Efficacy of drought-tolerant and insect-protected transgenic TELA® maize traits in Nigeria

Assessment of efficacy of drought tolerance (DT) and insect protection (Bt) genes in maize genotypes is invaluable for commercialization and production of transgenic maize in Nigeria. Seven maize hybrids, known as TELA® maize, with stacked events of Bt insect protection (MON89034) and drought tolerance (MON87460; DroughtGard®) and their respective non-GM versions (isohybrids) developed through the TELA Maize Project were evaluated in confined field trial site at Zaria in 2020 and 2021. The objective was to assess the efficacy of stacked DT and Bt genes to seek deregulation and commercialization of both traits in Nigeria. Significant (P < 0.05–0.01) differences were observed among genotypes (G), environments (E) and genotype × environment interaction (GEI) for grain yield and most other traits under stem borer (moth species) and fall armyworm infested, drought stress, and optimum-moisture conditions, except E and GEI under drought. TELA® GM hybrids with Bt MON89034 had 19% higher yield than their non-GM isogenic versions, and 40% higher yield than the commercial checks under the target pests infestation. The foliar damage score of all the TELA® GM genotypes was ≤ 2 relative to their non-GM isogenic versions which scored ≥ 4, indicating the effectiveness of the Bt MON89034 gene in conferring resistance against stem borer and fall armyworm. Under moderate drought, pairwise comparison showed TELA® GM Hybrid 1–1 and Hybrid 2–1 had 12.4–20.4% higher (P < 0.01) yield than their isogenic versions. Under optimum-moisture condition with pests controlled, the TELA® GM and their isogenic hybrids were similar, but both had 32% higher yield than the commercial checks. Adoption of TELA® GM technology by farmers as adaptation strategy to cope with climate change, will ensure sustainability of maize production and productivity in Nigeria.

Abstract Assessment of efficacy of drought tolerance (DT) and insect protection (Bt) genes in maize genotypes is invaluable for commercialization and production of transgenic maize in Nigeria. Seven maize hybrids, known as TELA® maize, with stacked events of Bt insect protection (MON89034) and drought tolerance (MON87460; DroughtGard®) and their respective non-GM versions (isohybrids) developed through the TELA Maize Project were evaluated in confined field trial site at Zaria in 2020 and 2021. The objective was to assess the efficacy of stacked DT and Bt genes to seek deregulation and commercialization of both traits in Nigeria. Significant (P < 0.05-0.01) differences were observed among genotypes (G), environments (E) and genotype × environment interaction (GEI) for grain yield and most other traits under stem borer (moth species) and fall armyworm infested, drought stress, and optimummoisture conditions, except E and GEI under drought. TELA® GM hybrids with Bt MON89034 had 19% higher yield than their non-GM isogenic versions, and 40% higher yield than the commercial checks under the target pests infestation. The foliar damage score of all the TELA® GM genotypes was ≤ 2 relative to their non-GM isogenic versions which scored ≥ 4, indicating the effectiveness of the Bt MON89034 gene in conferring resistance against stem borer and fall armyworm. Under moderate drought, pairwise comparison showed TELA® GM Hybrid 1-1 and Hybrid 2-1 had 12.4-20.4% higher (P < 0.01) yield than their isogenic versions. Under optimum-moisture condition with pests controlled, the TELA® GM and their isogenic hybrids were similar, but both had 32% higher yield than the commercial checks. Adoption of TELA® GM technology by farmers as adaptation strategy to cope with climate change, will ensure Introduction Maize (Zea mays L.) is one of the most important food crops in Sub-Saharan Africa (SSA) and has immense contribution to food security in the subregion. Maize is a staple food crop accounting for about 20% of the calorie intake of nearly half of the human population in this region. It is a multipurpose crop, as the grain can be used as food in several forms such as whole-maize foods, wet-ground maize foods, snacks and bread, porridge, and beverages. It is the most widely grown crop by smallholder farmers which covers over 30 million hectares in SSA (FAO 2019a, b). Maize ranks third in production after rice and wheat in the world (Olaniyan 2015). In Nigeria, maize is one of the most important staple crops predominately grown and consumed by smallholder farmers, accounting for 20% and 16% of national of calorie and protein needs, respectively (HarvestPlus 2004).
