Amylase/trypsin inhibitors (ATIs) levels in wheat event IND-ØØ412-7 are similar to non-transgenic wheat

TL;DR

This study found that a type of wheat protein linked to baker's asthma is present at similar levels in a transgenic wheat variety as in regular wheat.

Contribution

The study confirms that transgenic HB4 wheat does not alter ATI levels compared to non-transgenic wheat.

Findings

ATI levels in HB4 wheat were not significantly different from non-transgenic varieties.

ATI levels are influenced more by genetics and environment than transgenic modification.

HB4 wheat ATI levels fall within the natural variation observed in other wheat varieties.

Abstract

Despite being a main protein supplier in the human diet, wheat proteins represent a health challenge for some people. Besides the well-known celiac disease caused by gluten proteins, there is an occupational illness known as baker´s asthma. Amylase/trypsin inhibitors (ATIs) have been reported to be the major group of wheat proteins responsible for bakers´ asthma. As part of the characterization of stress-tolerant (HB4® technology) transgenic wheat (event IND-ØØ412-7, HB4 wheat), the level of the seven ATIs (0.28, 0.19 + 0.53, CM2, CM3, CM16, and CM17) was determined and compared to non-transgenic varieties. The materials tested in this study included the transgenic event in two different genetic backgrounds, their conventional counterparts (cv. Algarrobo and cv. Basilio), and five additional commercial varieties. Grain samples were obtained from field trials in Argentina in 2020 at six different locations. No significant differences were found in the ATIs levels between HB4 wheat and its isolines. ATIs levels in HB4 wheat were then analyzed within the natural variation given by the varieties and locations included in this study. Altogether, this study confirmed that ATIs levels, as previously reported for other allergens, are mainly affected by the genetic background and the environmental conditions, and that the ATIs levels measured in HB4 transgenic wheat are within the range of natural variability observed in the non-transgenic counterparts.

The online version contains supplementary material available at 10.1007/s11248-026-00487-8.

Figures

Amylase Trypsin Inhibitors (ATIs) levels in HB4 wheat and its non-transgenic isolines. The concentration of several ATIs (0.28, 0.19 + 0.53, CM2, CM3, CM16, and CM17) was measured by LC–MS in grain samples from wheat IND-ØØ412-7 (Alg-HB4 and Bas-HB4) and its non-transgenic isolines (Algarrobo and Basilio, respectively) in two genetic backgrounds: Algarrobo (a) and Basilio (b). Numbers represent the mean of 18 values measured in samples from field trials run during 2020 in six different locations (three replicates). Thin bars indicate the standard deviation of the mean

Sum of ATIs. The level of the seven ATIs measured in wheat event IND-ØØ412-7 introgressed into two different genetic backgrounds (Algarrobo: Alg-HB4, and Basilio: Bas-HB4) were combined. The results obtained for the respective conventional isolines (Algarrobo and Basilio) are indicated as a control. Bars represent the mean of 18 values, and the error bars indicate the standard deviation of the mean

Environmental variability of the sum of ATIs. The cumulative level of all the ATIs measured (y-axis) at different locations is depicted by genetic background (x-axis). Each dot represents an average of three replicated plots

Full text

Introduction

Wheat-based foods provide a significant proportion of the daily intake of dietary calories and substantially contribute protein, fiber, vitamins, and minerals to a wholesome diet (Weegels 2019). It has been proposed that bread, especially whole grain bread, would be the staple food of choice to face future food supply challenges, resulting from a growing global population, which is becoming more urbanized and sedentary (Weegels 2019). Nevertheless, some wheat proteins can provoke health-related conditions affecting the life quality of some subpopulations. The celiac disease associated with gluten proteins is the best-known affection triggered by wheat. The proteins and epitopes associated with clinical symptoms and the genetic diversity of the gliadin-coding genes, and the sensitive subjects have been characterized. In addition to gluten proteins, other components of wheat flour have been proposed as allergens (Tatham and Shewry 2008). Among them, the α-amylase/trypsin inhibitors (ATIs) have been pointed out as being responsible for triggering baker’s asthma (Sanchez-Monge et al. 1992). This is a respiratory allergy caused by inhalation of wheat flour and constitutes one of the most common types of occupational asthma, with an average prevalence of up to 8% among bakers (Zevallos et al. 2017). ATIs have also been recently reported to contribute to the etiology of celiac disease as well as non-celiac wheat sensitivity (Geisslitz et al. 2022).

