Risk assessment of pollen‐mediated gene flow from Ga1‐m field corn to dent‐sterile Ga1‐s popcorn

Abstract

The popcorn industry in the United States is at risk of genetic contamination because it utilizes the gametophyte factor 1 gene (Ga1) as a barrier against pollen‐mediated gene flow (PMGF) from field corn (Zea mays L.). Popcorn with the Ga1‐s allele accepts pollen only from Ga1‐s corn, allowing for field corn and popcorn to be nearby without isolation. Germplasm is being introduced to the United States to increase field corn diversity that unknowingly contains the Ga1‐m allele, which can overcome Ga1‐s selectivity and pollinate popcorn. The risk to the popcorn industry has been underassessed. Experiments were conducted to model the frequency of PMGF from Ga1‐m field corn to Ga1‐s popcorn under field conditions and to evaluate the role of wind speed and direction using a concentric donor–receptor design in 2017 and 2018 in Nebraska, USA. The PMGF to white popcorn was detected using a field corn pollen donor with yellow kernel color (dominate) and further confirmed with a molecular test. Popcorn kernels were harvested from cardinal and ordinal directions from 1 to 70 m. More than 7 million kernels were screened to detect the PMGF. Information‐theoretic criteria were used to select the best‐fit model. The greatest PMGF (1.6– 4.1%) was detected at 1 m and declined with distance. The PMGF was detected at 70 m, the maximum distance tested. Amplification of Ga1 followed by restriction enzyme digest confirmed that yellow kernelled progeny were hybrids from the result of PMGF. This is the first assessment of PMGF from field corn to popcorn, and the results are alarming for the popcorn industry and export market.

Pollen‐mediated gene flow from herbicide‐resistant yellow corn to non‐genetically engineered food‐grade white corn

“…The greatest PMGF of 0.0621 (6.21%) to 0.1950 (19.5%) was noted at the nearest distance of 1 m (Table 3), with a sharp exponential decline as distance increased from the pollen donor (Figure 5). Averaged across two site-years, PMGF decreased to 0.0450 (4.50%) at 2 m, 0.0305 (3.05%) at 3 m, 0.0111 (1.11%) at 5 m, 0.0076 (0.76%) at 10 m, 0.0029 (0.29%) at 30 m, and 0.0028 (0.28%) at 50 m. The observed frequency of PMGF and its relationship with distance from the pollen donor generally agree with other studies conducted in Canada (Ma et al, 2004), Mexico (Baltazar et al, 2015;Weber et al, 2007), Spain (Messeguer et al, 2006;Pla et al, 2006), and the United States (Barnes et al, 2020;Goggi et al, 2006;Halsey et al, 2005). This is the first study to model and highlight the role of wind in PMGF at closer distances (1-10 m; Table 4 and Figure 6).…”

“…However, corn pollen grains can cross-pollinate further distances and thus, PMGF from Enogen corn to white corn is the greatest concern. PMGF in corn is influenced by several factors, such as the size of the pollen donor and receptor blocks, the distance from the pollen donor, and the number of days of flowering synchrony (Barnes et al, 2020;Della Porta et al, 2008;Palaudelmàs et al, 2012;Viaud et al, 2008). Additionally, wind conditions as represented by wind speed, direction, and frequency can affect gene flow, especially for wind-pollinated crops such as corn (Della Porta et al, 2008;Halsey et al, 2005;Kuparinen et al, 2007).…”

“…Therefore, field experiments were conducted with the following objectives: (1) to determine the frequency of PMGF from GE yellow corn to non-GE white corn under field conditions and (2) to evaluate the significance of distance from pollen donor, wind speed, and direction on PMGF. The experimental design was a Nelder wheel, that is, a concentric donor-receptor design (Barnes et al, 2020). In this design, the central "pollen-donor" block of 20 × 20 m was surrounded by an outer "pollen-receptor" block of 120 × 120 m. Due to the unfulfilled request to buy Enogen corn seed, commercially available herbicide-resistant Enlist corn (GE yellow corn; resistant to 2,4-dichlorophenoxyacetic acid (2,4-D) choline, glyphosate, glufosinate, and quizalofop) was used as the pollen donor, while non-GE white corn served as the pollen receptor.…”

“…where n i represents the sample size of the ith group, and the sum of the squared differences between the observed (γ i ) and fitted ( μ𝑖 ) values for the ith group of observations is divided by the variance of the fitted values [μ 𝑖 (𝑛 𝑖 − μ𝑖 )/𝑛 𝑖 ], where μ i is estimated from μ𝑖 . The n−k−1 represents the degree of freedom for Pearson's χ2 test, with n and k indicating the total number of groups and parameters, respectively (Barnes et al, 2020;Ganie & Jhala, 2017;Sarangi et al, 2017). [Correction added on 6 April, after first online publication: In the paragraph below equation 3, k is replaced with j in sentence "k represents the combined number of models under comparison".]…”

“…The estimates of coefficients with their standard errors for the best-fit exponential decay models are given in Table 4. Ma et al (2004) and Barnes et al (2020) used exponential decay and double decay models to explain gene flow in white corn and popcorn, respectively. Similarly, the exponential decay or double decay model was successfully fitted to gene flow in rice (Oryza sativa L.) (Rong et al, 2010), wheat (Triticum aestivum L.) (Beckie et al, 2011;Loureiro et al, 2012), and weed species such as common lambsquarters (Chenopodium album L.) (Yerka et al, 2012), giant ragweed (Ambrosia trifida L.) (Ganie & Jhala, 2017), kochia [Bassia scoparia (L.) A. J. Scott] (Beckie et al, 2016), shattercane (Schmidt et al, 2013), and waterhemp (Sarangi et al, 2017).…”