Comparison of Wheat Quality, Antioxidant Activity, and Mycotoxins Under Organic and Conventional Farming

Abstract

Wheat (Triticum aestivum L.) is a major global staple crop, widely consumed in processed forms such as bread and pasta. As consumer demand for healthier food options increases, organic wheat production is gaining importance. However, organic farming excludes the use of synthetic pesticides and fungicides, potentially increasing the risk of fungal contamination and mycotoxin presence. At the same time, questions remain about whether organically grown wheat can match the grain quality needed for industrial processing, particularly in terms of protein content, gluten strength, and dough properties. This study aims to evaluate grain and flour quality parameters, as well as the occurrence of selected mycotoxins, in eight winter wheat cultivars grown under both organic and conventional farming systems in northern Serbia during the 2023/2024 season. Cultivars included modern premium varieties recommended for organic production, as well as one traditional and one conventional cultivar. Despite unfavourable weather conditions in the early stages of the vegetation in 2024, favourable conditions during grain ripening contributed to the good quality of organically produced varieties. The quality parameters of most varieties from organic production (PC > 17%, WG > 49%, WA > 62%) were significantly higher than the standard for organic wheat. NS Epoha stood out as the variety with the highest yield in organic production. As expected, varieties from the organic system had a higher incidence of mycotoxin contamination, but their concentrations were low.

1. Introduction

Common wheat (Triticum aestivum L.), along with other cereal grains, represents a vital source of nutrients and energy in the human diet. It is the second most produced cereal globally, following maize, and the predominant cereal crop in Europe, accounting for nearly half of the total cereal harvest [1]. In Serbia, wheat accounts for a similarly important share of the total cereal production, with an average annual output of 3.3 million tons during 2021–2023 [2]. Wheat serves as the main ingredient in a wide variety of food products, including bread, pasta, pastries, and numerous other processed foods. Globally, it ranked second in 2018 in terms of food supply available for human consumption, with an estimated per capita availability of 78.4 kg per year [1].

Globally, the majority of wheat is produced through conventional farming, while organic production accounts for a significantly smaller share. In Serbia, organic production has been expanding since the 2010s, with cereals representing the second most prominent category within organic crop production. In 2017, organic cereals were grown on 4161 hectares, representing approximately 31% of all certified organic agricultural land in Serbia [3]. However, the area cultivated with organic cereals in Serbia remains considerably smaller compared to neighbouring Hungary, where 33,246 hectares were cultivated in 2018 [4]. This is significantly less compared with 14.8 million hectares in 2020, organic farming accounted for 9.1% of the European Union’s total agricultural area of which cereals represent 16% [5].

The rise in organic production is driven by growing consumer awareness of the health benefits associated with organic foods, including reduced exposure to pesticides and a higher intake of beneficial nutrients and bioactive compounds [6]. As a result, many major retail chains now include organic products their assortments [7]. Organic production prohibits synthetic pesticides, mineral fertilizers, plant growth regulators, and genetically modified cultivars, while emphasizing the use of organic fertilizers, such as manure and compost, to maintain soil fertility and plant health [8,9,10]. However, the absence of fungicide use allows certain fungal strains to develop on cereals, producing secondary metabolites known as mycotoxins [11,12]. Consumption of cereals contaminated with mycotoxins above regulatory limits can cause serious health problems in humans, including gastrointestinal disorders, hormonal imbalances, weakened immune function, and potential mutagenic and carcinogenic effects [13,14]. The key mycotoxins in wheat are deoxynivalenol (DON), zearalenone (ZEA), and T-2/HT-2 toxins, which are produced by Fusarium species infecting cereal grains in the field [11,13,14]. In addition, Fusarium species can also produce the mycotoxin moniliformin (MON) in cereals [15,16]. Among mycotoxins produced by Alternaria fungi, tentoxin (TEN) is among the four most frequently detected in wheat grains [17,18], while α-ergocryptine is one of the six principal alkaloids of Claviceps purpurea that may also occur in cereal grains [19,20]. On the other hand, several studies [21,22] confirm that organic cereal production enhances antioxidant and polyphenolic activity. Dietary intake of polyphenols may play a crucial role in promoting health by regulating metabolism, body weight, the progression of chronic diseases, and cell proliferation [23]. As one of the major subclasses of polyphenols, flavonoids significantly contribute to the overall antioxidant capacity of cereals, protecting biological molecules from oxidative damage by neutralizing free radicals [24].