ATIs are found in the endosperm of plant seeds, where they account for up to 4% of the total protein in grain, and work as storage proteins as well as natural defense agents against different pests (Altenbach et al. 2011; Geisslitz et al. 2022). The currently available wheat genome (Zhu et al. 2021) revealed the presence of 35 ATI-related genes (Juhász et al. 2018). How these genes collaborate with the final allergenic potential associated with specific ATI proteins is still a work in progress. ATIs are a diverse group of similar low molecular weight proteins appearing in various isoforms: monomeric like 0.28 (named based on electrophoretic mobility), homodimeric (0.19 and 0.53), and chloroform–methanol (CM) soluble heterotetrameric isoforms (Geisslitz et al. 2021).

During the last few years, several publications analyzed ATIs levels mainly associated with different Triticum species (Geisslitz et al. 2018, 2020; Call et al. 2020) and wheat varieties (Prandi et al. 2013; Geisslitz et al. 2018; Bose et al. 2020; El Hassouni et al. 2021). In this study, the ATIs levels in two wheat cultivars carrying the stress-tolerant transgenic event IND-ØØ412-7 (HB4 wheat) were compared with those found in their conventional isolines. Wheat event IND-ØØ412-7 is the first transgenic wheat that has been commercially approved in several countries. It carries the HB4^®^ technology, based on the expression of the sunflower HaHB4 gene, and provides tolerance to environmental stresses including drought (González et al. 2019). The comparative analysis described in this study allows us to explore the potential putative differences in the level of endogenous allergens that might have been raised as unintended effects associated with genetic modification. Unintended changes that genetic modification might produce are one of the columns of risk assessment. Within this context, measuring the level of endogenous allergens might be part of the regulatory requests. A previous study has already confirmed that the gliadin content of HB4 wheat was similar to non-genetically modified wheat, according to the compositional equivalence reported for this event (Ayala et al. 2019).

Considering the diversity of genes and proteins associated with the different ATIs, the relationship of specific isoforms with allergenicity is not easy to determine. Consequently, ATIs tagged as predominant, based on their contribution of more than 5% to the total levels in wheat grain, were considered for this study (Geisslitz et al. 2020), to examine if wheat event IND-ØØ412-7 displays an altered content of these allergens.

