Introduction

Triticale (× Triticosecale) is an intergeneric hybrid of wheat (Triticum spp.) and rye (Secale cereale). It was originally bred to combine the best qualities of both crops, offering an excellent alternative for regions unsuitable for wheat production (^{Dobreva, et al., 2016}; ^{Góral et al., 2015}; ^{CIMMYT, 1976}). In Mexico, triticale was planted on 24, 152 ha in 2019, producing 450, 518 t of green forage and 23, 039 t of grain. The main forage-producing states were Durango, Querétaro, and Chihuahua, while the State of Mexico led in grain production (^{SIAP, 2023}). Although triticale is mainly grown for animal feed, recent genetic improvements have focused on enhancing grain quality for human consumption (^{Mcgoverin et al., 2011}; ^{Mergoum et al., 2009}).

When first developed, triticale was considered resistant to various pests; however, the expansion of its cultivation area increased the selection pressure from harmful organisms, making this cereal more susceptible to insect pests and diseases common to rye and wheat. Among these, diseases have become one of the main constraints on triticale production, with Fusarium head blight (FHB) standing out as particularly significant (^{Dobreva, et al., 2016}; ^{Góral et al., 2015}).

Fusarium head blight is one of the most aggressive diseases affecting small grain cereals (^{Backhouse, 2014}; ^{Magliano and Chulze, 2013}; ^{Zillinsky, 1984}). Severe FHB epidemics have been reported worldwide, reducing yields by up to 80% and significantly lowering grain quality (^{Alisaac and Mahlein, 2023}). The disease decreases both grain weight and protein content, alters carbohydrate composition, and leads to contamination with trichothecenes mycotoxins mainly deoxynivalenol (DON) and nivalenol (NIV) which are harmful to human and animal health and pose a threat to food safety (Magliano and Chulze, 2013; ^{Sobrova et al., 2010}; ^{Veitch et al., 2008}).

Until recently, Fusarium graminearum (teleomorph: Gibberella zeae (Schwein)) was considered the main causal agent (^{McMullen et al., 2012}). However, other species may also be associated with the disease, depending on climatic conditions (temperature and relative humidity), geographic location, and the crop involved (^{Xu and Nicholson, 2009}; ^{Qu et al., 2008}). In Mexico, the predominant species are F. graminearum and F. boothii, as well as F. culmorum, F. equiseti, F. avenaceum, and F. tricinctum (^{Cerón-Bustamante et al., 2018}; ^{Leyva-Mir et al., 2017}; ^{Osman et al., 2015}; ^{Ireta and Gilchrist, 1994}). Fusarium head blight is a difficult disease to control due to its aggressive nature. Nevertheless, the use of resistant varieties is key in planning and implementing an integrated management strategy, as it is considered the most economical, effective, and sustainable control method (^{Gilbert and Fernando, 2004}; McMullen et al., 1997).

The few studies conducted on triticale report wide variation in resistance to Fusarium head blight. ^{Góral and Arseniuk (2003}) noted significant variation among triticale genotypes in resistance to Fusarium. According to ^{Oettler and Wahle (2001}) and ^{Góral et al. (2021)}, triticale is less susceptible to Fusarium head blight than wheat. In a comparative study, ^{Góral (2009)} found that winter triticale lines infected with F. culmorum showed half the symptom severity observed in wheat lines.

Currently, there are no studies in Mexico on the levels of resistance that triticale has to Fusarium head blight. This underscores the need to conduct research to characterize the resistance of triticale varieties, which would also help identify new sources of resistance for use in breeding programs. The objective of this study was to evaluate, under greenhouse conditions, the resistance levels of seven triticale varieties to infection by F. graminearum, F. boothii, and F. camptoceras.

Materials and Methods

Experiment location. The experiment was conducted in 2018 and 2019 at the National Laboratory for Rusts and Other Cereal Diseases (LANAREC), part of the National Institute for Forestry, Agriculture and Livestock Research (INIFAP), at the Valle of México Experimental Station (CEVAMEX), located at 19° 29' N and 98° 54' W, at an altitude of 2,260 masl.