In addition to the positive impact of organic wheat on human health, the end-use quality of the wheat itself is also very important. Wheat quality is a complex trait defined by various parameters that provide a comprehensive understanding of its suitability for different end-use applications in the food industry. Among these, protein content is the most important parameter, as it largely determines the market value of wheat [25]. Another key indicator of wheat quality is the test weight. A low test weight reflects a higher proportion of small and shrivelled grains in wheat bulk, which are associated with a higher bran content, and, consequently, reduced milling yield [26]. Furthermore, the viscoelastic properties of wheat dough-crucial for its performance-are determined by proteins in the grain endosperm, specifically monomeric prolamins and polymeric glutenins [27]. These proteins constitute the main components of gluten, the complex polymer responsible for dough functionality. While glutenins account for about 70% of dough elasticity [28], gliadins primarily contribute to its extensibility [29]. Both elasticity and extensibility are critical properties in bread production. During fermentation, elastic dough can retain the gases produced without collapsing, while dough with high extensibility contributes to improving bread volume [30]. Therefore, developing a wheat cultivar that combines these desirable properties remains a challenging task for breeders.

The specific objectives of this study were to: (a) evaluate the end-use quality of different wheat cultivars intended for organic production and compare them with those cultivated under conventional farming, (b) assess the impact of production systems (organic vs. conventional) on antioxidant activity, and (c) compare the levels of mycotoxin in wheat grain from different cultivars grown under organic and conventional practices.

2. Materials and Methods

2.1. Materials

A total of eight winter wheat (Triticum aestivum L.) cultivars were included in the study, cultivated under both organic and conventional conditions. The conventional cultivars included the domestic varieties Zvezdana and NS Epoha (

Table 1. Type of farming, end-use quality and originate of wheat varieties.

Variety	Type of Farming	End-Use Quality	Origin
Zvezdana	conventional	Good quality	Institute of Field and Vegetable Crops in Novi Sad, Serbia
NS Epoha	Conventional/Organic	milling-quality	Institute of Field and Vegetable Crops in Novi Sad, Serbia
Bánkúti 1201	Organic	Good quality	old Hungarian variety
Genotype 1	Organic	Premium	University of Debrecen, Debrecen, Hungary
Genotype 2	Organic	Premium	Probstdorfer Saatzucht GmbH & Co. KG, Probstdorf, Austria
Genotype 3	Organic	Premium	Probstdorfer Saatzucht GmbH & Co. KG, Probstdorf, Austria
Genotype 4	Organic	Premium	Saatzucht Edelhof, Edelhof, Austria
Genotype 5	Organic	Premium	Probstdorfer Saatzucht GmbH & Co. KG, Probstdorf, Austria

). Zvezdana is one of the five most commercially widespread varieties, valued for its consistent performance and adaptability in standard farming systems, while NS Epoha is a modern high-yielding milling-quality variety and well-balanced end-use characteristics [31]. The organic cultivars included Bánkúti 1201, an old Hungarian heritage variety known for its distinctive quality traits as well as one of the most widely used varieties in organic production in Hungary [32,33]. Also, five premium Hungarian and Austrian wheat varieties (Genotype 1, Genotype 2, Genotype 3, Genotype 4, and Genotype 5) specifically recommended for organic production and recently introduced to Serbia were included (Table 1). NS Epoha was cultivated under both organic and conventional systems to enable a direct comparison of how different growing practices influence its end-use quality, antioxidative properties, and mycotoxin contamination. All cultivars were sown in the autumn of 2023 on the same organic farm, together with the standard varieties NS Epoha and Zvezdana.

A field experiment was conducted during the 2023/24 growing season in Kanjiža, northern Vojvodina, Serbia (46°00′ N, 19°54′ E), representing a typical Pannonian environment. The soil at the experimental site was classified as a haplic chernozem aric [34]. Prior to sowing, soil samples were collected from the 0–30 cm layer for analysis, and the basic soil properties are presented in

Table 2. Basic soil properties of the organic production field in 2023.

pH		CaCO3	Humus	Total N	AL-P2O5	AL-K2O
KCl	H2O	Scheilbler, %	Tyurin, %	CNS, %	Egner–Riehm, mg/100 g	Egner–Riehm, mg/100 g
7.75	8.22	11.55	4.01	0.27	39.57	36.05

.

The production fields follow a long-term crop rotation of small-grain cereals, maize, and soybeans, with one of the field certified for organic production. The preceding crop for winter wheat was maize in both organic and conventional production fields. Selected winter wheat cultivars were sown at the optimal regional time (13–14 October 2023) at a density of 400–500 viable seeds per m², as recommended for large-scale production.

