Implications of Crop Rotation and Fungicide on Fusarium and Mycotoxin Spectra in Manitoba Barley, 2017–2019

Fusarium head blight (FHB) is one of the most important diseases of barley in Manitoba province (western Canada), and other major barley producing regions of the world. Little is known about the Fusarium species and mycotoxin spectra associated with FHB of barley in Manitoba. Hence, barley grain samples were collected from 149 commercial fields from 2017 to 2019, along with information on respective cropping history, and analyzed with respect to Fusarium species spectra, abundance, chemotype composition, and mycotoxin profiles. Fusarium poae was the predominant Fusarium species associated with FHB of barley in Manitoba, followed by F. graminearum, and F. sporotrichioides; F. equiseti and F. avenaceum were also detected but at low levels. F. poae strains with the nivalenol (NIV) chemotype and F. graminearum strains with 3-acetyl deoxynivalenol (3-ADON) and 15-acetyl deoxynivalenol (15-ADON) chemotypes were commonly detected in the barley grain samples. Nivalenol (597.7, 219.1, and 412.4 µg kg−1) and deoxynivalenol (DON) (264.7, 56.7, and 65.3 µg kg−1) were the two most prevalent mycotoxins contaminating Manitoba barley in 2017, 2018 and 2019, respectively. A substantially higher DON content was detected in grain samples from barley fields with cereals as a preceding crop compared to canola and flax. Furthermore, F. poae proved less sensitive to four triazole fungicides (metconazole, prothioconazole+tebuconazole, tebuconazole, and prothioconazole) than F. graminearum. Findings from this research will assist barley producers with improved understanding of FHB threat levels and optimizing practices for the best management of FHB in barley.


Introduction
Barley (Hordeum vulgare L.) is one of the world's major cereal crops and known for many health attributes, including being high in soluble fibre, antioxidants, essential vitamins, and minerals that contribute to lowering cholesterol [1]. Globally, barley is harvested at approximately 49.5 million hectares and produces approximately 143 million metric tonnes annually (https://www.fao.org/faostat/en/#data/QCL/visualize, accessed on 6 June 2022). Canada is a major exporter of barley, contributing~9% of global production. Canadian breweries and livestock operations that rely on barley export over $2 billion in goods to the United States, Japan, South Korea, and Mexico (https://cafta.org/export/ barley/, accessed on 6 June 2022). Over 90% of Canadian barley is produced in the western Canadian provinces of Manitoba, Saskatchewan, and Alberta.
To better understand the FHB species spectra and associated mycotoxins affecting Manitoba barley, analysis was carried out on barley grain samples collected from producers' fields in three consecutive years (2017, 2018, and 2019), from five crop districts of Manitoba. This study investigates four main objectives: (i) Fusarium species spectra and chemotypes associated with FHB of barley; (ii) profile of naturally occurring Fusarium mycotoxins in barley grains and their main fungal source; (iii) impact of commonly used fungicides on the mycelial growth of F. poae and F. graminearum, and (iv) influence of crop rotation on DON and NIV levels in barley grains.

Fusarium Species Spectra and Chemotype Composition
PCR-based assays were used to investigate Fusarium species spectra associated with FHB of barley in Manitoba ( Figure 1A). A total of five Fusarium species were detected from 2017 to 2019. F. poae was the most common, detected in 66%, 62%, and 55% of barley samples in 2017, 2018 and 2019, respectively. F. graminearum was the next most common, detected in 40 to 57% of barley samples from 2017 to 2019. F. sporotrichioides was detected in 43% of barley samples in 2017, but at a much lower rate of 7% and 9% of samples in 2018 and 2019, respectively. Infections caused by F. avenaceum and F. equiseti were also found but at much lower frequencies compared to those attributed to F. poae, F. graminearum, and F. sporotrichioides.
Toxins 2022, 14, x FOR PEER REVIEW 3 of 15 To better understand the FHB species spectra and associated mycotoxins affecting Manitoba barley, analysis was carried out on barley grain samples collected from producers' fields in three consecutive years (2017, 2018, and 2019), from five crop districts of Manitoba. This study investigates four main objectives: (i) Fusarium species spectra and chemotypes associated with FHB of barley; (ii) profile of naturally occurring Fusarium mycotoxins in barley grains and their main fungal source; (iii) impact of commonly used fungicides on the mycelial growth of F. poae and F. graminearum, and (iv) influence of crop rotation on DON and NIV levels in barley grains.

