# ABSTRACT:

The intensive use of glyphosate can select resistant populations. This study aimed at evaluating the occurrence of glyphosate resistance in populations of D. insularis in agricultural areas. Three experiments were conducted in randomized block experimental design with four replications. The Resistance Degree (RD) was determined by a dose-response curve in the evaluated populations: Experiment 2012, São Camilo (R01 and R02) and Palotina (R03, R04 and S05-susceptible); Experiment 2013 Palotina (R06 and R12), Katueté/Paraguay (R07), Toledo (R08 and R10), Cascavel (R09), São Miguel do Iguaçu (R11), Dom Eliseu/Pará (S13-susceptible); Experiment 2015 Nova Aurora (R14), Iracema do Oeste (R15 and R16), Jesuitas (R17 and R18), Tupãssi (R19), Jotaesse (R20), Espigão Azul (R21) and Palmitopolis (S22- susceptible). In 2012, it was possible to confirm a moderate resistance (RD = 1.1 to 5.0) only in R03; and populations R01, R02 and R04 were highly resistant (RD>5.0). In 2013, populations R09, R10, R11 and R12 showed resistance, while R06, R07 and R08 were highly resistant. In 2015, the R14, R16, R18, R19 and R20 populations showed moderateresistance, and R15 and R18 were highly resistant. Only R21 was susceptible. It was found that 83% of producers use the no-tillage system, 100% use the succession soybean/corn with 90% transgenic varieties, 45% of producers cannot explain the procedure regulation/sprayer calibration. Around 62% use practices to prevent the selection of resistant biotypes. Resistance was confirmed in 89.5% of the populations. The adopted agronomic practices may explain these results.

Keywords:
sourgrass; agricultural practices; weed; herbicide

# RESUMO:

Palavras-chave:
capim-amargoso; práticas agrícolas; planta daninha; herbicida

# INTRODUCTION

Weed resistance to herbicides is an adaptive response to agricultural practices (Monquero and Christoffoleti, 2001Monquero PA, Christoffoleti PJ. Manejo de populações de plantas daninhas resistentes aos herbicidas inibidores da acetolactato sintase. Planta Daninha. 2001;19:67-74.). Thus, resistant weed biotypes were selected in response to the environmental disturbance caused by the selection pressure of herbicides provided by their intensive use over the last decades (Christoffoleti and López-Ovejero, 2003; Machado et al., 2006Machado AFL, Ferreira LR, Ferreira FA, Fialho CMT, Tuffi Santos LD, Machado MS. Análise de crescimento de Digitaria insularis. Planta Daninha. 2006;24:641-7.).

The concept of resistance consists in the natural and inheritable ability of certain biotypes within a population to survive and reproduce after the exposure to doses of herbicides that would be lethal to normal (susceptible) individuals from the same species (Christoffoleti and López-Ovejero, 2003Christoffoleti PJ, López-Ovejero R. Principais aspectos da resistência de plantas daninhas ao herbicida glyphosate. Planta Daninha. 2003;21:507-15.). Thus, the countries with the highest numbers of glyphosate-resistant plants are the United States, Argentina and Brazil, since they are large consumers of glyphosate and intensively use high doses, which can more frequently select resistant populations (Heap, 2016Heap I. The international survey of herbicide resistant weeds [Internet]. [accessed on: 26, oct 2016]. Available at: Available at: http://www.weedscience.com .
http://www.weedscience.com...
).

Among the species that present resistant biotypes in Brazil, it is possible to highlight Digitaria insularis, whose resistance to glyphosate has already been confirmed in the states of São Paulo, Minas Gerais and Paraná (Lacerda and Victória Filho, 2004Lacerda ALS, Victória Filho R. Curvas dose-resposta em espécies de plantas daninhas com o uso do herbicida glyphosate. Bragantia. 2004;63:73-9.).

It is worth mentioning that agricultural producers from the western region of Paraná empirically reported the occurrence of resistance to glyphosate in populations of D. insularis; in most cases, suspicions of resistance in these weed populations have not been scientifically proven.