In the organic system, 346 kg per ha of organic fertilizer derived from composted and heat-treated poultry manure (Aminor; NPK 4/3/3 + 10 CaO + 60% OM + 30% C + SE) was applied before ploughing. No additional fertilizers or pesticides were used in organic production. In the conventional production system, a total of 120 kg N, 30 kg P₂O₅, and 30 kg K₂O per ha was applied. Weeds, diseases, and pests were controlled using recommended herbicides, fungicides, and insecticides (tribenuron-methyl + fluroxypyr; cyproconazole + picoxystrobin; deltamethrin). Harvest was carried out on 3 July 2024.

2.2. Grain and Flour Quality Traits

The protein content (PC) and moisture content of the grain were measured using a handheld GrainSense Analyzer (Oulu, Finland). Prior to milling, wheat samples were tempered to 13.5% moisture for 24 h. Thirty min before milling, the samples were further adjusted to a final moisture content of 15% and then milled using a Bühler MLU-202 laboratory mill (Uzwil, Switzerland). Also, the test weight (TW) was determined according to ISO 7971-3:2009 [35].

Wet gluten content (WG) and gluten index (GI) were determined following ICC Method 137/1 [36], using a Glutomatic 2200 system (Perten Instruments, Huddinge, Sweden). Dough properties, including water absorption (WA), farinograph resistance (FR), and softening degree (SD), were evaluated with a farinograph (Brabender OHG, Duisburg, Germany) following the Hungarian Standard Method 6369/653 [37]. Alveograph parameters, including deformation energy (W), extensibility (L), and the over-pressure/extensibility ratio (P/L), were assessed using a Chopin alveograph (Paris, France), according to ICC Method 121 [38].

2.2.1. Total Phenolic Content

The total phenolic content (TPC) was determined using a modified Folin–Ciocalteu assay [39]. Briefly, 0.5 mL of the wheat extract was transferred into a 10 mL volumetric flask containing 6.0 mL of double-distilled water and 0.5 mL of Folin–Ciocalteu reagent. After 60 s of reaction, 2.0 mL of 15% (w/v) sodium carbonate solution was added, and the mixture was stirred for 30 s. The volume was then adjusted to 10 mL with double-distilled water. The solution was incubated in the dark for 45 min, after which the absorbance was measured at 760 nm using a UV–Vis spectrophotometer (UV-2100, Unico Instrument Co., Shanghai, China). A calibration curve was prepared using gallic acid, and the results were expressed as milligrams of gallic acid equivalents per 100 g of dry matter (mg GAE/100 g DM).

2.2.2. Total Flavonoid Content

The total flavonoid content (TFC) was determined using the aluminum chloride colorimetric method [40]. In a 10 mL volumetric flask, 1.0 mL of wheat extract was mixed with 4.0 mL of double-distilled water, followed by the addition of 0.3 mL of 5% (w/v) sodium nitrite solution. After 5 min, 0.3 mL of 10% (w/v) aluminum chloride was added. One minute later, 2.0 mL of 1 mol/L sodium hydroxide was added, and the volume was adjusted to 10 mL with double-distilled water. The mixture was incubated in the dark at room temperature for 15 min, and the absorbance was measured at 510 nm. Quercetin was used as the calibration standard, and the results were expressed as milligrams of quercetin equivalents per 100 g of dry matter (mg QE/100 g DM).

2.3. Mycotoxins Analysis

The modified liquid chromatography tandem mass spectrometry (LC-MS/MS) method established by Hofmann and Scheibner [41] was used for the quantification of mycotoxins in the collected wheat cultivars. Briefly, this method enabled the simultaneous monitoring of 22 mycotoxins. For this purpose, the following chemicals were used: methanol (MeOH) (Carlo Erba, Val de Reuil, France), water (Carlo Erba, Val de Reuil, France), and formic acid (Fluka Analytical, Sigma-Aldrich, Steinheim, Germany) of LC-MS/MS grade, and acetonitrile (ACN) (Fisher Scientific, Geel, Belgium) of HPLC grade. With the exception of T-2 toxin (100 μg/mL), which was acquired from LGC Standards (Wesel, Germany), mycotoxin standards were obtained as follows: DON (100 μg/mL), fumonisins (FUs) (FB1 and FB2, each 50 μg/mL), HT-2 toxin (100 μg/mL), ZEN (100 μg/mL), TEN (100 μg/mL), alternariol (5 mg) and alternariol monomethyl ether (5 mg) from Fluka Analytical (Steinheim, Germany); MON (100.4 μg/mL), mixed standards of six ergot alkaloids (100 μg/mL of each), and a mixed aflatoxins standard (Afs) (2 μg/mL of AFB1 and AFG1 and 0.5 μg/mL of AFB2 and AFG2) from Biopure, Romerlab (Tuln Austria). MeOH/water (50/50, v/v) was used to dissolve a mixed standard stock solution of 15 distinct mycotoxin standards. Calibration solutions were prepared by spiking matrix extracts with a mixture of standards. The individual standards were dried down and reconstituted in MeOH/water (50/50, v/v). Eleven calibration levels were used, ranging from 0.01 to 200 ug/kg, with correlation coefficients (r²) greater than 0.9915. All mycotoxin standards and stock standard solutions were stored at −18 °C until analysis.