Maize production in Nigeria is constrained by biotic, abiotic, and socioeconomic factors causing variation in grain yield (Olaoye et al. 2009;Badu-Apraku and Oyekunle 2012;Oyekunle et al. 2015). Drought is a major abiotic stress contributing to severe maize yield loss in the lowland savanna belt of West and Central Africa (WCA) (Fajemisin et al. 1985). The risk of drought stress is high across the various agroecologies particularly in the Sudan savanna due to unreliable and uneven distribution of rainfall (Eckebil 1991), resulting in significant yield loss (40-90%) when the drought occurs at the most sensitive stages of the crop such as flowering and grain-filling (Badu-Apraku and Oyekunle 2012; Oyekunle et al. 2015).
Insect pests, particularly stem borer (moth species) and fall armyworm are the most economically important biotic stress to maize production in all major maize growing areas of the country. There are three lepidopteran stem borer species, namely the African stem borer (Busseola fusca (Fuller)), the spotted stem borer (Chilo partellus (Swinhoe)) and the pink stem borer (Sesamia calamistis (Hampson)) that cause significant damage to maize production with yield loss ranging from 15 to 40% in Nigeria depending on the level of infestation. B. fusca is more predominant in the southern and northern Guinea savanna, S. calamistis is prevalent in the forest and derived savanna, while C. partellus is more in the mid-altitude and lowlands of Nigeria. In addition, fall armyworm (Spodoptera frugiperda), has become an important pest of maize since arriving in Africa in 2016. Fall armyworm poses a significant risk for more than 12.5 million maize-producing smallholder farmers in Nigeria (Abubakar 2021). Three quarters of the 7.8 million hectares of maize land area was infested by the pest in 2017, causing a loss of over one million tons of maize worth more than USD 268 million (FAO 2018).
Maize under drought suffers more from stem borer infestation, making the situation more complex (Pu et al. 2019). Thus, drought and insect pest damage may lead to complete crop failure under severe cases, further compounding the food insecurity and hunger situation in the country. Drought tolerant maize varieties have been developed through conventional breeding and deployed to farmers in Nigeria. It is hypothesized that the incorporation of drought tolerance (MON87460; CspB, DroughtGard®) gene from Bacillus subtilis will further increase the level of tolerance of the conventional drought tolerant maize with an additional yield advantage of at least 10%. This is based on the proof-of-concept studies that showed across four years of testing, the CspB event (MON87460) provided an average yield benefit of 10.5% under managed drought stress environments in the USA (Nemali et al. 2015;Castiglioni et al. 2008).
The twin problem of drought and insect pests ravaging maize production in Nigeria could be addressed through the development and deployment of climatesmart maize varieties with combined drought tolerance and insect protection for smallholder farmers using transgenic or genetically modified (GM) technology. Drought-tolerant (DT) GM maize, known as DroughtGard® hybrids were first commercialized in the USA in 2012. About four years later, the varieties were already making great impact with 22% of the total maize hectarage in the USA planted to DT GM maize, and up to 42% adoption rate (> 1.6 m ha) in the state of Nebraska (AgroNews 2019). This impact demonstrates the importance of the technology in addressing the challenge of frequent drought events.
Further, insect-protected Bt maize incorporating Bt (MON810 [Cry1Ab]/MON89034 [Cry1A.105/Cry2Ab2]; from Bacillus thuringiensis) has been commercialized globally including South Africa for over 20 years (ISAAA 2018). Therefore, the current advancement in variety development using transgenic technology to stack drought tolerance (DT) and insect protection (Bt) genes could contribute to a more effective and sustainable approach to overcome both drought and insect pests complex affecting maize production as an adaption strategy to cope with climate change in Nigeria.
In view of this, the Institute for Agricultural Research (IAR) Zaria, Nigeria in collaboration with other institutions in the TELA Maize Project partnership involved in the development and commercialization of the Bt maize hybrids and stacked DT × Bt traited maize hybrids, trademarked as TELA® maize hybrids is addressing the challenges posed by both drought and insect pests in Africa including Nigeria. Therefore, the objective of this study was to assess the efficacy of stacked drought tolerance and insectprotection genes to seek for environmental release and commercialization of both traits in Nigeria.

