INHERITANCE OF REACTION TO Leveillula taurica ( LEV . ) ARN . IN Capsicum annuum L .

The use of fungicides to control powdery mildew in sweet pepper has been ineffective and genetic resistance is the best alternative. Resistance sources identified in Capsicum annuum L. are rare and unsatisfactory. The purpose of this work was to study the inheritance of C. annuum reaction to powdery mildew. Three homozygous powdery mildew resistant parents, HV-12, Chilli and #124 and three susceptible lines, 609, 442 and 428 were used to obtain seven F 1 ’s and respectively their generations F 2 : HV-12 x 609, 442 × HV-12, 428 × HV-12, Chilli × 609, #124 × 609, Chilli × HV-12 and #124 × HV-12. The powdery epidemic was natural using inoculum from highly sporulating susceptible pepper host. Powdery mildew host reaction evaluations were carried out during the fruit production using a rating system based on disease severity scales varying from 1 (resistant) to 5 (highly susceptible). The experimental design was completely randomized. The following genetic parameters were estimated: locus numbers, gene action, heritability coefficient, expected selection gain and observed progress in F 3 generation, and possibly allelic relationship among resistance genes of different resistance sources. The HV-12 × 609 cross was the only one that showed absence of dominance. Other genetically analyzed crossings showed dominant and epistatic effects. Resistance was characterized as being due to at least four pairs of genes. The heritability and selection gains estimates were high. The resistance mechanisms of #124, Chilli and HV-12 showed differences in their expression.


INTRODUCTION
The powdery mildew, caused by Leveillula taurica (Lev.)Arn., is the most important and limiting disease for the sweet peppers crop in greenhouse.The anamorphic of asexual fungus stage found in sweet pepper crop is known as Oidiopsis taurica (Braun, 1987).
In Brazil, the first report of powdery mildew for sweet pepper was in the Federal District region in 1994 (Boiteux et al., 1994).Sweet pepper and tomato sucessive planting resulted the establishment of the powdery mildew.The symptoms begin with the mature tissue of chlorotic lesions on the down side leaves.Later, they advance Sci.Agric.(Piracicaba, Braz.), v.62, n.1, p.40-44, Jan./Feb.2005 and leaf abscission occurs, photosynthetic reduction activity and fruit sun burning (Palti, 1988;Daubeze et al., 1995).
The use of systemic fungicides for control has not been effective.Besides this, its indiscriminate use causes pathogen resistant strains development (Palti, 1988;Bergamin Filho et al., 1995).The best manner of control would be by means of resistant varieties.In Capsicum annuum L., partial resistance has not been satisfactory (Shifriss et al., 1992).
Heredity studies of C. annuum showed that resistance to powdery mildew is controlled by three pairs of genes (Shifriss et al., 1992) with additive as well as epistatic effects (Daubeze et al., 1995).For Murthy & Deshpande (1997), the resistance was dominant and poligenic and showed allelism differences among the resistant parents.
The inheritance of resistance to powdery mildew has not been studied and all the Brazilian sweet pepper hybrids worldwide are highly susceptible to powdery mildew.The purpose of this paper was to study the sweet reaction inherited of powdery mildew to support further breeding programs.
The experiment was carried out using a high tunnel of plastic in Piracicaba, SP, situated at 22º42'30" S 47º 38'00" W, at an altitude of 546 meters.The epidemic of powdery mildew occurred naturally.The initial inoculation was obtained and maintained by susceptible sweet pepper plants and with a high degree of sporulation.Plants that disseminated the pathogen were intercalated for a ratio one susceptible plant to each fifteen tested acesses.
Seedlings were transplanted into eight-liter pots with a substrate for vegetables from Multiplanta®.The experimental was entirely randomized design with five plants per pot and 25 treatments of which seven were hybrids (F 1 ), seven F 2 populations, three F 3 populations and six parents and two susceptible checks for a 8,766 plants.
Powdery mildew reaction evaluations were made at fruit development.A scale of scores was used in accordance with the leaf area affected, proposed by Ullasa et al. (1981), of which: 1 -resistant, no symptoms; 2moderately resistant, with 10% of the leaf area affected; 3 -moderately susceptible, with 11-20% of the leaf area affected; 4 -susceptible, with 21-50% of the leaf area affected and 5 -highly susceptible with 51% or more of the leaf area affected.Hybrids Margarita and Magali R were used as susceptible checks.The populations evaluation was made when referential check plants achieved maximum score 5.
The following estimates were obtained: (a) the number of segregating loci (based on the frequency in F 2 of the resistant extremes (score 1 or 1 and 2) showed by the resistant parent and by the formula given by Burton, 1951); (b) type of gene action, using the additive-dominant and epistatic model in accordance with Mather & Jinks, (1981), with significance verified by the t test according to Gomes (1990); (c) the broad sense heritability coefficient (h 2 ); (d) of the gain from selection (Gs) expected in F 3 , (e) observed progress in F 3 , estimated in accordance with Vencovsky & Barriga (1992), and (f) allelism relationships among the genes that control resistance.The broad sense heritability coefficient was employed for prediction of the selection expected gain in F 3 , being adequate in cases of absence of dominance.With any evidence of dominance, the value of Gs is overestimated, serving only as a reference.