2.3.1. Sample Preparation for LC-MS/MS Analysis

Sample preparation for LC-MS/MS analysis followed the protocol of Hofmann and Scheibner [41] with slight modifications. HPLC-grade acetonitrile purchased from Fisher Scientific (Geel, Belgium) and water purified using an Adrona Water Purification system (Riga, Latvia) were used for sample preparation. In brief, 5 g of ground wheat sample was placed into a 50 mL polypropylene tube and extracted with acetonitrile/water (80/20, v/v). During extraction, samples were shaken on a horizontal shaker (Biosan, Latvia) at 350 rpm for 60 min and centrifuged (Boeco, Hamburg, Germany) at 4000 rpm for 5 min. A 400 μL aliquot of the supernatant was diluted with 600 μL MeOH/water (50/50, v/v) and filtered through a 0.2 µm polytetrafluoroethylene (PTFE) disposable syringe filter into vials.

2.3.2. LC-MS/MS Analysis

An HPLC Vanquish Core system equipped with the Hypersil GOLD C18 Selectivity HPLC column (100 × 2.1 mm i.d., 1.9 μm particle size), coupled to a TSQ Quantis Triple Quadrupole mass spectrometer with a heated electrospray ionization (HESI) source (all from Thermo Fisher Scientific, Waltham, MA, USA), was used for the detection and quantification. The analysis was carried out as described in the ThermoFisher Scientific Application Note 65,969 [41], with several modifications. Briefly, the mobile phase consisted of water (95%) and methanol (5%), each containing 0.1% formic acid. The autosampler tray was maintained at 20 °C, and the column was set to 40 °C, with an injection volume of 10 μL. The retention times and the SRM transitions for the analyzed compounds are summarized in

Table 3. Retention times and SRM transitions for the different mycotoxins.

Compound	RT [min]	Polarity	Precursor m/z	Quantifier m/z	Collision Energy	Qualifier 1 m/z	Collision Energy	Qualifier 2 m/z	Collision Energy
Aflatoxin B1	9.47	+	313.06	285.07	23.08	241.00	37.22	270.03	27.40
Aflatoxin B2	9.16	+	315.06	287.07	26.07	243.01	39.68	271.08	32.22
Aflatoxin G1	8.56	+	329.01	243.07	26.98	200.04	41.31	215.07	32.97
Aflatoxin G2	8.41	+	331.05	313.04	24.33	189.04	42.30	285.01	27.93
Alternariol	10.61	−	257.05	213.00	22.85	214.97	25.35	146.99	32.63
Alternariolmethylether	11.75	−	271.05	255.99	22.25	228.04	29.14	212.95	36.84
Deoxynivalenol	1.01	+	297.04	249.07	10.23	231.07	13.53	175.04	19.86
Diacetoxyscripenol	9.81	+	367.09	307.07	10.23	349.05	10.23	229.05	12.73
Ergocornine	9.72	+	562.30	268.14	25.20	208.07	42.91	223.08	36.23
Ergocristine	10.30	+	610.30	268.16	26.23	208.08	43.66	348.10	25.62
Ergocryptine	10.25	+	576.30	268.15	25.77	208.07	44.23	223.08	36.84
Ergosine	9.47	+	548.30	208.07	40.82	223.08	33.62	268.16	24.52
Fumonisin B1	10.54	+	722.50	334.01	42.72	352.50	37.98	704.39	29.90
Fumonisin B2	11.5	+	706.30	336.22	38.89	688.45	27.63	354.15	34.64
Fusarenon X	0.98	+	355.09	229.07	16.56	175.00	22.09	277.00	12.39
HT-2 toxin	10.83	+	447.20	345.08	19.25	285.07	20.84	255.13	23.04
Moniliformin	0.94	−	96.85	41.03	13.18
Ochratoxin A	11.68	+	404.00	238.97	24.18	358.07	14.78	220.97	14.78
Sterigmatocystin	11.80	+	325.10	310.00	24.71	281.00	38.09	253.06	44.95
T-2 toxin	11.34	+	489.21	245.13	28.05	327.04	24.10	387.05	22.36
Tentoxin	10.99	−	413.18	271.16	16.10	141.00	20.05	369.24	18.49
Zearalenone	11.65	−	317.09	175.00	24.03	131.07	29.30	273.16	19.21