The observation of initial resistance is usually based on the unsatisfactory control of weeds after herbicide application, but this also occurs when there are usage defects in the application techniques. Thus, it is important to make careful observations on the field and in the laboratory, in the sense that any reduction in herbicide efficiency can be detected (Christoffoleti and López-Ovejero, 2003Christoffoleti PJ, López-Ovejero R. Principais aspectos da resistência de plantas daninhas ao herbicida glyphosate. Planta Daninha. 2003;21:507-15.).

Therefore, the hypotheses of this work are based on the fact that the management used to control weeds in agricultural areas is selecting glyphosate-resistant biotypes in populations of D. insularis. The objective of this study was to evaluate the occurrence and quantification of glyphosate resistance levels in populations of D. insularis, characterizing the agronomic practices adopted in the agricultural areas.

# MATERIAL AND METHODS

Three experiments were conducted in plastic pots maintained under field conditions. The first one was developed in the agricultural year of 2012, the second one in 2013 and the third one in 2015. A randomized block design with four replications was used. For each experiment, a susceptible population was used as a control treatment.

Table 1 describes the collection sites, as well as the year of the experiment, geographic coordinates and sowing system frequently used in the area. Specifically in the experimental year 2013, two populations did not originate in regions of the State of Paraná: Dom Eliseu, from the State of Pará (S13), and Katuete, from Paraguay (R07). All seeds were collected in areas of agricultural cultivation when seed dispersal by the wind was observed. At that moment, D. insularis panicles were wrapped in recyclable paper bags to collect as many mature seeds as possible.

Table 1
Characterization of collection points in each experimental year

In order to better understand the process of selecting populations of D. insularis that are resistant to glyphosate, at the time of seed collection from each population, a questionnaire was applied to producers, whose nine questions are described in Table 2.

Table 2
Questionnaire submitted to producers at the time of collection of Digitaria insularis populations

The resistance degree (RD) was determined using a dose-response curve with the following glyphosate doses for the experimental years 2012 and 2013: 0, 43, 90, 360, 720, 1,440, 2,880, 7,200 g ha-1. For the experimental year 2015, doses were 0, 28, 56, 112, 225, 450, 900, 1,800, 3,600 and 7,200 g ha-1. A commercial formulation containing 360 g L-1 of acid equivalent and without the addition of adjuvant was used. Applications were made when the plants had two to three tillers.

Populations were individually seeded in polystyrene trays for germination and subsequent transplanting of the seedlings in plastic pots filled with soil. The experimental units were represented by a plastic pot with a capacity of 3.0 dm³, filled with soil with the following chemical composition: 13.95 mg dm-³ of P; 0.36 cmol dm-3 of K +; 4.34 cmol dm-³ Ca; and 1.23 cmol dm-3 of Mg.

Transplanting was performed 20 days after sowing, when D. insularis plants had a leaf and a well developed root system. The standardization of four plants per experimental unit (plastic pot) was carried out.

In the application of the herbicide treatments, a CO2 pressurized backpack sprayer was used, working at a constant pressure of 2.0 kgf to obtain a spray volume corresponding to 200 L ha-¹. The bar was composed of four fan-type XR 110.02 nozzles spaced 0.5 m apart and positioned at 0.5 m from the target. Climatic conditions during the applications ranged from 26.2 to 30.1 oC, relative humidity from 54 to 60% and wind speed from 3.2 to 4.6 km h-1.

The visual control evaluation was carried out 30 days after the application (DAA), using a 0 to 100% scale, in which 0% corresponds to non-control and 100% to the plant death, proposed by the Brazilian Society of Weed Science (Sociedade Brasileira da Ciência das Plantas Daninhas - SBCPD, 1995Sociedade Brasileira da Ciência das Plantas Daninhas - SBCPD. Procedimentos para instalação, avaliação e análise de experimentos com herbicidas. Londrina: 1995. 42p. ). After completing the visual evaluations, the analysis of the dry matter (g) of plants was carried out, and plants were collected and stored in paper bags and placed in a forced air circulation oven at 65 oC for 72 hours, until reaching constant weight.

The results obtained in the experiments were submitted to the F test in the analysis of variance. Data were adjusted to the logic-biological non-linear regression, represented in Equation 1:

$y=\frac{a}{1+e-\left(\frac{x-{c}_{50}}{b}\right)}$

where y is the control percentage; “x” represents the herbicide dose (g ha-1); and “b” and “a” are curve parameters, so that “b” corresponds to the slope of the curve, “a” represents the upper limit of the curve and “C50” corresponds to the dose providing 50% of the variable response - the lower limit of the curve was considered zero.