Genetic base of heredity
The parent HV-12 showed high resistance (Table 1).According to Shifriss et al. (1992), the resistance expression of the dihaploid variety HV-12 was due to the restriction mechanism the infection, pathogen colonization and leaf retention.Pathogen penetration of Chilli and #124 parentals ocurred, however with colonization restriction as small necrotic lesions, indicating hypersensitivity reaction to the pathogen and consequent defoliation.
The parents 609, 442 and 428, behaved as susceptible to the pathogen with averages of 4.64, 4.56 and 4.60, respectively.The checks Magali R and Margarita confirmed their high susceptibility showing average scores of 5.00 (Tables 1 and 2).Behavior of the hybrids HV-12 × 609, 442 × HV-12 and 428 × HV-12 agreed with that observed by Shifriss et al. (1992), with an intermediate reaction, with averages of 2.87, 3.00 and 3.58, respectively (Table 2).This intermediate reaction might be able to help in the integrated control of the powdery mildew using fungicides or alternative control.
Plant reaction from the crosses Chilli × 609 and #124 × 609, whose parents with different resistance expression from the HV-12, were susceptible with averages of 4.96 and 4.17 in F 1 and 4.74 and 4.00 in F 2 , respectively.The parental choice greatly influences results and the breeding program success (Table 2).
The distribution of F 2 individuals for the different scores, in the crosses HV-12 × 609, 442 × HV-12 and 428 × HV-12 fell between the classes 2 and 3 with 66.8%, 75.6% and 50.7%, respectively (Table 1).Few plants had susceptibility degree of the parents 609, 442 and 428 as well few ones were resistant like HV-12.These results agree with those obtained by Daubeze et al. (1995), suggesting a complex genetic control.Crosses Chilli × 609 and #124 × 609, both had the largest proportion of plants showing class 5, with 83.0% and 46.7% respectively (Table 1).
The genetic analysis of the generations showed non-dominance for the cross HV-12 × 609 (Table 2).Crosses 442 × HV-12, 428 × HV-12, Chilli × 609 and #124 × 609 had significant heterosis value in relation to the parents average, indicating dominance expression at least in part of the loci (Table 2).All of the crosses with highly significant dominant gene action displayed epistatic effects, according with the results of Daubeze et al. (1995).
In relation to the averages of the F 1 generations and principally F 2 , there was indication of certain degree of dominance for susceptibility, indicating that the resistance alleles are recessive.These results confirm those observed by Shifriss et al. (1992) and Daubeze et al. (1995) (Table 2).To get powdery mildew resistant hybrids both parents have to be both resistant ones.The positive values of heterosis indicated greater character average in the F 1 generation than the parents.In relation to epistasis there was greater disease severity average of the F 2 generation than the expected one.
The number of segregating loci confirmed the results obtained by Shifriss et al. (1992), Daubeze et al. (1995) and Murthy & Deshpande (1997).For the crosses HV-12 × 609 and #124 × 609 at least two genes might be involved in the inheritance, while for the crosses 442 × HV-12 and 428 × HV-12 at least three genes were involved.In the cross Chilli x 609, the results calculated for the frequencies and by the formula of Burton (1951) were different.By the frequencies calculation, at least four genes were detected, while by the formula given by Burton (1951), nine genes might be controlling the resistance to the powdery mildew inheritance.Errors of estimates of the population variance can also be responsile for this discrepancy.
Considering estimated number of genes, it may be concluded that the inheritance of resistance to powdery mildew in C. annuum involves several loci (Table 2), which demand a greater selection generations cycles to get homozygosity.This kind of durable poligenic resistance is more difficult to be overcome by pathogenic strains (Van der Plank, 1968).