. Nitrogen was used as the sheath, auxiliary, and sweep gas, set to 30, 6, and 1 arbitrary units (Arb), respectively. Argon at 1.5 mTorr served as the collision-induced dissociation (CID) gas. The ion transfer tube and vaporizer temperatures were adjusted to 325 °C and 350 °C, respectively. The duration of each cycle was 0.5 s. Data acquisition and processing were carried out using TSQ Quantis 3.2 Tune and TraceFinder 5.1 (ThermoFisher Scientific, Waltham, MA, USA).

2.3.3. Method Validation

Method validation was performed in accordance with the Technical Report CEN/TR 16,059 [42] and the European Regulation, No 2002/657/EC [43]. Therefore, the performance of the applied LC-MS/MS method was evaluated in terms of linearity, recovery, repeatability, reproducibility, limit of quantification (LOQ), and matrix effects. The matrix effect was calculated as the signal suppression/enhancement ratio (SSE), expressed as the slope ratio of the matrix-matched calibration curve to solvent-based calibration curve. To distinguish between extraction efficiency and matrix-induced signal suppression/enhancement, the slope ratios of the linear calibration functions were computed to determine the apparent recovery (RA), representing the overall method recovery, and the signal suppression/enhancement (SSE) attributable to matrix effects. The extraction recovery (RE), or sample preparation recovery, was calculated by dividing the overall recovery by the matrix effect (

Table 4. Matrix effect (SSE) and Recovery data of the employed analytical method based on the solvent (R_A) and matrix-matched (R_E) calibration curves.

Mycotoxin	Spiking Level (μg/kg) a	SSE (%) b	RA (%) c	RE (%) d
ZEA	0.1–150	83.9	75.8	90.3
T-2 toxin	0.2–250	92.6	88.4	95.5
HT-2 toxin	0.1–100	101	93.0	91.8
MON	0.1–25	91.6	79.7	86.9
TEN	0.1–40	114	110	95.7
Ergocryptine	0.2–40	109	108	98.8

Abbreviations: ZEA—Zearalenone; MON—Moniliformin; TEN—Tentoxin; ^a Concentration range of analytes for standard, matrix-matched calibration curves and calibration curves of spiked samples (μg/kg); ^b SSE-matrix effect (%) calculated by the slope of matrix-matched calibration curve/slope of the solvent calibration curve; ^c R_A—Apparent recovery (%) calculated by the slope of spiked sample-prepared curve/slope of the solvent calibration curve; ^d R_E—Sample preparation recovery (%) calculated by the slope of spiked sample-prepared curve/slope of matrix-matched calibration curve.

). The repeatability and reproducibility were assessed for six detected mycotoxins by analyzing blank wheat samples spiked at six concentration levels, each in six replicates (

Table 5. Precision data of the employed LC-MS/MS method for detected mycotoxins.

Mycotoxin	Spiking Level (µg/kg)	RSD (%) (n = 6) a	RSDs (%) (n = 3 × 6) b
ZEA	0.1 (LOD) *	11.5	16.7
	1 (LOQ) **	9.32	14.0
	20	9.06	12.9
	50	8.96	11.3
	75	8.03	9.96
	150	5.92	8.65
T-2 toxin	0.2 (LOD)	14.0	19.9
	1 (LOQ)	9.90	16.0
	25	6.72	13.6
	75	6.51	11.8
	100	6.23	11.0
	250	3.30	7.08
HT-2 toxin	0.1 (LOD)	9.49	14.2
	2 (LOQ)	9.35	14.1
	25	8.70	11.0
	75	5.50	9.92
	100	3.77	9.34
	500	2.78	8.14
MON	0.1 (LOD)	12.9	19.7
	1.0	10.7	16.8
	2.5 (LOQ)	9.68	12.7
	5	9.56	12.2
	10	8.88	10.3
	25	8.53	9.89
TEN	0.5 (LOD)	14.8	16.3
	0.75	13.2	11.7
	1 (LOQ)	12.0	14.9
	20	7.30	13.9
	30	7.31	13.8
	40	7.06	10.2
Ergocryptine	0.2 (LOD)	10.8	19.8
	0.3	8.56	17.6
	0.4	7.73	13.9
	1 (LOQ)	6.95	9.87
	30	5.94	9.24
	40	4.03	7.49

Abbreviations: ZEA—Zearalenone; MON—Moniliformin; TEN—Tentoxin; * LOD—limit of detection; ** LOQ—limit of quantification; ^a RSD (%)—relative standard deviation of 6 replicates at six concentration levels using the spiked white wheat flour and the matrix-matched calibration (MMC) curve; ^b RSDs (%)—relative standard deviation of 6 replicates at six concentration levels using the spiked white wheat flour and the MMC curve, over the course of three days, using the same instrument and by the same operators.