Fusarium Species Spectra and Chemotype Composition
PCR-based assays were used to investigate Fusarium species spectra associated with FHB of barley in Manitoba ( Figure 1A). A total of five Fusarium species were detected from 2017 to 2019. F. poae was the most common, detected in 66, 62, and 55% of barley samples in 2017, 2018 and 2019, respectively. F. graminearum was the next most common, detected in 40 to 57% of barley samples from 2017 to 2019. F. sporotrichioides was detected in 43% of barley samples in 2017, but at a much lower rate of 7% and 9% of samples in 2018 and 2019, respectively. Infections caused by F. avenaceum and F. equiseti were also found but at much lower frequencies compared to those attributed to F. poae, F. graminearum, and F. sporotrichioides. The chemotype composition of Fusarium pathogens associated with FHB of barley in Manitoba was also evaluated ( Figure 1B). The percentage of barley samples infected with F. graminearum 3-ADON strains was slightly higher (65, 50, and 45% of samples) than for 15-ADON strains (58,38, and 40% of samples) from 2017 to 2019, respectively. Fusarium graminearum strains with the NIV chemotype were not detected. In comparison, F. poae strains with the NIV chemotype were detected in 100% of barley samples from 2017 and 2019 and in 82% of samples from 2018.

Abundance of Fusarium DNA
Quantitative PCR was used to analyse the abundance of  The chemotype composition of Fusarium pathogens associated with FHB of barley in Manitoba was also evaluated ( Figure 1B). The percentage of barley samples infected with F. graminearum 3-ADON strains was slightly higher (65%, 50%, and 45% of samples) than for 15-ADON strains (58%, 38%, and 40% of samples) from 2017 to 2019, respectively. Fusarium graminearum strains with the NIV chemotype were not detected. In comparison, F. poae strains with the NIV chemotype were detected in 100% of barley samples from 2017 and 2019 and in 82% of samples from 2018.

Abundance of Fusarium DNA
Quantitative PCR was used to analyse the abundance of 0.16 pg Fusarium gDNA ng −1 total gDNA, significantly higher than mean concentrations of F. graminearum gDNA at 0.20 and 0.06 pg Fusarium gDNA ng −1 total gDNA, respectively.
In 2019 barley samples, the mean concentrations of F. poae and F. graminearum g were similar at 0.08 and 0.09 pg Fusarium gDNA ng −1 total gDNA, respectively. The a dance of F. sporotrichioides gDNA in barley grains was much lower and varied from to 0.01 pg Fusarium gDNA ng −1 total gDNA from 2017 to 2019. Due to low Ct values (b 39) in most samples, data for F. avenaceum and F. equiseti were excluded from the ana

Fusarium Mycotoxins
Naturally occurring mycotoxins in barley grains were analysed using UH HRMS ( Table 1)  In 2019 barley samples, the mean concentrations of F. poae and F. graminearum gDNA were similar at 0.08 and 0.09 pg Fusarium gDNA ng −1 total gDNA, respectively. The abundance of F. sporotrichioides gDNA in barley grains was much lower and varied from 0.003 to 0.01 pg Fusarium gDNA ng −1 total gDNA from 2017 to 2019. Due to low Ct values (below 39) in most samples, data for F. avenaceum and F. equiseti were excluded from the analysis.