Selection for resistance
The estimated gains in selection for the F 3 generations were high, showing good prospects for obtaining lines of sweet pepper resistant to powdery mildew.In the F 3 generations of the cross Chilli × 609 and #124 × 609 resistance increase of 35.2% and 34.8%, are expected respectively (Table 2).
Resistance gains seen in the F 3 generations confirm the high heritability coeficient and the selection efficiency done in the F 2 generations.The expected average without selection for the F 3 generation of the cross HV-12 × 609 would be 2.89, but with selection it was 1.59, with a real gain on the average of 1.30 or 45%.In the F 3 population of 442 × HV-12 cross, the expected average without selection would be 2.85.However with selection it was 1.80, a gain in resistance of 1.05 or 36.8%.For the F 3 generation of the cross 428 × HV-12, the average expected without selection would be 3.14, with selection the average value obtained was 2.05, which corresponds 1.09 gain or 34.7% increase (Table 2).

Reaction of hybrids and their F 2 generations among resistant parents
Reaction expressions of the resistant parents HV-12, Chilli and #124 showed distinct genetic systems determining heritability with no allelic differences.The resistance reaction of the HV-12 proved to be reliable one and fulfilled the requirements for a vertical inheritance.On the other hand, resistance expression of hot pepper plants Chilli and #124, with strong restriction and sporulation reduction with small lesion number, indicates horizontal resistance machanism.
Results shown by the F 1 and F 2 generations indicated differences in genetic mechanisms among the resistant parents.In the cross #124 x HV-12, the parents displayed averages of 1.57 (#124) and 1.00 (HV-12), while the F 1 and F 2 generations showed themselves to be more susceptible than the parents, displaying averages of 2.71 and 2.88, respectively.For Chilli x HV-12, the plants of the F 1 and F 2 generations were even more susceptible.The average for the F 1 generation was 4.25 while the Chilli and HV-12 parents were 1.51 and 1.00, respectively.In F 2 , there was wide segregation for reactions to powdery mildew with 3.05 average (Table 2).
For both crosses, average contrasts involving the P 1 , P 2 , F 1 and F 2 generations were significant according to the t test, for heterosis and epistatic effects detections (Table 2).This fact confirms that the parents HV-12 and the hot peppers resistant source to powdery mildew are genetically different with respect to the alleles that control resistance to powdery mildew.

Table 1 -
The number of segregating plants for reaction to Leveillula taurica under greenhouse conditions, using severity disease score scale.Capsicum annuum.Piracicaba, 2002.

Table 2 -
Number of plants (n)and estimatives of averages disease severity scale, respective variances, numbers of segregating loci, heritability (h 2 ), heterosis (H and H%) and epistasis (C) with significance by the t test and expected selection gain in F 3 (Gs).C. annuum.Piracicaba, 2002.