Varieties. Seven triticale varieties cultivated in Mexico were evaluated: Caborca TCL79, Eronga TCL83, Secano TCL96, and Impulso TCL2016, released by INIFAP; Pollmer and Siglo TCL21, developed by the International Maize and Wheat Improvement Center (CIMMYT); and Bicentenario TCL08, released by the Institute for Agricultural, Aquaculture and Forestry Research and Training (ICAMEX).

Inoculum. Three species were evaluated: F. graminearum, F. boothii, and F. camptoceras, provided by the Universidad Autónoma Chapingo, State of Mexico. These species were propagated through monospore cultures on carnation leaf agar (CLA) medium at the Mycology Laboratory of the Department of Agricultural Parasitology at the Universidad Autónoma Chapingo. The CLA medium consisted of 2% water agar (20 g of agar in 1 L of water) plus 10 to 12 sterile 5 mm carnation leaf pieces placed in Petri dishes. This method was used to promote the formation of Fusarium colonies producing macroconidia of uniform size and form (^{Leslie and Summerell, 2007}).

Experimental design. A completely randomized design with a split-plot arrangement was used, where the main plots were the Fusarium species and the subplots were the triticale varieties. A total of 21 treatments were established (seven varieties × three Fusarium species), with three replications plus controls.

Planting and agronomic management. Under greenhouse conditions, the seven varieties were planted in seven-inch plastic pots filled with a sterilized substrate (60% soil and 40% peat moss). Five seeds were sown per pot and later thinned to retain the three most vigorous plants. A total of 12 pots were used per variety (three pots per replication). Fertilization was carried out using urea (46-00-00) and Tricel-20 (a foliar fertilizer), based on the crop’s nutritional requirements.

Inoculation. Plants were inoculated at the anthesis stage (approximately 65 days after sowing). A spore suspension was prepared at a concentration of 1 × 10⁶ conidia mL⁻¹, and the point inoculation method using the cotton technique was employed. Three previously marked spikes per pot were selected, and two 20 mg cotton tufts soaked in the spore solution were placed under the glumes at the midpoint of each spike. The inoculated spikes were then covered with glassine paper bags, following the methodology described by ^{Gilchrist-Saavedra et al. (2005}) and ^{Imathiu et al. (2014}). The control for each genotype was inoculated with sterile distilled water using the same technique.

After the spikes were inoculated with Fusarium species, the plants were placed in an incubation chamber for 96 hours at an average temperature of 25 ± 10 °C, with a 12-hour photoperiod. Dew was produced for eight hours during the first 24 hours and for four hours over the following 72 hours to maintain relative humidity below 95%. The plants were then transferred to a greenhouse, where they were kept at an average temperature of 25 ± 10 °C until symptoms appeared.

Evaluations. Five evaluations of disease severity were carried out for each treatment at 4, 9, 14, 19, and 26 days after inoculation (DAI), using a visual scale for Fusarium head blight in triticale (Figure 1), adapted for assessing severity under greenhouse conditions. This scale was based on the visual scale for estimating Fusarium severity in wheat spikes proposed by ^{Stack and McMullen (1998}). Based on the average severity percentage obtained, triticale varieties were classified as: Resistant (0-10), Susceptible (20-40), and High Susceptible (41-100).

Figure 1 Visual scale for determining the percentage of Fusarium head blight severity in triticale spikes under greenhouse conditions. 

Variables. The evaluated variables were: a) disease severity (DS), the percentage of the spike showing Fusarium infection; b) grain weight (GW), the weight of the grain obtained from each inoculated spike once it reached harvest maturity; and c) Area Under the Disease Progress Curve (AUDPC), calculated from the five DS readings taken at set time intervals.

Statistical analysis. The data for the evaluated variables were statistically analyzed jointly using SAS 9.4 software (^{SAS, 2018}), and mean comparisons for the studied varieties were performed using Tukey’s test (p≤0.05).

RESULTS AND DISCUSSION

Analysis of variance. The analysis of variance showed significant differences in the evaluations expressed in days after inoculation (DAI), among varieties, and in their interaction. This indicates that at least one variety exhibited a different pattern of severity across the five evaluations for each Fusarium species. For the variable Area Under the Disease Progress Curve (AUDPC), there were no significant differences among species; however, the sources of variation-varieties and the interaction between varieties × species-showed significant differences. This suggests that the varieties displayed different levels of disease development across the species. In contrast, for grain weight per spike (GW), significant differences were found among species, among varieties, and in the varieties × species interaction, indicating that the average GW varied both among species and among varieties.