Materials and methods
Experimental Site and development of confined field site A confined field trial (CFT) was developed with stateof-the-art drip irrigation facilities covering two-hectare land area at Institute for Agricultural Research Farm Samaru, Zaria. Samaru, Zaria, Kaduna State is in the northern Guinea savanna agro-ecology of Nigeria. Confined field trial is a field trial conducted under special regulatory conditions with limited access of unauthorized individual to the experimental materials. The state-of-the-art irrigation facilities was done by providing all the necessary facilities for uninterrupted supply of irrigation water to the plants using drip irrigation system. It is on latitude 11° 11′ N and longitude 7° 38′ E with altitude of 640 m asl. The site receives an average of 1200 mm annual rainfall with minimum and maximum temperature of 13.5 °C and 42 °C. The site is characterized by sandy loam soil type.

Experimental maize genotypes
Seven maize genotypes with breeding stacked events of Bt insect protection (Bt gene, Event MON89034) and drought tolerance (DT, CspB gene -Drought-Gard®, Event MON87460) and their respective non-GM versions (iso-hybrids) were imported from Bayer, USA in 2020. The seeds were received at the Department of Plant Science, Ahmadu Bello University, Zaria and stored in the confined storage facility of Biotechnology laboratory of the Department of Plant Science, Samaru, Zaria until planting. The seven transgenic (GM) maize hybrids, seven isogenic (non-GM) hybrids, and two commercial hybrids as local checks were the 16 experimental materials used in the trials (Table 1). The CFTs were conducted during the 2019/20 and 2020/21 dry seasons to evaluate the efficacy of the GM genotypes for drought tolerance and insect protection against stem borers and fall armyworm. Three management conditions were applied in the trials: (a) Fully irrigated with insecticide application for protection against the target pests and no artificial infestation (Optimum Trial); (b) Fully irrigated with artificial infestation with the target insect pests but no insecticide application (Bt Trial). The Bt trial was conducted with artificial infestation using 20 larvae of stem borer (S. calamistis) per plant at 3, 4 and 5 weeks after planting (WAP) while fall armyworm was under natural infestation; and (c) Managed drought by withholding irrigation two weeks before flowering until 3 weeks after flowering, with no artificial infestation and application of insecticide to protect against the target pests (DT Trial). Randomized complete block design was used for all the three trials. In all the CFTs, the number of replications for DT, Bt, and Optimum were six, four, and three, respectively. Each experimental unit consisted of 2 rows of 5 m long with spacing of 0.75 between rows and 0.25 m between plants. Two seeds per hill were planted, and the seedlings were thinned to one per stand about 2 weeks after emergence to give a target plant population density of 53,333 plants ha −1 . A compound fertilizer (NPK 20:10:10) was applied at the rate of 60 kg N ha −1 2 weeks after planting (WAP) for all experiments. An additional 60 kg N ha −1 urea was top-dressed three weeks later. In all the experiments, the trials were kept weed-free by manual weeding.

Data collection
In each experimental plot, the following data were recorded: (1) days to anthesis as the number of days from planting to when 50% of the plants had shed pollen; (2) days to mid-silk as the number of days from planting to when 50% of the plants had emerged silks; (3) anthesis-silking interval (ASI) was computed as the interval in days between 50% anthesis and 50% silking; (4) plant height was measured as distance from the base of the plant to the first tassel branch; (5) ear height was measured as the distance from the base of the plant to the node bearing the upper ear; (6) husk cover was rated on a scale of 1-5, where 1 = husks tightly arranged and extended beyond the ear tip and 5 = ear tips exposed; (7) plant aspect was rated on a scale of 1-9, where 1 = excellent plant conformation and 9 = poor plant conformation; (8) ear aspect was scored on a 1-9 scale, where 1 = clean, uniform, large, and well-filled ears and 9 = rotten, variable, small and partially or poorly filled ears; (9) number of ears per plant (EPP) was computed as the total number of ears with at least one grain at harvest divided by the number of plants at harvest; and (10) foliar damage in Bt trial was rated on a scale of 1-9, where 1 = No visible leaf feeding damage and 9 = plant dying as a result of foliar damage; (11) Leaf death (LD) due to the impact of drought stress was scored at 80 days after planting on a scale of 1-10, where 1 = less than 10% dead leaf area and 10 = more than 90% dead leaf area; (12) grain yield, adjusted to 15% moisture content, was computed from the shelled grain weight.

