Cysteine (CYS), reduced (GSH) and oxidized (GSSG) glutathione, and trifluoroacetic acid (TFA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Phytochelatins: (γ-Glu-Cys)2-Gly phytochelatin2 (PC2), (γ-Glu-Cys)3-Gly phytochelatin3 (PC3), (γ-Glu-Cys)4-Gly phytochelatin4 (PC4), and (γ-Glu-Cys)5-Gly phytochelatin5 (PC5) were synthesized at Clonestar Biotech (Brno, Czech Republic) with a purity above 90%. HPLC-grade methanol (>99.9%; v/v) came from Merck (Dortmund, Germany). Other chemicals were purchased from Sigma-Aldrich, unless noted otherwise. Stock standard solutions of the thiols (1 mg/mL) were prepared with ACS water (Sigma-Aldrich) and stored in darkness at −20 °C. Working standard solutions were prepared daily by dilution of the stock solutions. All solutions were filtered through 0.45 μm nylon filter discs (Millipore, Billerica, MA, USA) prior to analysis using HPLC-ED. The pH was measured using an inoLab WTW Level 3 instrument equipped with a Level 3 probe (inoLab, Weilheim, Germany).

Fungi species were identified based upon cultivation on specific media and specific metabolic markers. The infection was assessed on a scale from 1 (highest level of infection) to 9 (plots free of disease). Degree 1 represents very low resistance to fungal diseases (>90% diseased plant leaves per plot), degree 3 low resistance (65–90% diseased plant leaves per plot), degree 5 intermediate resistance (40–65% diseased plant leaves per plot), degree 7 high resistance (15–40% diseased plant leaves per plot), and degree 9 very high resistance (<15% diseased plant leaves per plot). Evaluated on two dates (31 May and 12 July), the levels of pathogen infection in experimental plots ranged on the 9-point scale from 7.3 to 8.0.

The plants were sampled at specific time intervals. Immediately after sampling, the plants were washed three times with distilled water, dried with filter paper, and frozen using liquid nitrogen prior to further experimental procedures. To determine fresh weight, the plants were weighed on Sartorius R160P balances (Sartorius GmbH, Goettingen, Germany) immediately after collection from the filter paper.

Weighed plant tissues (approximately 0.2 g fresh weight) were transferred to a test tube and prepared according to Supalkova et al. [41]. Liquid nitrogen was then added to the test tube, and the samples were frozen to disrupt the cells. Subsequently, 1 mL of 0.2 M phosphate buffer (pH 7.2) was added to the test tube. The mixture was prepared using an ULTRA-TURRAX T8 hand-operated homogenizer (IKA, Staufen, Germany) at 25,000 rpm for 3 min. Homogenate was then transferred to a new test tube, where it was shaken on a Vortex-2 Genie (Scientific Industries, New York, NY, USA) at 4 °C for 30 min. The homogenate was centrifuged (14,000 g) for 30 min at 4 °C using a Universal 32 R centrifuge (Hettich-Zentrifugen GmbH, Tuttlingen, Germany). The supernatant was filtered through a 0.45 μm nylon membrane filter disk (Millipore, Billerica, MA, USA) prior to analysis.

The HPLC-ED system consisted of two solvent delivery pumps (Model 582, ESA Inc., Chelmsford, MA, USA) operating in the range of 0.001–9.999 mL/min, a Zorbax eclipse AAA C₁₈ (150 × 4.6; 3.5 μm particles, Agilent Technologies, Palo Alto, CA, USA) column and a CoulArray electrochemical detector (Model 5600A, ESA Inc.). The electrochemical detector included one flow cell (Model 6210, ESA Inc.). The cell consisted of four working porous carbon electrodes, each with auxiliary and dry Pd/H₂ reference electrodes. Both the detector and the reaction coil/column were thermostated at 32 °C. The sample (20 μl) was injected using an autosampler (Model 542 HPLC, ESA Inc.). Samples were kept in the carousel at 8 °C during the analysis. The mobile phase consisted of 80 mM TFA (A) and methanol (B). The compounds of interest were separated by the following linear gradient: 0 → 1 min (3% B), 1 → 2 min (10% B), 2 → 5 min (30% B), and 5 → 6 min (98% B). The mobile phase flow rate was 1 ml/min, and the working electrode potential was 900 mV [42,43,44,45,46]. Time of analysis was 30 min.