Fusarium Mycotoxins
Naturally occurring mycotoxins in barley grains were analysed using UHPLC-HRMS ( Table 1)

PCA of the Concentrations of Fusarium gDNA and Mycotoxins in Barley Grains
The results from the quantification of Fusarium gDNA and mycotoxins in barley grain samples were subjected to PCA ( Figure 3). The first two dimensions account for 45.74% of the variability: component one on the x-axis describes 25.11% of the variability and component two on the y-axis represents an additional 20.63%. A high correlation was found between the concentrations of F. graminearum gDNA and DON/CULM as they clustered in the same quadrant. Concentrations of F. poae gDNA and NIV/BEA /DAS also clustered in the same quadrant. Furthermore, F. sporotrichioides gDNA and T-2/HT-2 concentrations also showed a high correlation. No clear correlation was established between any of the Fusarium species and ZEA/ENNs (Figure 3). Notes: Limit of quantitation (LOQ) and limit of detection (LOD) were 2.5 and 1 µg kg −1 , respectively. For calculation of the mean, mycotoxin concentrations below the respective LOQ or LOD were calculated as LOQ/2 or LOD/2, respectively. ENNs is the sum of ENN A, A1, B, and B1.

PCA of the Concentrations of Fusarium gDNA and Mycotoxins in Barley Grains
The results from the quantification of Fusarium gDNA and mycotoxins in barley grain samples were subjected to PCA (Figure 3). The first two dimensions account for 45.74% of the variability: component one on the x-axis describes 25.11% of the variability and component two on the y-axis represents an additional 20.63%. A high correlation was found between the concentrations of F. graminearum gDNA and DON/CULM as they clustered in the same quadrant. Concentrations of F. poae gDNA and NIV/BEA /DAS also clustered in the same quadrant. Furthermore, F. sporotrichioides gDNA and T-2/HT-2 concentrations also showed a high correlation. No clear correlation was established between any of the Fusarium species and ZEA/ENNs (Figure 3).

Sensitivity of F. poae and F. graminearum Strains to Triazole Fungicides
The sensitivity of F. poae and F. graminearum to metconazole, prothioconazole+tebuconazole, tebuconazole, and prothioconazole was analysed using PDA plate-based assays supplemented with two fungicide concentrations (0.01 and 0.1 mg of active ingredient L −1 ) ( Table 2). Compared to Proline and Folicur, Caramba and Prosaro displayed a higher inhibitory effect against the mycelial growth of F. poae and F. graminearum. Interestingly, F. poae strains showed lower sensitivity to all four fungicides than F. graminearum strains at both concentrations, except for Caramba at 0.1 mg active ingredient L −1 and Folicur at 0.01 active ingredient L −1 ( Table 2).
Note: Mean ± SE refers to combined average radial growth of 15 F. poae and 15 F. graminearum selected strains. Means of % inhibition followed by the same letter within the same fungicide are not significantly different at p = 0.05 (*) and p = 0.01 (**) based on Student's t-test.

Effects of Preceding Crop and Year on Fusarium Mycotoxins in Barley Grains
The effect of preceding crop and year on DON and NIV concentrations in barley samples was investigated using the GLIMMIX procedure (Table 3)