Statistical analysis
Combined analysis of variance (ANOVA) across environments was done on plot means for grain yield and other agronomic traits with PROC GLM in SAS using a RANDOM statement with the TEST option (SAS Institute 2002). Pairwise comparison of GM versus Non-GM genotypes was carried out using a t-test. Shapiro-Wilk test was used to establish a normal distribution of the data and Bartlett's test was used to test homogeneity of variances before paired t-test was done. In the tests, the probability values were greater than 0.05, indicating that the data met the assumptions for conducting paired t-test. In the combined ANOVA, genotype was considered as fixed factor while year hereafter referred to as environment, and replicates were considered as random factors. The linear model for the combined ANOVA is as follows: where Y klmi is the observed measurement of trait i of m genotype in l replicate within k environment, μ i is mean effect, E ki is the effect of environment k on trait i, R(E) l(k)i is the effect of replication l within environment k on trait i, G mi is the effect of genotype m on trait i, GE kmi is the effect of the interaction between genotype m and environment k on trait i, and ε klmi is the experimental error effect associated with genotype m and replication l within environment k on trait i.

Analysis of variance of transgenic TELA maize hybrids
The results of combined analysis of variance revealed that there were highly significant (P < 0.01) differences between environments for all measured traits under infested and optimum conditions, except EPP under infested condition and husk cover under optimum conditions (Table 2). In contrast, there was no significant difference in environmental effect under drought condition. There were highly significant (P < 0.01) differences among genotypes for all measured traits under infested, drought and optimum conditions, except ASI and plant height under optimum condition (Table 2). Furthermore, the mean squares for genotype × environment interaction (GEI) was significant (P < 0.05-0.01) for grain yield, days to anthesis and silking, ear height, husk cover, and foliar damage under infested condition. In contrast, there was no significant difference in GEI under drought conditions. However, the GEI was significant (P < 0.05) for grain yield, days to anthesis and silking, husk cover, plant and ear aspects, and EPP under optimum condition ( Table 2).
Pairwise comparison for foliar damage score also revealed highly significant (P < 0.01) difference between GM and non-GM for all tested genotypes. The foliar damage score of all the test GM genotypes was ≥ 2 relative to their non-GM isogenic versions (score ≥ 4) under artificial infestation of stem borer and natural infestation of fall armyworm (Table 3). There was no significant difference in the foliar damage of the two commercial checks, with a mean damage score of 4.2.
All hybrids attained anthesis on < 70 days, except commercial hybrid check 2 (73 days) ( Table 5). The TELA hybrids exhibited greater synchronization of anthesis and silking as measured by ASI, ranging from 0.9 to 1.8 days, relative to the commercial check hybrids with ASI of 1.9-2.0 days suggesting that they may be more drought tolerant than the commercial check hybrids. In general, TELA GM hybrids were taller and had greater ear height by 4-8 cm than the non-GM hybrids (Table 5).
On average, under moderate drought, the GM and non-GM hybrids had similar yield (3.2-3.3 t ha −1 ). However, pairwise comparison revealed two GM genotypes, Hybrid 1-1 and Hybrid 2-1, had 12.4-20.4% (0.42-0.46 t ha −1 ) higher (P < 0.01) yield than their isogenic (non-GM) hybrids (Table 4). The grain yield reduction of non-GM hybrids 1-2 and hybrid 2-2 under moderate drought was 11.2% and 17%, respectively when compared with their isogenic versions, indicating that the drought stress level imposed resulted in meaningful yield reductions of up to 17%. Grain yield of the hybrids ranged from 2.2 t ha −1 for commercial hybrid (Check 1) to 3.8 t ha −1 for Hybrid 2-1 GM and Hybrid 4-1 GM with a mean of 3.2 t ha −1 (Table 5). On average, the GM hybrids had 3% higher yield than their non-GM isogenic versions, and 22% higher yield than the commercial checks under managed drought stress (Table 5).
Similarly, pairwise comparison among the GM and their isogenic hybrids showed in general, as required for regulatory trials, that the genotypes were similar under optimum condition with chemical control of the target pests (Table 4). Grain yield of the hybrids ranged from 5.0 t ha −1 for Hybrid 2-2 Non-GM to 8.1 t ha −1 for Hybrid 1-1 GM with a mean of 6.2 t ha −1 (Table 5). On average, the GM hybrids had    9% higher yield than their non-GM isogenic versions, and 32% higher yield than the commercial checks (Table 5).

