Effects of the glyphosate-resistance gene and herbicides on soybean: Field trials monitoring biological nitrogen fixation and yield

• Effects of the RR transgene, herbicide application and weed management on BNF with soybean were evaluated.
• In general, the transgene negatively affected BNF parameters, but not grain yield.
• No differences were observed between the effects of glyphosate and conventional herbicides.
• No differences were observed between RR soybean + glyphosate, and conventional soybean + conventional herbicides.
• Differences were attributed mostly to geographic location, cropping season and cultivar.

The commercial use of glyphosate-resistant (also known as Roundup Ready®, GR or RR) soybean was initiated in 1996 in the United States. This genetically engineered crop now occupies 75.4 million ha worldwide, 20.6 million of which are in Brazil where it occupies 86% of the total area cropped with the legume. Biological nitrogen fixation (BNF) is critical for economic sustainability of soybean in Brazil; therefore, to investigate the effects of the RR gene—using pairs of nearly isogenic cultivars—and herbicides on BNF, we conducted an extensive series of field experiments for three cropping seasons. The experiments were set up at six sites with five treatments, three pairs of nearly isogenic cultivars, and six replicates. The treatments consisted of: (T1) RR soybean + glyphosate; (T2) RR soybean + conventional herbicides; (T3) conventional parental soybean + conventional herbicides; (T4) RR soybean + hand weed control; and (T5) conventional parental soybean + hand weed control. Parameters of nodulation, plant biomass production, total N and ureide-N were evaluated at the V4 and R2 stages of growth, and grain yield and total N in grains were evaluated at crop maturity. Data were analyzed by ANOVA, analyses of contrasts, and multivariate analyses considering a pool of six variables, denominated as symbiotic efficiency (SyEf). The comparison of the pairs of non-transgenic and RR soybean cultivars showed that the transgenic trait negatively affected some BNF variables, but over a three-year period these effects had no significant impact on soybean grain yield. No consistent differences between glyphosate and conventional herbicide application were observed on BNF-associated parameters. When compared to conventional soybean and conventional herbicides, weed-management strategy with RR soybean and glyphosate did not affect symbiotic efficiency. In addition, at three sites, grain yields increased in the treatments with glyphosate and RR soybean over the three cropping seasons. The results from the multivariate analyses indicate that BNF and yield parameters were more affected by location, cropping season and cultivar than by the transgene, herbicides, or weed-management strategy. Despite the lack of effects of the transgene on yield in the three-year period, longer-term effects on BNF and N accumulation should be monitored.