After the statistical analyses, it was possible to the resistance degree (RD) related to the use of glyphosate doses and required to provide 50% control on plants, adapted from Seefeldt et al. (1995Seefeldt SS, Jensen JE, Fuerst EP. Loglogistic analysis of herbicide dose-response relationship. Weed Technol. 1995;9:218-27.). According to the results, populations were classified as susceptible with RD?1.0, moderately resistant with an RD between 1.1 and 5.0 and highly resistant with RD>5.0.

# RESULTS AND DISCUSSION

Table 3 shows the resistance levels of the populations, as well as the parameters evaluated for dose-response curves.

Table 3
Resistance degree (RD) of populations of D. insularis collected in agricultural areas, based on control percentage, 30 days after application

During the experimental year 2012, populations R01 and R02 were classified as highly resistant (RD>5.0), population R03 presented moderate resistance (RD = 1.1 to 5.0), while population R04 was classified as susceptible (RD?1.0).

In the experimental year 2013, the populations from Palotina (R06 and R12), Toledo (R09 and R08) and Cascavel (R10) presented high resistance; 62% of the populations were evaluated. The population of Paraguay (R07) presented high resistance (RD = 7.9). Some researchers point out that resistant biotypes found in Paraná come from this region of the neighboring country, since the first occurrence of resistance in D. insularis was found in Paraguay in 2006 (Heap, 2016). However, this information needs to be confirmed with genomic studies between Brazilian and Paraguayan biotypes.

In the experiment conducted in 2015, populations R15 (Iracema do Oeste) and R17 (Jesuítas - Paraná state) were the ones with the highest resistance degrees, with an RD of 7.9 and 12.5, respectively. These locations are about 50 km away from the region of Toledo and São Camilo, where populations showed a high degree of resistance (RD>5.0). Probably, seed dispersal may have occurred among these locations, since they are easily carried by the wind due to the caryopsis included in their chaff (Kissmann and Groth, 1997Kissmann KG, Groth D. Plantas infestantes e nocivas. 2ª.ed. São Paulo: BASF; 1997.825p.). Another important point to be observed is the chemical management used to control D. insularis, which is probably selecting resistant biotypes of this species.

It is possible to observe that the great majority of the evaluated populations ranged from moderately resistant to highly resistant. Licorini et al. (2015Licorini LR, Gandolfo MA, Sorace MA, Osipe R, Cossa CA, Osipe JB. Identificação e controle de biótipos resistentes de Digitaria insularis (L.) Fedde ao glyphosate. Rev Bras Herb. 2015;14:141-7.), in studies with different populations of glyphosate-resistant D. insularis, confirmed the occurrence of this weed’s resistance in the cities of Cascavel (RD = 16.88), Maringá (RD = 9.12) and Palotina (RD = 6.32), in Paraná.

Figures 1, 2 and 3 show data about the dry matter reduction (%) at 30 DAA for the experimental years 2012, 2013 and 2015, respectively.

Figure 1
Dry matter reduction (%) of resistant and susceptible D. insularis populations, 30 days after the application of glyphosate in the experimental year of 2012.

Figure 2
Dry matter reduction (%) of resistant and susceptible D. insularis populations, 30 days after the application of glyphosate in the experimental year of 2013.

Figure 3
Dry matter reduction (%) of resistant and susceptible D. insularis populations, 30 days after the application of glyphosate in the experimental year 2015.

In the 2012 experiment, like what was verified in visual control data, it is possible to observe that populations R01 and R02 were considered highly resistant (RD = 22.14 and 21.56, respectively), while the R03 population was considered moderately resistant (RD = 4.73). Based on the dry matter the population R04 was considered highly resistant (GR = 24.24). This can be explained by the variation in plant size, where the behavior of the populations with higher dry matter accumulation at the beginning of the evaluations adjusted to models that determined the need to use high doses for biomass reduction, even though they were considered susceptible by visual control data (Table 3), and vice versa. However, for the R04 population the dry matter reduction data confirm the high resistance in disregard of visual control data at 30 DAA.