Discussion
The presence of significant environmental effects for most measured traits under infested and optimum conditions indicated that each environment (year) is unique for the expression of the performance of the genotypes under different management conditions. The lack of statistically significant differences in environmental effect under drought condition indicated consistency in the performance of the genotypes under managed drought condition. This result suggests that the drought management imposed over years were similar for the expression of the genotypes. The presence of genetic variation for characters of interest in crops is invaluable in breeding for crop improvement. The significant variation observed among the genotypes for all measured parameters under infested, drought, and optimum conditions, except ASI and plant height under optimum condition indicated that adequate genetic variability existed among the genotypes to allow significant progress from selection for improvements. If both MON89034 and MON87460 Events are deregulated by the regulatory authority, the presence of variability among the genotypes under different management conditions will allow for the identification of suitable genotypes    (2012), and , who suggested the possibility of identifying high-yielding maize hybrid(s) with either narrow or broad adaptation in Nigeria. The significant GEI observed for grain yield and most other measured characters under infested and optimum conditions revealed the differential ranking of the genotypes in different environments under both management conditions. The differential responses of the genotypes with both Events to varying environments suggested the need for extensive testing of maize genotypes in multiple environments before genotype recommendations as previously suggested (Badu-Apraku et al. 2011b, 2015. However, the lack of significant GEI for all measured parameters under drought indicated that the expression of hybrids with both Events would be consistent in drought stress environments when the drought Event MON87460 is deregulated. An important objective of the present study was to assess the efficacy of the insect protection trait (MON 89034) and drought tolerance trait (MON 87460) in maize genotypes. The observed grain yield advantage of 21.7 to 32.6% especially in two GM genotypes (Hybrid 1-1 and Hybrid 5-1) over their non-GM, isogenic hybrids, implied the positive efficacy of the transgene and further strengthened the case Table 5 Influence of Bt gene (MON 89034) and DT gene (MON 87460) on growth and yield parameters of maize genotypes under drought, infestation of stem borer and fall army-worm and optimum conditions in confined field trials across 2020 and 2021 dry seasons in Zaria, Nigeria ASI anthesis-silking interval (days), DS manage drought, OP optimum condition, Inf infested condition a Foliar damage (scale 1-9), where 1 = where 1 = No visible leaf feeding damage and 9 = plant dying because of foliar damage b Leaf death (scale 1-10), where 1 = less than 10% dead leaf area and 10 = more than 90% dead leaf area to deregulate the transgene for the benefit of Nigerian farmers to cultivate the TELA® maize hybrids with the Events. In addition, the observed 19% yield advantage of GM over non-GM further suggests the efficacy in the expression of the Bt MON89034 gene in the maize genotypes for controlling the target pests, especially the invasive fall armyworm. Furthermore, the presence of highly significant differences in the foliar damage scores of GM and non-GM further confirm the efficacy of the Bt gene in the GM genotypes in conferring resistance against stem borer and fall armyworm pests in Nigeria. These observations confirmed that Bt gene, being the only difference between the GM and their isogenic hybrids, confers protection against target insect pests. The lack of significant difference in the foliar damage of the two commercial checks was expected and indicated the efficacy of the artificial infestation with stem borer and natural infestation of fall armyworm methods in discriminating among the genotypes.
The yield advantage of two GM genotypes, Hybrid 1-1 and Hybrid 2-1 of 12.4-20.4% over their isogenic (non-GM) hybrids under drought stress indicated the expression of the efficacy of MON87460 gene in the two genotypes. However, on average, the lack of significant differences between GM genotypes and their isogenic non-GM under drought indicated low levels of genetic variability among the paired genotypes for drought tolerance. The plausible explanation for the observed results could be due to the high intensity of drought imposed during flowering and grain filling periods. This observation is evidence in the grain yield reduction of 50.8% recorded under drought when compared with optimum condition, thus suggesting that when the drought tolerant (MON87460) trait is deregulated, hybrids with the trait would need further multilocation performance evaluation under random drought stress before making genotype recommendations for commercialization.
The observed similarity in pairwise comparison among the GM and their isogenic hybrids under optimum condition was expected. The slight difference in grain yield of 0.6 t ha −1 of GM over the non-GM, isogenic hybrids could be due to incomplete protection from the use of chemical insecticide against the target pests, especially the fall armyworm, among the non-GM hybrids. On average, the commercial checks had 16% lower yield than the test genotypes (GM and isogenic, non-GM hybrids) under optimum condition, indicating that they are likely to have much lower yield potential than the test genotypes, in addition to the incomplete protection from the use of the insecticides against the target pests.