Data were processed using Microsoft Excel^® (Microsoft Corporation, Redmond, WA, USA) and Statistica CZ Version 8.0 (Prague, Czech Republic). Results are expressed as mean ± standard error of the mean, unless noted otherwise (Excel^®). Statistical significances were determined using Statistica CZ. Differences with p < 0.05 were considered significant and were determined using multifactorial ANOVA (in particular, Scheffé’s test), which was applied for comparing means.

In this study, three grass species—F. pabulare (Figure 1A and B), L. perenne (Figure 1C) and F. braunii—were studied to determine whether low molecular mass thiols are involved in protective mechanisms against molds in grasses. Detecting the presence of molds on the species during the cultivation period was of primary interest. It is common knowledge that grasses can be attacked by many species of fungi.

In the spring of 2011, there was widespread infection of grasses by Fusarium nivale. Additionally, many plants experienced multiple infections by Colletotrichum graminicola, F. pabulare and Ascochyta. In late summer and autumn, Puccinia graminis and Puccinia coronata occurred on all experimental species (Figure 1C). To a lesser extent, Blumeria graminis was also detected. Such other fungi as Alternaria spp., Cochliobolus spp., Aspergillus niger, and Ramularia spp. also were observed, but their occurrence was negligible as compared to the aforementioned species. Fungal diseases altered forage production (volume) and quality due to the fact that grasses actively defend themselves against the presence of a pathogen. There are two main types of defense pathways: (i) mechanical, and (ii) biochemical. Mechanical defense is based on an increase in the content of solid fibrous elements, which subsequently reduces the digestibility of organic matter and leads to a decline in forage quality [47]. The stress-induced biochemical pathways involve the production of several types of compounds which are able to detoxify the xenobiotics produced by fungi. Such low molecular mass thiols as cysteine, reduced glutathione, and phytochelatins are among these compounds. Thiols have been shown able to detoxify both inorganic [48,49,50,51,52] and organic [53,54,55,56,57] xenobiotics. One may also consider the fact that the occurrence of fungi may be associated with the risk of mycotoxins which reduce the health safety of forage [2]. Therefore, we were interested in evaluating the behavior of grasses attacked by a pathogen and the influence of treatment and mold presence on the content of low molecular mass thiols, because treatment itself by this fungicide, and other strobilurins, increases yield and delays plant senescence even in the absence of disease [58,59,60]. Respiration and the plant redox status are critical to the production of the thiols [61], and thus the potential for the fungicide treatment to alter this redox status is of great significance [62,63]. The fungicidal activity of strobilurins relies on their ability to inhibit mitochondrial respiration by binding at the so-called Qo site (the outer quinol-oxidation site) of the cytochrome bc1 enzyme complex (complex III). This inhibition blocks electron transfer between cytochrome b and cytochrome c1, which, in turn, leads to an energy deficiency in fungal cells by halting the production of ATP [64]. Mitochondrial glutathione, as a key survival antioxidant, could serve in the forming of some resistance of fungi against these fungicides [65].

Attention was primarily focused on plant development. Growth of the grasses was accompanied by a lengthening stalk formation and generative organs in the first crop. At the time of harvest, grasses were at earring stage. After the first cutting, a period of intense tillering followed. As shown in Figure 2, temperature and precipitation at the research station in Vatin were sufficient for growth within the period from June to July. An absence of vernalization was reflected in the formation of leafy shoots and absence of fertile stalks. Similar conclusions apply also for grass growth after the second cutting. No visual difference (growth, color) was present in the growing of plants treated with fungicide as compared with those not treated. Assessment of disease infection level of the studied species prior to and after the first fungicide application is shown in Table 1. While during the summer months the species showed comparable levels of disease infection, that situation changed somewhat when stronger pathogen pressure was apparent in late September. No visible differences in infection level were found between grasses treated with fungicides and those untreated. A tendency toward greater pathogen pressure was evident in L. perenne, but the differences shown in Table 1 were not statistically significant, even though L. perenne is a species more susceptible to pathogens as compared to the more resistant Festuca arundinacea and its hybrids. [66,67]. Plants in the experiment were not artificially infected, and the pathogens did not occur. The level of infection can be influenced by many factors. Comparable and low levels of infection in the evaluated species could have been affected by weather conditions adverse for the growth of mold. Had higher pressure from pathogens been evident, that might have substantiated a benefit from use of fungicides and showed the a priori anticipated differences between treated and untreated variants. The plants sown under controlled conditions served as a control group and showed no occurrence of fungal diseases. Under the controlled conditions, grass sprouted between 4 d and 10 d. Due to the fact that these are winter species, they formed leafy shoots, with no tendency to produce fertile stalks. Biomass sampling was performed 60 d after sowing in the tillering stage.