Materials and methods

Samples

Wheat event IND-ØØ412-7 was originally generated by the genetic transformation of the winter cultivar Cadenza (SASA 2017). As part of the breeding process, the event was introgressed into different wheat commercial varieties using conventional backcross, marker-assisted selection (MAS), and field phenotypic selection, to recover the event and the background of the commercial varieties. Two cycles of backcrosses, including one MAS and five cycles of field phenotypic selections, were performed to generate the HB4 lines. The wheat genotypes used in this study include two IND-ØØ412-7 lines introgressed in the facultative cultivars Algarrobo and Basilio, and their non-genetically modified parental varieties (isolines) that were used as controls. In addition, other five commercial varieties (Aca 360, Alhambra, Baguette 620, Buck Destello, and Cedro) were included to provide a range of natural variability. The two HB4 lines have been registered for commercialization as Iruya HB4 and Traful HB4, respectively, in the Argentina cultivars catalog and were launched in 2024 (INASE 2024). Field trials were carried out in the 2020 season at six locations covering the different environments for wheat production in Argentina: Arias and Monte Buey, both in Cordoba Province; Pergamino, Tandil, and Tres Arroyos, located in Buenos Aires Province; and Victoria, in Entre Rios Province (see a map in Supplementary information, Fig. S1). Favorable fields for wheat production with predominant soils classified as mollisols (Eswaran and Reich 2005) were selected for field trials (Table S1 in Supplementary information). A randomized complete block design with three replications was used. The experimental unit consisted in 7 m2 plots that were fertilized with 100 kg/ha of ammonium phosphate (%NPK: 18-20-0) at sowing and with 250 kg/ha of urea at tillering. Diseases were preventively controlled at jointing or flag leaf growth stage with wide spectrum fungicides based on strobilurins and triazoles. Weeds were controlled before sowing with herbicides based on glyphosate, sulfonylureas, and 2,4-D or Dicamba. All the chemical inputs were used at the recommended doses for wheat, approved by the Argentinean National Service of Agri-Food Health and Quality (SENASA). At maturity, whole plots were harvested, and a sample of kernels was kept at 4°C until tested. Aliquots from each sample were shipped to Eurofins Food Chemistry Testing Madison, Inc. (Madison, Wisconsin) for analysis.

Samples processing

Grain samples were milled to flour and kept at − 70 °C until use. Proteins were extracted with ammonium bicarbonate (Junker et al. 2012). This procedure was selected as the most similar used in the ATIs quantification procedure previously described (Geisslitz et al. 2018). It removes trypsin inhibitory activity and proved to be more efficient for detecting some of the target peptides. Aliquots of wheat flour (50 mg) were extracted twice with 0.5 mL of 50 mM ammonium bicarbonate and vortexed for one min at RT, centrifuged at 16,100 rcf for five min at 5 ͦC. The supernatants from these two sequential extractions were placed in the same well of a 96-well plate and evaporated under nitrogen at 50 ͦC. Dried extracts were resuspended with 320 µl of 0.5 M Tris–HCl and the same volume of 1-propanol, vortex for one minute, sonicated for 5 min, and vortexed again for 5 min. Samples were reduced by adding 50 µl 0.05 M of Tris(2-Carboxyethyl)phosphine in 0.5 M Tris–HCl and incubating 30 min at 60 ͦC in a water bath. Alkylation solution (100 µl of 0.5 M Chloroacetamide in 0.5 M Tris–HCl) was added, and incubation continued for 45 min at 37 ͦC in the dark. After vortexing, aliquots (80 µl) were placed in new plates and evaporated under nitrogen at 40 ͦC. Samples were reconstituted with 0.5 mL of 40 mM urea, 100 mM Tris, pH 7.8, vortexed, sonicated, and supplemented with heavy peptides solution (20 µg/ml each). Digestion started by adding 50 µl of trypsin solution (475 µg/µl) and incubated for 19–21 h at 37 ͦC in the dark. After evaporation under nitrogen at 50 ͦC, samples were resuspended in 0.1% formic acid and used for ATIs quantification.

ATIs quantification

The quantification of the different ATIs was set up by Eurofins (USA, Madison, Wi, USA) based on procedures previously described (Geisslitz et al. 2018; Bose et al. 2020). A High-Performance Liquid Chromatography-Mass Spectrometry (HPLC–MS) method was validated for the analyses of the primary protein targets 0.19, 0.28, 0.53, CM2, CM3, CM16, and CM17. Analytical procedures were performed under GLP compliance and with laboratory-validated standard operational procedures. Briefly, proteotypic peptides useful to unequivocally identify the proteins under study were synthesized using labeled heavy isotopes (JPT Peptide Technologies GmbH, Germany). These “heavy” peptides (Table S2, Supplementary Information) were used to set up the conditions for isolation and detection by LC–MS. These peptides were then used to correlate endogenous peptide levels in extracts produced during tryptic digestion. A pooled sample of wheat flour containing the different conventional wheat varieties included in the study was used to set extraction and digestion conditions adequate to detect the endogenous (light) peptides.