Response of the varieties. Figure 2 shows the disease development pattern in the seven varieties over the five evaluations for the species F. boothii. Significant differences were found among varieties (p<0.0001) in the final evaluation at 26 DAI. In other words, the severity percentage was significantly lower in the varieties Bicentenario TCL08 and Siglo TCL21, with values below 20%, compared to Secano TCL96, Impulso TCL2016, Caborca TCL79, and Eronga TCL83, which showed greater disease development.

For the species F. camptoceras, significant differences were found among varieties (p<0.0001) (Figure 3). Fusarium head blight severity was significantly higher in the varieties Caborca TCL79 and Impulso TCL2016, with values exceeding 80% at 26 DAI, while Bicentenario TCL08 and Siglo TCL21 showed significantly lower severities of 13.6% and 16.7%, respectively.

Figure 2 Performance of the varieties inoculated with F. boothii over the five evaluations. 

Figure 3 Performance of the varieties inoculated with F. camptoceras over the five evaluations. 

Regarding disease severity caused by F. graminearum, the lowest final severity values were observed in Siglo TCL21 (1.56%) and Bicentenario TCL08 (7.78%), which were significantly different (p<0.0001) from those of the other varieties. The highest severities were recorded in Pollmer (81.11%) and Caborca TCL79 (75.78%) (Figure 4).

Figure 4 Performance of the varieties inoculated with F. graminearum over the five evaluations. 

Most of the varieties began showing characteristic disease symptoms starting at four days after inoculation (DAI) with each Fusarium species. However, Bicentenario TCL08 began to show symptoms at 14 DAI for F. graminearum, and Siglo TCL21 showed delayed symptom development for both F. boothii and F. graminearum. Both varieties also exhibited severity percentages below 17% for all three Fusarium species. As suggested by ^{Bai and Shaner (2004}), ^{Imathiu et al. (2014}), and Schroeder and Christensen (1963), this behavior may be due to the fungus penetrating plant tissues without spreading through the spike, resulting in the absence of visible symptoms. In other words, Fusarium hyphae may have failed to invade the cells adjacent to the infection site, likely because these varieties possess genetic traits that prevented fungal progression or colonization.

Combined response of the varieties across the five evaluations. Table 1 presents the mean comparisons among varieties for the variable average severity percentage for each of the Fusarium species.

For F. boothii, significant differences were found among varieties (p<0.0001). Bicentenario TCL08 (3.7%) and Siglo TCL21 (4.8%) were statistically different from Secano TCL96 (47.6%), Caborca TCL79 (41.2%), and Eronga TCL83 (38.7%), showing the lowest severity percentages. Similarly, for F. camptoceras, significant differences were observed among varieties (p<0.0001). Siglo TCL21 and Bicentenario TCL08 had the lowest severity values, with 6.4% and 6.8%, respectively, while Caborca TCL79 showed the highest severity at 54.3%. Finally, for F. graminearum, significant differences were also found among varieties (p<0.0001). Bicentenario TCL08 showed the lowest severity at 0.8%, and Caborca TCL79 the highest at 43.3%. The results indicated that average severity across varieties ranged from 3.7% to 47.6% for F. boothii, from 6.4% to 54.3% for F. camptoceras, and from 0.8% to 43.3% for F. graminearum.

Disease severity varied widely due to the different interactions established between the pathogen and the host. These variations were observed in the seven triticale varieties inoculated with the three Fusarium species, where significant differences were found in the average severity percentage (p<0.0001). Notably, Siglo TCL21 and Bicentenario TCL08 consistently showed significantly lower severity values for all three fungal species, indicating resistance, while the remaining varieties were susceptible, with Caborca TCL79 being the most susceptible. This confirms the existence of a wide range of resistance levels among the evaluated varieties, in line with findings by ^{Oettler and Wahle (2001}), who reported variation in resistance to F. culmorum among 100 triticale genotypes despite infection pressure. Similarly, ^{Kalih et al. (2014}) observed broad variation in Fusarium head blight resistance in dwarf triticale lines, explained by the presence of QTLs that confer resistance despite the presence of the Ddw1 gene, which reduces plant height. ^{Góral and Arseniuk (2003}) also found significant variation in resistance levels to Fusarium head blight among parental triticale genotypes.