Next, low molecular mass thiols were determined. Determination was carried out as in our previously published work [42,43,44,45,46], in which our group had demonstrated detection limits for these thiols measurable down to nanograms per gram of fresh material (ng/g), with a linear detection range down to four orders of magnitude in concentration and recovery of the thiols exceeding 80%. This method was used to analyze the experimental plants. GSH and GSSG are well-known markers of oxidative stress and therefore can be used as good markers of the negative effect of mold on grasses. In relation to phytochelatins, they are mainly considered as metal pollution markers, but their richness in thiol moieties could give them the ability to scavenge reactive oxygen species [68,69]. Festulolium pabulare had higher levels of GSH and PC3 in the pot experiment (Table 2). Moreover, lower GSSG content resulted in a high GSH/GSSG ratio.

In the field experiment (Table 3 and Table 4), PC2, PC4, and PC5 content was higher in F. pabulare. Conversely, F. braunii showed higher content of CYS. The species differed significantly (p < 0.05), however, in PC4 and PC5 content. Lower contents of GSH and GSSH and a lower GSH/GSSG ratio were recorded in September. The differences between months were not statistically confirmed. An exception was the low content of GSSG (p < 0.05) in September, taking into account the treatment factor (Table 4). Fungicide application did not result in demonstrable differences in the average content of low molecular mass thiols (Table 4).

A few papers reporting on the evaluation of thiol content in the biomass of grasses have been published. For example, Raab et al. detected PC2 and PC3 in the biomass of Holcus lanatus [70]. This species is resistant to heavy metals and its arsenic contamination showed a higher content of PC3 than did Pteris cretica, which is vulnerable to heavy metals. In our pots experiment (Table 2), disease-resistant F. pabulare showed higher PC3 levels than did the more pathogen-susceptible L. perenne and F. braunii. In our field experiment, the content of PC4 and PC5 was significantly higher also in F. pabulare. Hunaiti et al. evaluated the influence of heavy metals and sulfate content on thiols in Cynodon dactylon, Amaranthus gracilis, Chenopodium murale, and Typha domingensis [71]. The increase of metal and sulfate levels in the soil at their contaminated site was correlated with a rise of total glutathione and cysteine in plants. However, the individual effects of metals and sulfate on glutathione, short-chain phytochelatin, and long-chain phytochelatin levels were dissimilar. According to Jia et al., this fact depends on plant species [72].

The content of GSSG found in the pots experiment was comparable to the values determined under field conditions in L. perenne and F. braunii during the growing season. That assessment does not apply to F. pabulare, where an obvious decrease of GSSG from June to September and low GSSG content in plants grown under control conditions was observed (Figure 3A). The relationship between Date and Species was significant (p < 0.05) in the case of GSSG (Table 5). Comparable small changes in GSSG content during the growing season are reported in those species susceptible to fungal diseases (i.e., L. perenne and F. braunii). In contrast, F. pabulare, which can be considered resistant to infection, clearly showed a high GSSG content in June. Moreover, the content of GSSG in F. pabulare was low under control conditions.

Similarly, differences between species susceptible to fungal diseases (L. perenne and F. braunii) and resistant species (F. pabulare) in the assessment of GSH/GSSG are shown in Figure 3B. While the more susceptible species show a higher ratio of GSH/GSSG in June, decreasing thereafter in the continuing growing season, F. pabulare shows comparable values from June to September. A higher ratio of GSH/GSSG was found, however, in plants of F. pabulare grown under control conditions. Biomass obtained from plants grown under controlled conditions had a higher content of PC3. This fact, however, applied only to L. perenne and (especially so) to F. pabulare (Figure 3C). The relationship between Species and Treatment was significant (p < 0.05, Table 5). Even more substantial are the differences between PC3 values. Lolium perenne and F. braunii show comparable PC3 content from June to September. Lolium perenne grown under control conditions exhibited a higher PC3 content, but this discrepancy was not statistically significant. Conversely, F. pabulare grown in control conditions had high (p < 0.05) PC3 content (Figure 3D). Differences in PC3 content between plants grown under control conditions and in field experiments are very obvious. Biomass obtained from plants grown under controlled conditions had a higher PC3 content. This fact, however, applies only to L. perenne and (especially so) to F. pabulare. The relationship between Species and Treatment was significant (p < 0.05, Table 5).