Discussion
The current study investigated populations of Fusarium pathogens and mycotoxin profiles in grains from naturally infected commercial barley fields from five crop districts in Manitoba. The results indicate that the Fusarium species composition associated with FHB of barley in Manitoba is diverse and dominated by several toxigenic Fusarium species. F. poae was the predominant species, followed by F. graminearum and F. sporotrichioides. This pattern agrees with the results of a 13-year barley FHB survey (2005 to 2017) in eastern Canada, which showed F. poae and F. graminearum were the predominant Fusarium pathogens in barley at that time, representing 42% and 40%, respectively, of the total Fusarium population in the region [29]. F. poae has also been reported as the predominant Fusarium pathogen associated with FHB of barley in several South American and European countries [22,30].
The Fusarium species spectra associated with FHB of barley appear be different than that affecting wheat. Fusarium graminearum is typically considered the most prevalent and aggressive FHB pathogen of wheat in the temperate and warmer regions of the United States and Canada [31,32], while in cooler regions of Europe F. graminearum, F. culmorum and F. avenaceum are predominant [33]. Therefore, research on Fusarium pathogens associated with FHB of barley in these regions has primarily focused on F. graminearum. Our findings, together with others, show Fusarium species other than F. graminearum, such as F. poae and F. sporotrichioides, play a significant role in the FHB disease complex on barley. This result is of particular importance and suggests more focus should be directed towards understanding the impact of Fusarium species previously considered less aggressive but still economically important due to their association with FHB and mycotoxin accumulation and subsequent cumulative negative impacts on the grain yield and quality.
Compared to other Fusarium species involved in FHB, F. poae causes the least reduction in seed germination [34] and its infection on the cereal head is mainly asymptomatic [35]. Therefore, F. poae had been considered a weak pathogen on cereals. However, recent studies show F. poae has become omnipresent in Europe, South America, and North America [36][37][38], which raises concerns about the importance of this pathogen in the FHB disease complex.
Climatic conditions, specifically temperature and humidity, can significantly impact the distribution and predominance of Fusarium species. An extended period of wet conditions during plant anthesis is well documented as crucial for the infection of F. graminearum [39]. On the contrary, humidity duration has no significant effect on the infection of F. poae and its mycotoxin production [40]. The optimal temperature of F. poae growth is between 25 and 30 • C, similar to that of F. sporotrichioides but slightly higher than F. graminearum (20-25 • C) [41]. F. poae may proliferate well in hot and dry conditions rather than the wet and cool conditions preferred by F. graminearum. Manitoba experienced relatively warm and dry conditions during growing seasons and barley anthesis between 2017 and 2019. The predominance of F. poae over other Fusarium species, especially F. graminearum, can be partially explained by the occurrence of thermo-hygrometric conditions that do not favour F. graminearum, allowing for a shift in the population dynamics towards F. poae.
Concentrations of NIV and F. poae gDNA were highly correlated in barley grain samples from Manitoba. These results agree with the findings of Yli-Mattila et al. [42], where a similar high correlation in barley samples from Finland and Russia was observed. Islam et al. [37] reported correlations between NIV content and the concentration of F. poae DNA in oat samples from Manitoba. These results suggest F. poae is a potent NIV producer in barley. In addition, BEA and DAS are also highly correlated with F. poae infection. However, these two mycotoxins are only produced at low levels.
In this study, 16.7% to 29.5% of barley grain samples were contaminated with NIV at concentrations higher than 500 µg kg −1 . The maximum concentration of NIV was 4136.6, 3252.1, and 2245.4 µg kg −1 in 2017, 2018, and 2019, respectively. These values are in agreement with the results of Tucker et al. [43], who detected a high level of NIV contamination in grain samples from various mock-inoculated (water) barley genotypes in an FHB screening disease nursery using F. graminearum as the inoculum in Brandon, Manitoba. The extent of NIV contamination found in barley in this study is higher than that reported in the oat crop. Specifically, a survey of Fusarium mycotoxins in oat grains from Manitoba, collected between 2016 to 2018, found maximum concentrations of NIV ranged between 581 to 865 µg kg −1 [37]. To date, NIV contamination in wheat has been rarely reported in Western Canada and is mainly due to the predominance of F. graminearum on wheat and the absence of F. graminearum strains with the NIV chemotype in the region [31].
The significant level of NIV contamination in naturally infected barley grain is a cause of concern for food safety and security. NIV is more toxic to animals than DON due to the presence of a hydroxyl group, instead of hydrogen, at the C-4 position. The presence of a hydroxyl group in NIV increases its toxicity to animals by ten-fold compared to DON [6]. The European Food Safety Administration established a lower tolerated daily intake of 0.7 µg kg −1 body weight for NIV compared to 1 µg kg −1 for DON. Tolerance limits for NIV in food cereals have not been identified yet in Canada [44] and currently, NIV concentrations in barley grains are neither routinely determined nor legislatively regulated in Canada [19]. The discovery of NIV contamination associated with F. poae in barley shows the importance of routine testing for NIV in naturally affected barley grain samples and the need for more extensive research on F. poae. These results also have practical implications, such as the  [2,4,45].