). Limits of detection (LOD) and quantification (LOQ) were estimated by injecting decreasing concentrations of ma-trix-matched standards and measuring signal-to-noise (S/N) ratios of ≥3 and ≥10, respectively. The established LOD and LOQ fulfil the validation criteria for recovery and repeatability (Table 5).

2.4. Statistical Analysis

Experimental data related to technological parameters and antioxidant activity were subjected to one-way analysis of variance (ANOVA), supplemented by Tukey’s Honestly Significant Difference (HSD) test, to discern notable differences at a significance level of p < 0.05. ANOVA analyses were performed using InfoStat software, Version 2016e (UNC, Córdoba, Argentina; trial version). The relationship among the studied parameters was further evaluated using Pearson’s correlation analysis, with correlations considered statistically significant at p < 0.05. Principal Component Analysis (PCA) was also performed to identify patterns and groupings among the variables. These were conducted using XLSTAT BASIC, version 5.1, 2022 software (Addinsoft, Paris, France).

3. Results

The 2023/24 growing season was the warmest on record in Serbia. According to data from the local meteorological station in Kanjiža, the autumn-winter period (October–December) was characterized by temperatures considerably above the long-term average (Figure 1). The total amount of precipitation during this period was about 10% lower than the multiannual average (378 mm vs. 410 mm), with substantial variability in distribution. Precipitation in October was markedly below average, hindering soil tillage and crop emergence, whereas November recorded the highest precipitation total in recent decades, replenishing soil moisture reserves. Unusually warm weather persisted throughout the spring vegetation period in 2024, accompanied by below-average precipitation, particularly in February and March.

Physical, chemical, and functional characteristics of the investigated wheat cultivars are summarized in

Table 6. Physical, Chemical and Functional Properties of Wheat Cultivars.

Cultivars	GY	TW	PC	WG	GI	WA	FR	SD	W	L	P/L	TPC	TFC
NS Epoha O	3.14 b ± 0.31	74.5 c ± 0.78	14.1 e ± 0.12	35.1 d ± 0.17	100.0 a ± 0.00	57.3 e ± 0.10	3.5 e ± 0.7	55 b ± 8	296 a ± 38	117 ab ± 30	0.63 ab ± 0.17	119.09 ab ± 19.7	116.81 ab ± 0.83
Genotype 1 O	2.80 b ± 0.30	76.0 bc ± 0.60	17.6 bc ± 0.42	53.9 a ± 0.28	64.8 f ± 0.31	64.9 b ± 0.14	7.5 d ± 0.9	85 a ± 5	184 bc ± 15	98 ab ± 28	0.73 ab ± 0.25	129.94 ab ± 8.48	111.71 ab ± 2.22
Bánkúti 1201 O	1.95 c ± 0.21	77.7 ab ± 0.38	18.1 abc ± 0.14	53.7 a ± 0.12	74.1 de ± 0.78	63.5 c ± 0.30	9.0 cd ± 0.3	55 b ± 7	198 bc ± 15	105 ab ± 15	0.60 ab ± 0.12	121.92 ab ± 1.79	123.70 a ± 1.27
Genotype 2 O	2.75 b ± 0.29	76.4 bc ± 0.53	15.5 d ± 0.39	39.4 c ± 0.30	96.7 b ± 0.46	57.4 e ± 0.28	15.0 a ± 0.4	0 d ± 0	187 bc ± 10	71 b ± 11	0.91 ab ± 0.22	118.71 ab ± 0.17	116.00 ab ± 1.89
Genotype 3 O	2.45 bc ± 0.27	77.1 b ± 0.42	18.9 a ± 0.16	54.4 a ± 0.57	73.0 e ± 0.28	67.2 a ± 0.17	10.0 c ± 0.0	50 b ± 9	218 abc ± 27	98 ab ± 26	0.72 ab ± 0.20	126.94 ab ± 8.14	119.42 ab ± 3.66
Genotype 4 O	2.50 bc ± 0.25	74.6 c ± 0.71	18.3 ab ± 0.44	54.4 a ± 0.33	75.3 d ± 0.39	62.4 d ± 0.25	9.5 c ± 0.5	45 b ± 4	194 bc ± 10	128 a ± 21	0.41 b ± 0.09	154.48 a ± 9.13	122.22 ab ± 2.10
Genotype 5 O	2.45 bc ± 0.25	79.3 a ± 0.45	17.0 c ± 0.30	49.1 b ± 0.14	72.7 e ± 0.42	63.7 c ± 0.29	12.5 b ± 0.6	20 c ± 0	244 ab ± 70	78 ab ± 23	1.12 a ± 0.47	126.99 ab ± 1.04	116.06 ab ± 5.13
Zvezdana C	6.80 a ± 0.37	72.1 d ± 0.40	12.0 f ± 0.28	26.5 e ± 0.42	82.4 c ± 0.85	55.5 f ± 0.21	3.0 e ± 0.2	50 b ± 0	171 bc ± 44	90 ab ± 17	0.56 ab ± 0.06	115.68 b ± 10.4	104.75 b ± 7.05
NS Epoha C	6.95 a ± 0.38	75.6 bc ± 0.57	12.0 f ± 0.13	26.2 e ± 0.26	83.5 c ± 0.71	57.6 e ± 0.28	3.0 e ± 0	95 a ± 3	158 c ± 7	65 b ± 10	1.09 a ± 0.21	117.24 ab ± 9.17	114.18 ab ± 8.66

