Comparison of the efficacy of QoI fungicides , alone or in mixture with triazoles , in Asian soybean rust control , 2016 / 17 growing season

Facultad de Agronomia, Universidad de Buenos Aires, Av. San Martín 4453, C1417DSE, Argentina; Agrocarregal, Av. Pres. Vargas. Condomínio Centro Empresarial Le Monde., 266, sala 401, 4 andar, Setor Jardim Marconal, Rio Verde, CEP 75.901-551, Goiás GO, Brasil; Agroservice, Rua Miguel Vargas, 291, CEP 99035-380, Passo Fundo, RS, Brasil. Autor para correspondência: Erlei Melo Reis (erleireis@upf.br). Data de chegada: 15/01/2018. Aceito para publicação em: 15/07/2018. 10.1590/0100-5405/190157


ABSTRACT
compared.Severity was evaluated in four phenological stages, and control was calculated based on final severity data, area under rust progress curve, defoliation, one-thousand-grain mass, grain yield and damage caused by the disease.The lowest mean of rust control efficacy resulted from applications of azoxystrobin (15.8 and 11.19%) and pyraclostrobin (15.4 and 16.76%) and the highest mean was found for picoxystrobin (54.1 and 54.35 %) and trifloxystrobin (69.7 and 64.46%), calculated based on final severity and AUDPC, respectively.Such efficacy is not enough to cover the fungicide application cost, although there was a positive effect of treatments on disease control and grain yield.
In the 2014/15 soybean crop, ASR control was done by spraying fungicide mixture containing SDHI (benzovindiflupyr + azoxystrobin), which had an efficiency greater than 90%.After three seasons of use, data from experimental fields showed reduced control by this mixture (9).Therefore, in 16 years, the three site-specific fungicides became inefficient in controlling ASR in Brazil.Moreover, DMIs and QoIs alone or in co-formulation with each other or with SDHIs showed a reduction in ASR control efficacy, indicating that double or triple mixtures do not work as an anti-resistance strategy.
In parallel to the reduction in P. pachyrhizi sensitivity to the three active ingredients (DMI, QoI and SDHI), the respective fungal mutations that confer cross and multiple resistance to the three mechanisms of action (MOA) were identified (4, 6, 14).
We hypothesized that the fungitoxicity of the four QoIs currently in use against P. pachyrhizi, with the same mechanism of action, has differently reduced, resulting in differences in efficacy among them.
The aim of this study was to quantify the current contribution of QoIs, alone or in mixture with triazoles, to Asian soybean rust control.
Experimental design was in completely randomized blocks with four replicates and experimental units of 2.25 x 6.00 m long; data were analyzed according to Scott-Knott test.
ASR control was calculated based on the data from the third severity assessment and from AUDPC.As to severity, the efficacy of treatments ranged from 11.0% (pyraclostrobin 360 mL/ha) to 73% (trifloxystrobin 240 mL/ha).When calculated by using AUDPC, the control efficacy varied from 14.8% (pyraclostrobin 360 mL/ha) to 68.1% (trifloxystrobin 240 mL/ha).Considering QoI mixed with DMIs, when calculated based on severity, the efficacy ranged from 44.3 for azoxystrobin + cyproconazole, 71.5 for pyraclostrobin + fluxapyroxad to 79.3 for trifloxystrobin + prothioconazole, respectively.Similarly to the calculation based on AUDPC data, the control efficacy was 46.0, 68.3 and 76.0%, respectively (Table 1).
If not transformed into efficacy, the values for AUDPC make comparisons difficult among treatments.In the unsprayed plots, the highest value was 1216.37 and the lowest one was 291.1 units, only allowing the identification of the best treatment.However, when expressed as control, such difficulty was eliminated.For example, the maximum control of 76.0%, close to 80%, is considered the minimum to match the fungicide spraying cost in order to control ASR (11).
Defoliation of soybean plants ranged from 77.25% to 100% and was a function of ASR intensity (Table 2).
Regarding one-thousand-grain mass, the lowest value was 129.7g, determined for grains of the unsprayed plots, while the highest value was 162.9 g for grains from plots that were sprayed with pyraclostrobin + fluxapyroxad (Table 2).Data regarding one-thousand-grain mass, as well as AUDPC, have little importance in the selection of treatments with greater efficacy for disease control.
