition, some fungi from the IAC bank of fungal isolates (1436-1 and 1595-0) were used (Table 1). New isolates were given the numbers 11576-0 and 11576-1, respectively. The P. arachidis spores were collected and stored in jelly capsules in refrigerator (about 5ºC), for about one week. It was periodically necessary to inoculate susceptible peanut leaves and re-isolate the spores.
VSPmSv 13832 sl (0) 0.00372a (1) 0.02715b (2) -3 Continue... Five isolates were prepared: two of Cercospora arachidicola, two of Cercosporidium personatum and one of P. arachidis. For Cercospora arachidicola and Cercosporidium personatum isolates were replicated by the use of oat-agar medium into Erlenmeyers that were shaked for about 20 minutes. The cultures were replicated at each 14 days, during approximately three months.

A. stenosperma
A detached leaf technique was used for the establishment of bioassays (Moraes & Salgado, 1982) mostly using leaves of the main stem of the plants. Cotton and germitest paper layers and a slide were used to keep the leaves well conditioned in Petri dishes. Destilated water was used to maintain leaves alive and turgid for several weeks. For the inoculation of Cercospora arachidicola and Cercosporidium personatum, the first expanded leaves were placed in Petri dishes, with the adaxial side upward. For Puccinia arachidis, the abaxial side was set upward. Inoculation was done with a mixture of the isolates of each leaf spot by spraying Tween 20 at 0.5% in a concentration of 50.000 spores mL -1 . Plates were maintained at 23-25ºC, photoperiod of 10 h light and 14 h darkness, in four random blocks. During the first 48 h, the Petri dishes were sealed in a plastic bag.
Puccinia arachidis bioassays were analysed at 20 and 27 days, Cercospora arachidicola at 27 days and Cercosporidium personatum at 42 days. For P. arachidis evaluation, the number of lesions per leaflet area (mm 2 ) was recorded and the lesions scored for presence or absence of spores. For Cercospora arachidicola and Cercosporidium personatum, the ratio of lesion area to leaflet area was evaluated (Foster et al., 1981).
The SAS program was used for the analysis of mathematic model according to the random blocks experiments. Different transformations were used for each group of data, according to ANOVA conditions. For Cercospora arachidicola, the transformation was arcsin (x + 0.5). For Cercosporidium personatum and P. arachidis a ) 1 log( + x transformation was used. The ANOVA presuppositions were verified: independence, homogeneity, and normal distribution of residues. Only data from accessions with lesions were used in the analyses. The Scott & Knott method (Scott & Knott, 1974) was used at 5% probability of Type I Error for the multiple comparisons among averages of different accessions. A sum of points, adapted to an index of selection (rank sum) of Mulamba & Mock (1978), was done to determine the most resistant and most susceptible genotypes. The t test was used to observe significant differences between the three groups, the species that have the A genome, species with non-A genome, and species that possess the AB genome. For Cercospora arachidicola and Cercosporidium personatum, this test was done using the average data of the species. For P. arachidis, the t test was used in the punctuation data to have the same screening for accessions that had lesions without pustule and lesions with pustule.

RESULTS AND DISCUSSION
Results for all diseases studied are summarized in Table 2.

Cercospora arachidicola
From 97 accessions investigated, 22 were shown to be highly resistant (Table 2 -third column)  (1) (1)). The "d" group presented accessions of A. kuhlmannii (2) and A. hypogaea (3). The "e" and "f" groups showed accessions with more lesions than the other groups, so these were the most susceptible. In these two final groups, only accessions of A. hypogaea (6) were observed. These results were already expected, confirming the greater resistance of many wild species when compared to A. hypogaea and their potential use in the improvement of the cultivated peanut, like A and "non-A" peanut genome substitutes. Resistance to C. arachidicola was also found in accessions of A. stenosperma, A. diogoi, A. correntina (Burkart) Krapov. & W.C. Greg. and A. duranensis by Foster et al. (1981).
The resistances are very heterogeneous among accessions of the same species, as in A. kuhlmannii and A. stenosperma. So, a precise analysis of each accession is necessary, as the data do not support the species as always resistant or susceptible. A high coefficient of variation of 49.68%, was obtained. However, it is commom to find values as great as this in disease evaluation data.

Cercosporidium personatum
From 91 accessions submitted to the C. personatum resistance test, 54 appeared highly resistant. No lesion was observed (Table 2 - (1)). As for resistance to Cercospora arachidicola, it was observed that wild germplasm accessions associated to both A. hypogaea genomes are available to be used in the introgression of Cercosporidium personatum resistance genes into the cultivated peanut. The results may suggest that the above accessions are immune to the pathogen. However, it seems too hasty to come to this conclusion based just on a laboratory test. It would be necessary to repeat the bioassay or to test the material in the field.
In the "a" group, which includes the most resistant of those accessions statistically analyzed, 22 accessions associated to the A genome species were observed, A. kuhlmannii (11 accessions), A. stenosperma (5), A. duranensis (1), A. helodes (1), A. kempffmercadoi (1), A. schininii (1), A. simpsonii (1), A. villosa (1) and three of "non-A" genome, A. magna (2) and A. gregoryi (1). An accession of A. hypogaea (US 224) showed higher resistance than the others, also being located in the "a" group. It is interesting to mention that this peanut accession, from Rondonia State, in Brazil, which also presented some resistance to C. arachidicola, is the source of resistance to Tomato Spotted Wilt Virus/TSWV incorporated in the Tamrun 96 cultivar (Smith et al., 1998). The "b" group only includes one accession of A. stenosperma and six of A. hypogaea. The "c" and "d" groups only include accessions of A. hypogaea (4), confirming that the cultivated species is significantly more susceptible than the wild species used in the experiment.
The coefficient of variation was 59.06%. As in the tests for Cercospora arachidicola resistance, it is normal to find high values like this in disease evaluation data.