Abbreviations: O—organic production; C—convectional production; GY—grain yield (t/ha); TW—test weight (kg/hl); PC—protein content (% on dry matter); WG—wet gluten content (%); GI—gluten index (%); WA—water absorption (%); FR—farinograph resistance (min); SD—softening degree (BU); W—alveograph deformation energy (10⁻⁴ J); L—extensibility (mm); P/L—over-pressure/extensibility ratio; TPC—total polyphenols content (mg/100 g DM) and TFC—total flavonoid content (mg/100 g DM), all results are expressed as mean value ± SD. Means with at least one common letter were not significantly different at p = 0.05 by Tukey’s test.

. The NS Epoha cultivar exhibited nearly a twofold statistically higher yield under conventional than under organic production, which was an expected outcome (Table 6). Among the tested cultivars, NS Epoha achieved the highest GY in organic production (3.14 t ha⁻¹), with values ranging between 1.95 and 3.14 t ha⁻¹. This is in line with the findings of Földi et al. [33], who reported similar levels in 2018. Despite a lower yield, organic production resulted in wheat with superior quality traits. The PC of five other wheat varieties grown organically reached 17% or more and was statistically higher than other two cultivars from organic production, which is considered high even for organic production, given that Földi et al. [33] reported PC levels of about 15% in both examined years. Only one wheat variety of premium quality from organic production (Genotype 2) showed slightly lower PC—15.5%. The PC of organically produced premium wheat varieties was higher than average PC of bread wheat conventionally produced in Martonvásár and Serbia; 15.2 and 12.9%, respectively [44]. A similar trend was observed for WG and WA among wheat cultivars from organic production, with five varieties exhibiting values above 49% and 62%, respectively. The values of these parameters with five organically grown varieties were statistically higher from other cultivars. Among organically grown wheat cultivars, FR was markedly higher in six cultivars, whereas NS Epoha O was an exception, exhibiting a value of 3.5 min, closer to the values observed in the remaining cultivars from conventional production. The value of this parameter in NS Epoha O did not differ significantly from that of conventionally grown varieties. SD of Genotype 2 O and Genotype 4 O from organic production was significantly lower that of other organically and conventionally grown varieties. Five organic cultivars exhibited a GI below 80%, that was significantly lower than organic cultivars NS Epoha and Genotype 2 showed values of 100% and 96.7%, respectively. Conventional cultivars had a GI higher than 80% significantly lower than organic cultivars NS Epoha and Genotype 2. W was generally higher for organic wheat cultivars, ranging from 184 × 10⁻⁴ J to 296 × 10⁻⁴ J, whereas cultivars from conventional farming showed values below 172 × 10⁻⁴ J. Regarding extensibility, organic cultivars generally exhibited higher values, with three cultivars exceeding 100 mm, while the others ranged from 71 to 98 mm. Among the conventionally grown cultivars, Zvezdana showed the highest extensibility (90 mm), whereas NS Epoha C had lower value of 65 mm. There was no statistical differentiation of this parameter between varieties from organic and conventional production. The P/L for almost all wheat varieties was between 0.6 and 1, which, according to the Italian quality regulation, classifies these varieties between good and improvers [45]. Considering antioxidant activity, organic cultivars generally had higher polyphenol and flavonoid content. Among them, the organic cultivar Genotype 4 was particularly prominent, with total phenolic content (TPC) and total flavonoid content (TFC) of 154.48 and 122.22 mg/100 g DM, respectively, indicating a particularly strong antioxidant potential. There was no statistical differentiation of these two parameters between varieties from organic and conventional production.