In this study, F. graminearum and DON were detected at lower levels than F. poae and NIV in barley grains. These observations were expected and align with the prevailing weather conditions in Manitoba i.e., relatively dry conditions in June and July of 2017, 2018, and 2019 that are not very conducive to infection by F. graminearum resulting in low DON concentrations. However, the maximum concentrations of DON in barley samples reached 8921.6, 1488.8, and 2051.1 µg kg −1 in these three respective years. This indicates F. graminearum can still pose a significant threat to growers when regional microenvironment conditions favour it as a causal agent.
Both F. graminearum 3-ADON and 15-ADON chemotype strains were detected in barley samples from Manitoba, with 3-ADON strains being detected at a higher frequency. Similarly, a higher percentage of F. graminearum strains with the 3-ADON chemotype were found in wheat samples from Western Canada [46,47]. Before 1994, F. graminearum strains with the 15-ADON chemotype were the principal pathogen causing FHB in wheat in North America. However, a shift to F. graminearum strains with the 3-ADON chemotype was documented between 1998 and 2004, attributed to changes in environmental conditions, host distribution, and various agricultural practices [46]. F. graminearum strains with the NIV chemotype were not detected in this study. This result agrees with previous studies on the F. graminearum population in Western Canada that suggest the lack of this group of F. graminearum strains in the region [31,37].  [50]. In a field study at an experimental farm in southern Manitoba, with barley varieties inoculated with F. sporotrichioides, HT-2/T-2 concentrations reached 670 to 3720 µg kg −1 [4]. These results suggest a significant level of HT-2/T-2 contamination could occur when environmental conditions are conducive to F. sporotrichioides. Due to barley being a major component of many feed formulations, livestock toxicity of HT-2/T-2 might arise in years with high F. sporotrichioides infection. Therefore, feed mill operators should consider chemical assays to ensure their feed components contain minimal amounts of HT-2/T-2.
To evaluate the potential effect of agrochemicals on the Fusarium population, we analysed the sensitivity of F. poae and F. graminearum to four triazole fungicides commonly used in Manitoba to manage FHB. The results showed F. poae has lower sensitivity to these fungicides than F. graminearum. Similarly, Tini et al. [51] show F. poae strains from Italy are more resistant to fungicides containing a similar group of active ingredients than F. graminearum and F. culmorum. In addition, Audenaert et al. [52] report the treatment of wheat fields with azole fungicides caused the Fusarium population to shift from one dominated by F. culmorum and F. graminearum to one dominated by F. poae. These results suggest the pressure exerted by the repeated use of fungicides may impact Fusarium population dynamics, leading to the shift in the pathogen population towards those more resistant to these fungicides.
In this study, cereal-barley rotations resulted in higher DON levels in barley grains than canola-barley or flax-barley rotations. This result agrees with findings by Schöneberg et al. [53], in which the cultivation of corn before barley caused a higher F. graminearum incidence and DON content than a canola-barley rotation. Pageau et al. [54] also report the DON content in barley grains was higher when cereals, rather than dry pea, were seeded in the rotation. Interestingly, no significant impact of previous crops on NIV concentration in barley was observed. Similarly, Vogelgsang et al. [38] show cropping history had no impact on the incidence of F. poae and NIV content in wheat. Differences in the effects of the previous crop on the concentration of DON and NIV in barley suggest F. poae and F. graminearum adapt differently to crop residues. Crop rotations that are effective in managing F. graminearum/DON might not be suitable for limiting F. poae infection and NIV accumulation in barley. We speculate the lesser impact of cropping rotation on the abundance of F. poae and associated mycotoxins could be due to the broad host range of F. poae, which includes small grain cereals, soybean, sunflower, and various broad-leaf grasses [55][56][57].
The multi-species nature of FHB on barley also raises the question about the impact of interactions between different Fusarium species during infection and consequent mycotoxin production in barley. Tan et al. [36] show pre-inoculation of wheat ears with F. poae before F. graminearum led to a disappearance of FHB symptoms and reduced mycotoxin levels compared to a single F. graminearum infection, while F. poae exhibited increased growth. Moreover, F. poae can asymptomatically induce salicylic acid and jasmonic acid related defences, which could hamper a later infection by F. graminearum [58]. Additionally, Walkowiak et al. [59] report co-inoculation with F. graminearum 3-ADON and 15-ADON strains results in a reduction in infection and toxin accumulation in planta. Evidence also suggests the co-occurrence of DON and NIV might synergistically affect their cytotoxicity [60].