LC–MS

An ABSciex 6500 coupled to a triple quadrupole mass spectrometer (SCIEX, Framingham, MA) equipped with an Acquity BEH C18 column (50 × 2.1 × 1.7 µM, Waters, Framingham, MA) was used for the study. Chromatography was carried out using 0.1% formic acid in water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B) and a gradient: 0–3 min 5% B, 3–20 min 5–30% B, 20–21 min 30% B, 21–22 min 30–90% B, 22–25 min 90% B. The amount of each ATI was calculated based on the isotope dilution method (Kippen et al. 1997) and is expressed as µg/g of dry weight. The results obtained for each ATI were used to calculate a value (sum of ATIs) representative of the total ATIs content since the set analyzed in this study have been reported to represent 84% of the total (Geisslitz et al. 2020).

Statistical analysis

ATI data from randomized complete block design field trials conducted at six sites were analyzed using InfoStat v.2020 (Di Rienzo et al. 2020). For each distribution of ATI values, observations with \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left|{x}_{i}-\overline{x }\right|>4s$$\end{document} (Eq. 1) were considered outliers and seven data points from Tres Arroyos were excluded from the analysis. Specifically, for Cedro replicate 2 in CM16 and replicate 3 in both CM16 and CM17 were removed; and for ACA 360 replicates 1 and 3 in both CM16 and CM17 were removed.

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ s = \sqrt {\frac{1}{n - 1} \mathop \sum \limits_{x = 1}^{n} \left( {x_{i} - \overline{x}} \right)^{2} } $$\end{document}

A combined site analysis was performed according to the model described by the Eq. (2):

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ y_{ijk} = \mu + \alpha_{i} + \beta_{j\left( i \right)} + { }{ gamma }_{{\mathrm{k}}} + \left( {\alpha \times { gamma }} \right)_{ik} + { }{ varepsilon }_{ijk} $$\end{document}

where μ is the grand mean; α_i_ is the effect of the ith environment; β_j(i)_ is the effect of the jth block nested within the environment, γ_k_ is the effect of the kth cultivar, (α × γ)ik is the term for the interaction effect between the cultivar and the environment and ε_ijk_ is the error. Individual site analyses were also performed according to Eq. (3):

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ y_{jk} = \mu + \beta_{j } + { }{ gamma }_{{\mathrm{k}}} + { }{ varepsilon }_{jk} $$\end{document}

where the individual values y_jk_ are estimated based on fixed effects of blocks (β_j_) cultivars (γ_k_) and a random error effect (ε_jk_).

Assumptions of normality were assessed using the Shapiro–Wilk test (Balzarini et al. 2008). For variables that were not normally distributed (0.28, CM16 and CM17), values were transformed with the Box-Cox method (Daimon 2011) (Eq. 4; \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda =0.3$$\end{document} )

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ y_{ } = \frac{{x^{\left( \lambda \right)} - 1}}{\lambda } $$\end{document}

The homoscedasticity assumption for ATI variables was confirmed by Levene’s test, according to the methodology described in Balzarini et al. (2008). R Core Team (2020) comparisons between IND-ØØ412-7 and its isolines were performed using Tukey's Honest Significant Difference (HSD) test as a post-hoc analysis, with a 95% confidence level. All the cultivars and experimental lines were included in the analysis. However, because the aim of this study was to evaluate whether the genetic modification affected the ATI content of wheat, the five commercial references were excluded from direct comparisons. Instead they were considered together with the non-transgenic isolines to stablish a reference range for natural variability. Results are reported as the mean ± standard error, expressed in µg/g of dry weight.