Area Under the Disease Progress Curve (AUDPC). For the AUDPC variable, significant differences were found among varieties for all three species (p<0.0001) (Table 2). The varieties Siglo TCL21 and Bicentenario TCL08 were statistically different from the others, showing the lowest disease development values for F. boothii, F. camptoceras, and F. graminearum. In contrast, Secano TCL96 showed the highest disease development for F. boothii, with an area of 1111.9 units (u), while for F. camptoceras and F. graminearum, the highest values were observed in Caborca TCL79, with 1262.2 u and 1015.4 u, respectively. Among the varieties, AUDPC ranged from 87.5 u to 1111.9 u for

F. boothii, from 141.2 u to 1262.2 u for F. camptoceras, and from 18.7 u to 1015.4 u for

F. graminearum.

Grain weight per spike. Table 3 shows the average grain weight per spike for the evaluated varieties under each fungal species and the control. A variability in grain weight per spike was observed among the triticale varieties in response to the different Fusarium species.

For F. boothii, significant differences were found among varieties (p<0.0002), with Siglo TCL21 showing the highest grain weight per spike at 0.73 g. In contrast, for F. camptoceras, there were no statistically significant differences among varieties (p<0.0652), indicating that the damage to grain was similar across varieties for this species. For F. graminearum, significant differences were observed (p<0.0046), with Bicentenario TCL08 showing a significantly higher grain weight (0.7 g) compared to Caborca TCL79 and Secano TCL96.

All evaluated varieties showed lower grain weight per spike when inoculated with F. boothii, F. camptoceras, and F. graminearum compared to the control. This reduction was due to damage caused by Fusarium, which affected the pericarp, induced structural changes in the endosperm, and altered starch morphology, resulting in deformed and lightweight grains, as reported by ^{Packa et al. (2008}).

In this study, the variety Bicentenario TCL08 showed resistance and recorded the highest grain weight, while Secano TCL96 was the most susceptible, with grain weight reduced by up to 90% (Table 3). In contrast, ^{Oettler and Wahle (2001}) reported that infection by F. culmorum caused a 48% reduction in average grain weight per spike in triticale genotypes. Although grain can still be harvested from plants affected by Fusarium head blight, it is essential to assess the levels of mycotoxins present in the damaged grain. In a study by ^{Góral et al. (2016}), it was found that triticale grain can accumulate higher levels of trichothecene mycotoxins than wheat. For this reason, it is important to avoid planting genotypes that show high disease severity and high accumulation of these mycotoxins.

Although few studies have examined resistance to Fusarium head blight in triticale compared to other cereals, triticale may be as resistant as wheat and less resistant than rye, as noted by ^{Gaikpa et al. (2019}), ^{Góral et al. (2016}), ^{Góral (2009)}, ^{Miedaner et al. (2004}), and ^{Miedaner et al. (2001)}. However, since there is no research in Mexico on the levels of resistance that triticale, varieties have to Fusarium head blight; this study identified moderate variation in resistance levels among varieties. This information can be useful for triticale breeding programs in Mexico aimed at developing new varieties with resistance to this disease.

CONCLUSIONS

The varieties Siglo TCL21 (3.1 to 6.4%) and Bicentenario TCL08 (0.8 to 6.8%) showed the lowest severities of Fusarium head blight caused by Fusarium boothii, F. camptoceras, and F. graminearum, and were considered resistant. The most susceptible variety to F. boothii was Secano TCL96 (47.6%), while for F. camptoceras (54.3%) and

F. graminearum (43.3%) it was Caborca TCL79. Grain weight per spike was reduced by up to 90% in the case of Secano TCL96, which was considered susceptible to the disease. It is important for cereal breeding programs to incorporate resistance to different Fusarium species in triticale cultivation in Mexico.