Conclusions
The current study investigated barley grain collected in Manitoba for Fusarium species spectra and naturally occurring mycotoxins. We demonstrate that Fusarium spectra associated with FHB of barley in Manitoba are diverse and mainly included F. poae, F. graminearum, and F. sporotrichioides. NIV and DON are the two most common mycotoxins detected in barley grains in Manitoba. Results also demonstrated that disease management practices, such as fungicide and crop rotation, and weather conditions can impact the Fusarium species spectra and their respective mycotoxins in barley. Further monitoring of FHB in barley is imperative to confirming our findings on a greater temporal and spatial scale and to observe potential changes in the Fusarium species spectra and their respective mycotoxins. Our results also address the need for more research on the pathology of F. poae and F. sporotrichioides on barley.

Barley Sample Collection
Manitoba is located in the centre of the North American continent and its climate mainly falls into the humid continental climate zone. In general, the anthesis period of barley in Manitoba is between 1 to 15 July. In July, the mean temperatures of Manitoba were 19.   Figure S1) were surveyed for FHB between July 18 and August 5 of each year. The fields were surveyed between the late milk growth stage (BBCH 77) and full maturity (BBCH 89). Barley spikes (n = 40 to 60) were randomly collected from five locations in each field and stored in paper envelopes for subsequent laboratory analysis. Information on the preceding crops (canola, flax or cereal) was also collected for each field/sample location.

DNA Extraction and PCR Analysis of Fusarium Genomic DNA
A subsample of 20 g of barley grains per field was ground to flour using a Retsch grinder (Retsch ZM 200, Scientific Inc. Newtown, PA, USA). Total genomic DNA (gDNA) was extracted from 1 g of flour using a QIAGEN DNeasy Mini Kit (QIAGEN Mississauga, ON, Canada) following the manufacturer's procedure.
Quantitative PCR was used to determine the abundance of Fusarium gDNA in barley using a CFX96™ Real-Time PCR Detector System (BioRad, Mississauga, ON, Canada). All standards and the negative control (double-distilled water) were run in triplicate. The PCR reaction was carried out in a total volume of 20 µL, consisting of 10 µL of Fast EvaGreen ® PCR Master Mix (BioRad, Mississauga, ON, Canada), 1 µL of each primer, 6 µL of doubledistilled water, and 2 µL of template DNA with a 37-cycle threshold (Ct) detection limit. PCR conditions were as follows: initial preheating at 98 • C for 2 min; 40 cycles of 95 • C for 15 s and 62 • C for 1 min; and dissociation curve analysis at 60 to 95 • C. Three technical replicates were performed for each sample. and Agri-Food Canada. Ground barley samples (1 g) were extracted for mycotoxins using a solvent mixture (10 mL; 75% acetonitrile: 10% methanol: 15% water) in a 10-mL flatbottomed tube. The sample-solvent mixture was thoroughly mixed by inversion to wet the flour sample completely, followed by sonication (30 min). The tubes were then loaded onto a rotary shaker and extracted for 90 min at 40 rpm. The sample-solvent mixture was centrifuged (4000× g for 30 min) to separate the solvent from the ground material and the supernatant was filtered through a 0.2-µm-nylon syringe filter (ThermoFisher, Mississauga, ON, Canada) into a clean 10-mL flat-bottomed tube. An aliquot of the filtered extract was dried under a gentle stream of nitrogen gas using a sample evaporator (RapidVap Labconco, Kansas City, MO, USA) at 100% speed for 90 min at 70 • C. Each dried-sample extract was dissolved and re-suspended in 1 mL of solvent mixture (50:50, water + 0.1% formic acid + 5 mM ammonium formate: methanol + 0.1% formic acid + 5 mM ammonium formate) by vortexing and brief sonication. The extract was transferred into an amber liquid chromatography vial before analysis using ultra-high performance liquid chromatography (UHPLC; Vanquish, Thermo Fisher Scientific, Mississauga, ON, Canada) coupled with highresolution mass spectrometer (HRMS; Orbitrap ID-X Tribrid Mass Spectrometer, Thermo Fisher Scientific Inc., Mississauga, ON, Canada).