To reduce by 50% the dry matter of the shoot of the R04 population, a dose of 1,224.61 g ha-1 was necessary, while in the susceptible population S05 the same reduction was provided by a dose of 50.52 g ha-1. Similarly, Reinert et al. (2013Reinert CS, Prado ABCA, Christoffoleti PJ. Curvas de dose-resposta comparativas entre os biótipos resistentes e suscetível de capim-amargoso (Digitaria insularis) ao herbicida glyphosate. Rev Bras Herb. 2013;12:260-7.) found that a glyphosate dose of 108.70 g ha-1 reduced by 50% the shoot dry matter of susceptible D. insularis populations, requiring the use of 1,811,20 g ha-1 to obtain the same effect in the resistant population; these values are similar to those obtained in this study.

Melo et al. (2012Melo MSC, Rosa LE, Brunharo CACG, Nicolai M, Christoffoleti PJ. Alternativas para o controle químico de capim-amargoso (Digitaria insularis) resistente ao glyphosate. Rev Bras Herb. 2012;11:195-203.), while evaluating the control of glyphosate-resistant and -susceptible D. insularis, observed that the glyphosate doses required for a satisfactory control were 310.9 and 632.6 g ha-1 for resistant plants, and 175.5 and 474.1 g ha-1 for susceptible plants.

In the experimental year 2013, only the R07 and R08 populations showed the lowest dry matter reductions with the highest glyphosate dose, in relation to the others (Figure 2). For the 2015 experiment, it was observed that the R21 population was considered highly resistant (RD = 5.84), but it is possible to observe that at the highest evaluated dose, the reduction was equivalent to that of the susceptible population S22 (Figure 3). The other populations were considered moderately resistant.

Generally speaking, these results confirm the occurrence of resistance to glyphosate in populations of D. insularis in agricultural areas of western Paraná.

After submitting the questionnaire to producers from the collection areas of the period from 2012 to 2015, data were transformed into percentage and presented graphically (Figure 4).

Figure 4
Results of the questionnaires submitted to agricultural producers for the experiments conducted in 2012, 2013 and 2015.

Approximately 83% of the 22 interviewed producers practiced the no-tillage system, and 100% used the soybean/maize succession system. Producers justify the use of this system as the best economic alternative for their property.

The main herbicides were glyphosate at the mean dose of 928.8 g ha-1 and 2,4-D at the mean dose of 443.3 g ha-1, applied alone or in a tank-mix in the desiccation management of weeds for the no-tillage system; they were used by 90% of the producers.

It is also worth mentioning that 96% of the producers use transgenic varieties with glyphosate resistance technology. This contributes to the excessive use of glyphosate applications per year (³3) and at higher doses than the recommended one (³720 g ha-1), since only 78% of the producers use the recommended dose. These results support those obtained by Vargas et al. (2013Vargas L, Nohatto MA, Agostinetto D, Bianchi MA, Paula JM, Polidoro E et al. Management practices x Euphorbia heterophylla resistance to ALS-inhibitors and tolerance to glyphosate in Rio Grande do Sul. Planta Daninha. 2013;31:427-32.), which found that approximately 80% of farmers in Rio Grande do Sul performed three or more applications of glyphosate per year, and 76% cultivated RR soybean for five years or more. These authors also emphasized that the intensive use of glyphosate can exert a high selection pressure on weed populations, and the advantages derived from the control of these weeds in transgenic crops can be lost in the short or medium run, making it impossible to use this technology.

It is worth highlighting that failures in the control of D. insularis populations with the use of glyphosate have occurred in areas where the number of applications of this herbicide has gradually increased. Thus, the intensified use of glyphosate increases the risk of selection of resistant biotypes (Beckie and Reboud, 2009Beckie HJ, Reboud X. Selecting for weed resistance: herbicide rotation and mixture. Weed Technol. 2009;23:363-70.; Melo et al., 2012Melo MSC, Rosa LE, Brunharo CACG, Nicolai M, Christoffoleti PJ. Alternativas para o controle químico de capim-amargoso (Digitaria insularis) resistente ao glyphosate. Rev Bras Herb. 2012;11:195-203.).

Only 36% of the interviewed farmers understand the concept of resistance, although 60% affirm that they adopt practices to avoid weed resistance selection in their cultivated areas.