Results

Analysis of different ATIs

The insertion event that characterized HB4 wheat (IND-ØØ412-7), originally developed by the transformation of the Cadenza cultivar, was introduced into two commercial genetic backgrounds (Algarrobo and Basilio) by conventional breeding techniques. ATIs contents were measured in grain from these two genetic backgrounds (with and without the event) and in five additional commercial varieties grown in six locations within the Argentina wheat cultivated area.Fig. 1. Amylase Trypsin Inhibitors (ATIs) levels in HB4 wheat and its non-transgenic isolines. The concentration of several ATIs (0.28, 0.19 + 0.53, CM2, CM3, CM16, and CM17) was measured by LC–MS in grain samples from wheat IND-ØØ412-7 (Alg-HB4 and Bas-HB4) and its non-transgenic isolines (Algarrobo and Basilio, respectively) in two genetic backgrounds: Algarrobo (a) and Basilio (b). Numbers represent the mean of 18 values measured in samples from field trials run during 2020 in six different locations (three replicates). Thin bars indicate the standard deviation of the mean

In the combined site analysis with all the commercial references, the HB4 lines and their isolines showed no significant differences in any ATI (Fig. 1). However, all the ATIs showed a significant genotype-by-site interaction (Table S3) indicating the HB4 lines and their isolines showed no significant differences in any ATI (Fig. 1). However, all the ATIs showed a significant genotype-by-site interaction (Table S3) indicating that genotype performance differed across sites.

Site-by-site analysis detected no significant differences between HB4 lines and their respective isolines for any ATI at any location, in either Basilio or Algarrobo backgrounds (Tables S4 to S9). Thus, the significant interaction in the combined analysis resulted from the commercial references variability, while the HB4 lines and their conventional isolines showed similar responses across the sites.

Sum of ATIs

The results obtained for each ATI were added to obtain the sum of ATIs, an overall value representative of the content of total ATIs. The combined site analysis showed no statistical differences between the lines carrying the event IND-ØØ412-7 and their respective conventional isolines (see Fig. 2).Fig. 2. Sum of ATIs. The level of the seven ATIs measured in wheat event IND-ØØ412-7 introgressed into two different genetic backgrounds (Algarrobo: Alg-HB4, and Basilio: Bas-HB4) were combined. The results obtained for the respective conventional isolines (Algarrobo and Basilio) are indicated as a control. Bars represent the mean of 18 values, and the error bars indicate the standard deviation of the mean

Variability in ATI contents across environments and genetic backgrounds

To address these results within the context of natural variability, data dispersion by location and genotype was examined. The value of the sum of ATIs for each genotype at different locations is shown in Fig. 3.Fig. 3. Environmental variability of the sum of ATIs. The cumulative level of all the ATIs measured (y-axis) at different locations is depicted by genetic background (x-axis). Each dot represents an average of three replicated plots

The results indicate that the sum of ATIs can vary up to 4.1-fold when the same variety is cultivated at different locations (see Cedro levels in Fig. 3). In addition, this analysis helps to show that the overall content of ATIs measured in HB4 wheat is not only lower than those displayed by commercial varieties across different locations but also seems to narrow the among-site variability. The coefficient of variation for the sum of ATIs (calculated as the ratio between the standard deviation and the mean) was higher in the conventional varieties than in their respective HB4 isolines (42% vs. 27% for Algarrobo and 33% vs. 17% for Basilio). These values indicate that HB4 lines showed less variation for the sum ATIs across sites than their conventional counterparts.

When the variability among different genetic backgrounds is examined, the sum of ATIs can differ up to 2.5-fold within the same location among the different wheat varieties included in this study. The highest variation was observed in Tres Arroyos (Fig. 3).