Grain yield in the unsprayed plots was 2508 kg/ha and, in the best treatment (pyraclostrobin + fluxapyroxad), it was 3579 kg/ha, corresponding to a maximum damage of 1071.6 kg/ha or 29.9%.The average yield of sprayed plots was 2528.9 kg/ha, 2613.8 kg/ha for pyraclostrobin, 2991.6 kg/ha for picoxystrobin, and 3110.4 kg/ha for trifloxystrobin.
Defoliation of soybean plants was a positive and linear function of ASR severity expressed as the function y = 0.4729x + 68.815, with coefficient of correlation, R 2 = 0.8316, where 'y' is plant defoliation and 'x' is rust severity (Fig. 1).The equation indicates that every 1% rust severity caused 0.4729% defoliation.The report by Reis et al. (10) confirms that defoliation is a function of ASR severity and that its evaluation is not necessary since severity data alone are sufficient to explain the effect of disease intensity on soybean yield, a reflex of the efficacy of treatments.
The relationship between grain yield and ASR severity was Our results confirmed (6, 8, 9, 14) the low field performance of fungicides containing strobilurins in ASR control.After detection of the mutation at position F129L, which confers a reduction in P. pachyrhizi sensitivity to QoIs, studies have been carried out to quantify whether this mutation differently affected the fungitoxicity of the four strobilurins mostly used for ASR control.Although they have the same mechanism of action, inhibition of electron transfer in the respiratory chain of complex III in the mitochondria, their efficacy was different: one group showed 15.8% (azoxystrobin) and 15.4% (pyraclostrobin) control, and the other group had superior mean control, picoxystrobin 54.1% and trifloxystrobin (69.7%) (Table 1).
Azoxystrobin + cyproconazole co-formulation, used in the 2009/10 season, was sprayed on 10 million ha in the 2006/07 season, and pyraclostrobin + epoxiconazole mixture in 15 million ha.Therefore, exposure to the fungus for a long time may have determined the most pronounced reduction in fungitoxicity to P. pachyrhizi; in addition, differences in the intrinsic characteristics of the molecules of QoIs can determine differences in fungitoxicity among them.
Finally, it can be inferred that the greater efficacy of picoxystrobin and trifloxystrobin explains the better performance of commercial mixtures of picoxystrobin + cyproconazole and trifloxystrobin + prothioconazole in ASR control.In contrast, the reduced fungitoxicity of azoxystrobin and pyraclostrobin explains the inferior performance of the mixtures in which they are incorporated.expressed as the function y = -16.177x+ 3613.9,R0.574 = ², where 'y' is grain yield, 'x' is ASR severity, and R is the coefficient of determination.According to the equation, every 1.0% severity reduces 16.177 kg/ha grains in 3613.9 kg/ha potential yield (Fig. 2).
Considering the AUDPC ratio of rust on grain yield, the obtained damage function was y = -1.061x+ 3665.5, R2 0.875, where each unit of AUDPC reduced 1.061 kg/ha in 3666.5 kg/ha yield (Fig. 3).The AUDPC has been recommended especially for the evaluation of the reaction of cultivars to diseases, having no practical importance to estimate the damage caused by a disease in a crop (5).In this case, the determination and practical use of severity has an advantage over AUDPC, which is more time-consuming.

Figure 2 .
Figure 2. Negative relationship between grain yield (y) and Asian soybean rust severity (x).

Figure 3 .
Figure 3. Negative relationship between soybean grain yield (y) and area under Asian soybean rust severity progress curve (AUDPC) (x).AUDPC

Table 1 .
Effect of treatments on rust severity, area under the disease progress curve (AUDPC), and disease control.
Means followed by the same letter in each column did not differ statistically from one another according to Scott-Knott test at 5%.

Table 2 .
Effect of treatments on the defoliation of soybean plants, one-thousand-grain weight (GW), and grain yield.
Means followed by the same letter in each column did not differ statistically from one another according to Scott-Knott test at 5%.