Approximately 82% of the producers stated that they perform the correct regulation of the spraying equipment. However, when asked if they could explain the adjustment and calibration procedure of sprayers, only 65% could explain them clearly and correctly. The lack of attention by producers on the regulation and calibration of sprayers before the applications can indicate the use of doses above or below the recommendation, besides increasing the risk of contamination of the applicator and of the environment.

Bauer et al. (2009Bauer FC, Pereira FAR, Scheeren BR, Braga LW. Diagnostico das condições, tempo de uso e manutenção de pulverizadores no estado de Mato Grosso do Sul. Eng Agríc. 2009;29:501-7.) verified that, after diagnosing conditions, time of use and maintenance in sprayers in the State of Mato Grosso do Sul, 51.1% of them had nozzles that should have been replaced, since they had a flow variation higher than the established limits and the equipment was being used outside the legal specifications.

In general, these results may contribute to a better understanding of the selection process of herbicide-resistant weed biotypes in agricultural production areas, since it was verified that most producers use only a system of culture succession and doses of glyphosate, successively, throughout the year and for several years.

Based on the results, glyphosate resistance was confirmed in 89.5% of the evaluated D. insularis populations, and it is possible to infer that the agronomic practices adopted in the agricultural areas contribute to the selection process of glyphosate-resistant D. insularis.

# ACKNOWLEDGEMENTS

This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) - Finance Code 001 and by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil, with scholarships.

• Bauer FC, Pereira FAR, Scheeren BR, Braga LW. Diagnostico das condições, tempo de uso e manutenção de pulverizadores no estado de Mato Grosso do Sul. Eng Agríc. 2009;29:501-7.
• Beckie HJ, Reboud X. Selecting for weed resistance: herbicide rotation and mixture. Weed Technol. 2009;23:363-70.
• Christoffoleti PJ, López-Ovejero R. Principais aspectos da resistência de plantas daninhas ao herbicida glyphosate. Planta Daninha. 2003;21:507-15.
• Heap I. The international survey of herbicide resistant weeds [Internet]. [accessed on: 26, oct 2016]. Available at: Available at: http://www.weedscience.com
» http://www.weedscience.com
• Kissmann KG, Groth D. Plantas infestantes e nocivas. 2ª.ed. São Paulo: BASF; 1997.825p.
• Lacerda ALS, Victória Filho R. Curvas dose-resposta em espécies de plantas daninhas com o uso do herbicida glyphosate. Bragantia. 2004;63:73-9.
• Licorini LR, Gandolfo MA, Sorace MA, Osipe R, Cossa CA, Osipe JB. Identificação e controle de biótipos resistentes de Digitaria insularis (L.) Fedde ao glyphosate. Rev Bras Herb. 2015;14:141-7.
• Machado AFL, Ferreira LR, Ferreira FA, Fialho CMT, Tuffi Santos LD, Machado MS. Análise de crescimento de Digitaria insularis Planta Daninha. 2006;24:641-7.
• Melo MSC, Rosa LE, Brunharo CACG, Nicolai M, Christoffoleti PJ. Alternativas para o controle químico de capim-amargoso (Digitaria insularis) resistente ao glyphosate. Rev Bras Herb. 2012;11:195-203.
• Monquero PA, Christoffoleti PJ. Manejo de populações de plantas daninhas resistentes aos herbicidas inibidores da acetolactato sintase. Planta Daninha. 2001;19:67-74.
• Reinert CS, Prado ABCA, Christoffoleti PJ. Curvas de dose-resposta comparativas entre os biótipos resistentes e suscetível de capim-amargoso (Digitaria insularis) ao herbicida glyphosate. Rev Bras Herb. 2013;12:260-7.
• Sociedade Brasileira da Ciência das Plantas Daninhas - SBCPD. Procedimentos para instalação, avaliação e análise de experimentos com herbicidas. Londrina: 1995. 42p.
• Seefeldt SS, Jensen JE, Fuerst EP. Loglogistic analysis of herbicide dose-response relationship. Weed Technol. 1995;9:218-27.
• Vargas L, Nohatto MA, Agostinetto D, Bianchi MA, Paula JM, Polidoro E et al. Management practices x Euphorbia heterophylla resistance to ALS-inhibitors and tolerance to glyphosate in Rio Grande do Sul. Planta Daninha. 2013;31:427-32.

# Publication Dates

• Publication in this collection
2018