Discussion

ATIs are a family of grain storage proteins associated with baker’s asthma (Sanchez-Monge et al. 1992). During the last few years, ATIs levels have been measured in different Triticum species (Geisslitz et al. 2018, 2020; Call et al. 2020; Simonetti et al. 2022), and different varieties of T. aestivum (Prandi et al. 2013; Geisslitz et al. 2018; Bose et al. 2020; Sielaff et al. 2021; El Hassouni et al. 2021). Some of these studies also considered the effect of environment (Prandi et al. 2013; Geisslitz et al. 2020; El Hassouni et al. 2021; Simonetti et al. 2022)and genetic background on ATIs content (Prandi et al. 2013; Geisslitz et al. 2018; Bose et al. 2020; Sielaff et al. 2021). In this study, we aimed to analyze if ATI levels may have been altered as an unintended effect derived from the genetic modification associated with HB4 wheat.

The levels of seven ATIs were measured in grains obtained from field trials carried out in six locations covering a wide range of environments within the Argentina wheat cultivation area. Entries included the transgenic event in two genetic backgrounds, their conventional isolines, and five additional commercial varieties. The ATI levels measured in this study are in accordance with those previously published (Geisslitz et al. 2018, 2020). Overall, our results revealed that the content of the different ATIs in HB4 wheat lines is comparable to those in conventional wheat varieties. Similar results were found for every one of the wheat components tested in the original transgenic event obtained in the Cadenza variety (Ayala et al. 2019).

The combined-sites analysis did not reveal differences in the content for any of the ATIs evaluated (Fig. 1). The analysis by site did not reveal any significant difference in ATI levels for the transgenic event in any of the environments. In all cases, the ATI levels measured in wheat event IND-ØØ412-7 fall within or even below the reference range provided by the commercial varieties grown in parallel (Fig. 3 and Tables S4–S9). Moreover, when the effect of other variables was analyzed, the interaction between the genetic background and the environment revealed their major role in the expression of ATIs. The variability in ATIs level associated with different locations and genotypes far exceeds any difference observed between the transgenic event and its isoline. This finding is consistent with previous studies analyzing the effect of transgenesis on several wheat allergens (Lupi et al. 2013, 2014, 2015).

In conclusion, the results of this study indicate that the level of ATIs, a protein family of allergens associated with baker´s asthma, in HB4 wheat is similar to that found in conventional wheat.

Plenty of evidence supports how genetic background and environmental conditions affect grain composition (Stevenson et al. 2012; Natarajan et al. 2016). As an example, soybean allergen levels have been reported to increase up to 19-fold due to environmental conditions (Geng et al. 2017). Specifically in wheat and for ATIs, previous reports support the changes associated with growing conditions and genotype (Prandi et al. 2013; Geisslitz et al. 2018, 2020; Bose et al. 2020). A wide study on 149 European cultivars stated an up to sixfold variation in some of the ATIs (El Hassouni et al. 2021). In our study, the reference range limits for the sum of ATIs reach up to 4.1 and 2.5-fold among locations and commercial varieties, respectively (Fig. 3). These differences are even higher when specific ATIs are considered (up to 18-fold for CM17 among sites for Cedro cultivar: 67–1202 µg/g). Moreover, being defense-associated proteins, stress-associated increases up to threefold in some ATIs levels have been reported (Hajheidari et al. 2007). However, even when the genetic modification introduced into HB4 wheat provides protection against environmental stresses, it did not modify the ATI levels. Moreover, it seems to reduce ATI levels variability among environments (Fig. 3).

Baker´s asthma has been associated with several proteins belonging to different families. In addition, each family is composed of several polypeptides. When adding the multiplicity of factors involved in the immune response and the variability of reactions among individuals, the association of allergies to a specific chemical component is not straightforward. In any case, no difference was found either in a single ATI or in the sum of ATIs between HB4 wheat and its conventional isolines.

In summary, the results presented in this study indicate that ATI levels in wheat event IND-ØØ412-7 are similar to those in conventional wheat. Considering that ATIs are the best-studied allergens associated with baker´s asthma, the evidence presented in this study indicates that the contribution of HB4 to this affection would be equivalent to that of conventional wheat varieties already present in the market.

Supplementary Information

Below is the link to the electronic supplementary material.Supplementary file 1 (PDF 626 kb)