Puccinia arachidis
In the rust resistance tests, seven out of 100 accessions appeared highly resistant, without evidence of any lesion (Table 2 -fifth column), 67 accessions showed different lesion levels, however, without pustules but with hypersensitivity reactions (Table 2 -fifth column), and 26 accessions were shown to be more susceptible, presenting pustules (Table 2 - sixth column).
In the same way as for resistance to Cercospora arachidicola and Cercosporidium personatum, wild germplasm accessions relating to both genomes of A. hypogaea were observed, and are available to be used in the introgression of P. arachidis resistance genes in the cultivated peanut.
All the accessions in the sixth column showed sporulation of P. arachidis. In the "a" group, there were 12 accessions of A. hypogaea, 6 of A. stenosperma, 2 of A. valida, 1 of A. monticola, 1 of A. magna, 1 of A. ipaënsis and 1 of A. williamsii (Table 2). In the "b" group, only 2 accessions were observed, 1 of A. stenosperma and 1 of A. valida. All accessions of A. hypogaea showed pustules, so the cultivated peanut is more susceptible than most of the wild species accessions in this study. Arachis monticola, an allotetraploid wild species, considered by some authors to be the ancestor or, alternatively, a derivative of A. hypogaea, also showed susceptibility to P. arachidis. Arachis ipaënsis, one of the original diploid ancestors of A. hypogaea (Fávero et al., 2006), also showed susceptibility to P. arachidis. The variation coefficient was 71.43%.
A sum of points was done to determine the most resistant and most susceptible genotypes and the added values for each accession are shown in Table 2 (Seventh column). Some data were missed, so with these data, the ranking may change in some cases. For Cercospora arachidicola, the letters from Scott & Knott test varied from "a" to "f", and some genotypes were not included in the variance analysis as they presented no lesion, therefore earning a zero score. To rank all the lesions and all the genotypes, the categorization by letters was converted to numbers: where it varied from zero to "f", the ranking varied from zero to six. For Cercosporidium personatum the values were from zero to "d", so they became zero to four. For P. arachidis, group one included the categories zero to "d", changing to 0 to 4; for group two, "a" and "b" groups were replaced by, respectively, scores 5 and 6. So, all test values obtained for each genotype were added, according to the conversions above. Most of the accessions of wild species were more resistant than the accessions of A. hypogaea. The allotetraploid A. monticola, also has been more susceptible to Cercosporidium personatum and P. arachidis when compared with most of the wild species. Arachis monticola was not tested for C. personatum because the leaves were not in appropriate condition for the analysis, nor were other accessions without group letters. Arachis ipaënsis, one of the diploid ancestors of A. hypogaea, was also shown to be more susceptible to Cercospora arachidicola and P. arachidis when compared to many of the wild species.
For Cercospora arachidicola, Cercosporidium personatum and P. arachidis, the results presented by Stalker & Moss (1987) in a table of species and respective resistance results for several diseases and pests show a marked similarity to those found in the present study.
The resistance to late leaf spot and rust were studied by Pande & Rao (2001) in 74 accessions of wild species of Arachis under greenhouse conditions. The accession KG 30006 of A. hoehnei did not show symptoms of either of the two diseases. Twenty-six accessions were classified as resistant to late leaf spot.
Sixty-eight accessions were considered rust resistant. Although most of the accessions appraised by Pande & Rao (2001) are not the same as those used in the present work, results can be extrapolated in some cases, as there are several common sites for the collections. Our accession of A. monticola (V 14165) , although collected more recently, but from the same site of Pande & Rao's accession, and the coincident accession of A. ipaënsis (K 30076) were also susceptible to rust, corroborating the results of Pande & Rao (2001).
There is resistance to Cercosporidium personatum, Cercospora arachidicola and Puccinia arachidis in many accessions of wild species and these resistances may be different among accessions of the same species.
Species with A genome are more resistant than A. hypogaea under the conditions of the present Cercospora arachidicola, Cercosporidium personatum and Puccinia arachidis bioassays (Table 3). The same was observed for species that have "non-A" genome. On the other hand, species with A genome were not significantly different from species with "non-A" genome, showing that resistance genes for the three fungal diseases are at both genomes, and it is possible to introgress them from the both genomes, doing the gene piramidization.