Detection and Quantification of Fusarium Mycotoxins
Separation of various mycotoxins was achieved using a reverse-phase C-18 core-shell silica column (particle size 1.7 µm, 100 × 2.1 mm, Kinetex, Phenomenex, CA, USA) held at 35 • C. Gradient elution was achieved with 100% water (mobile phase A) and 100% methanol (mobile phase B), with both phases containing 0.1% formic acid and 5 mM ammonium formate, at a flow rate of 0.2 mL min −1 for 20 min. Retention times and recoveries for various mycotoxins are identified in Supplementary Table S2. Heatedelectrospray ionization (H-ESI) was used to achieve a steady state electrospray from the chromatographic separation column. Mass spectra for each sample was obtained in positive mode with an obitrap mass resolution of 120,000 for precursor ion mass-to-charge (m/z) values with a mass tolerance of ±5 ppm; product ions mass spectra were acquired at with an orbitrap resolution of 30,000. Precursor and product ion m/z values are noted in Supplementary Table S2. Fragmentation of precursor ions was achieved using a higher energy collision-induced dissociation (HCD) with stepped collision energies (15%, 20%, and 25%). The precursor ion mass was used for analyte quantification while product ions were used for analyte confirmation. Spiked reference material was simultaneously extracted with each batch of samples for quality control.

Sensitivity of F. poae and F. graminearum to Triazole Fungicides
The sensitivity of F. poae and F. graminearum to triazole fungicides, including Caramba ® (metconazole, BASF Canada), Prosaro ® (prothioconazole + tebuconazole, BAYER Canada), Folicur ® (tebuconazole, BAYER Canada), and Proline ® (prothioconazole, BAYER Canada), was analysed using a potato-dextrose-agar (PDA)-based petri plate assay, supplemented with different concentrations of fungicide active ingredients. Fifteen F. poae and 15 F. graminearum strains were included in the analysis. These Fusarium strains were randomly selected from the Fusarium strains isolated from barley samples collected from 2017 to 2019. The in vitro assay was carried out in triplicate at two doses (0.01 and 0.10 mg of active ingredient (ai)/L PDA). The incubation protocol of the plates was as described by Spolti et al. [15]. The petri plates were incubated at 25 • C for 3 days, and then the mycelial growth of Fusarium strains measured. Mycelial growth inhibition (colony diameters on each plate) was determined according to: % inhibition = radial growth o f untreated control plate − radial growth o f treated plate radial growth o f untreated control plate × 100

Statistical Analysis
For statistical analyses, the homogeneity of Fusarium species abundance (qPCR) and mycotoxin concentrations (UHPLC-HRMS) from the barley fields/samples (n = 149 over 3 years) was verified using the Kolmogorov-Smirnov test of normality, which measures the divergence of the field sample distribution. The skewness and kurtosis were determined and plotted for each year of survey data to determine whether the data distribution was normal. When necessary, the data were transformed (log, square root, arcsine) or dataset outliers (extreme low and high values) were removed. NIV and DON concentrations were described with box and whisker plots using Microsoft Excel 2016, showing the distribution of mycotoxin concentrations in terms of the mean, median, minimum, maximum, and outliers. The least significant difference (LSD) test was performed to compare mean concentrations of Fusarium species gDNA. Student's t-test was used to compare the pair means of mycotoxins (DON and NIV) within each sampling year.
To determine the impact of crop rotation and survey year on the mean concentration of DON and NIV in barley grains, the GLIMMIX procedure in SAS v.9.4 was used to estimate the fixed and random effects of previous crop (PC), year (Y), and PC × Y interaction. Multiple comparisons of means were made using the Tukey-Kramer test (alpha = 0.05). A one-way ANOVA was performed to test the significance of previous crops on DON and NIV concentrations using data from all three years. SAS v.9.4 was also used to identify the significance of fungicides and doses in controlling the growth of Fusarium strains. Differences in the inhibition of the mycelial growth of F. graminearum and F. poae against triazole fungicides were analysed using Student's t-test.