The correlations among the physicochemical properties of wheat grains, their rheological behaviour, and antioxidant activity are presented in

Table 7. Correlation among wheat grains’ physicochemical properties, rheological behaviour with antioxidant and polyphenolic activity. Correlation matrix (Pearson).

Variables	GY	TW	PC	WG	GI	WA	FR	SD	P	L	W	P/L	TPC	TFC
GY	1	−0.6196	−0.8895	−0.8700	0.2204	−0.6796	−0.7122	0.4374	−0.2094	−0.4559	−0.6182	0.2357	−0.4554	−0.7153
TW		1	0.5955	0.5839	−0.3552	0.6518	0.6695	−0.2664	0.6899	−0.1852	0.4262	0.4587	0.0428	0.5471
PC			1	0.9918	−0.5817	0.8979	0.6561	−0.2234	0.0654	0.6059	0.3721	−0.3758	0.6313	0.6883
WG				1	−0.6470	0.9084	0.6035	−0.1512	0.0661	0.6299	0.3505	−0.3903	0.6497	0.6502
GI					1	−0.7801	−0.1372	−0.3326	0.0059	−0.4107	0.1458	0.2627	−0.4497	−0.0966
WA						1	0.4428	0.0727	0.2760	0.4492	0.3149	−0.1841	0.4893	0.5224
FR							1	−0.7578	0.1435	0.0096	0.2532	0.0799	0.2918	0.4512
SD								1	0.0339	0.0904	−0.3838	−0.0279	−0.0426	−0.1690
P									1	−0.5395	0.5667	0.8046	−0.3052	0.1265
L										1	0.1220	−0.8967	0.7739	0.4519
W											1	0.1223	0.0926	0.3818
P/L												1	−0.5330	−0.1631
TPC													1	0.4795
TFC														1

Abbreviations: GY—grain yield; TW—test weight; PC—protein content; WG—wet gluten content; GI—gluten index; WA—water absorption; FR—farinograph resistance; SD—softening degree; P—over-pressure; L—extensibility; W—alveograph deformation energy; P/L—over-pressure/extensibility ratio; TPC—total polyphenols content and TFC—total flavonoid content, Values in bold are different from 0 with a significance level alpha = 0.05.

. The study revealed that the two key parameters of wheat, GY and PC, were strongly negatively correlated, as previously reported by Silva et al. [46] and Živančev et al. [30]. GY also showed a strong negative correlation with WG, consistent with the findings of Soorninia et al. [47], as well as with farinograph parameters, WA and FR. In contrast, PC was expectedly strongly positively correlated with end-use quality parameters WG and WA, which themselves exhibited a statistically significant positive correlation. WA was strongly negatively correlated with GI, indicating that wheat varieties with high WA values tend to have softer or weaker WG, whereas TW showed a statistically significant positive correlation with FR and P. As expected, alveograph parameters P and L were strongly correlated with the P/L ratio, with P positively and L negatively associated with it. Finally, TFC exhibited a strong positive correlation with PC and a negative correlation with GY.

To investigate the influence of key physical, chemical, functional, and antioxidant properties of wheat cultivars grown under organic and conventional farming systems, a comprehensive principal component analysis (PCA) was conducted (Figure 2). The first principal component (PC1) accounted for more than 45% of the total variability, while the second principal component (PC2) explained over 24%.

PC was strongly positivelly correlated with WG, WA, and TFC, while strongly negatively correlated with GY, which confirm values of F1 factor (

Table 8. Loading values of first two factors from PCA biplot of the physicochemical, rheological behaviour, and antioxidant properties of wheat grains.

	F1	F2
GY	−0.9083	−0.1832
TW	0.6374	0.6869
PC	0.9876	−0.0544
WG	0.9811	−0.0921
GI	−0.5453	0.2803
WA	0.8728	0.0242
FR	0.6678	0.3891
SD	−0.2691	−0.3556
P	0.1192	0.8647
L	0.5985	−0.7477
W	0.4535	0.4675
P/L	−0.3281	0.8744
TPC	0.6608	−0.4938
TFC	0.7388	0.0834

Abbreviations: GY—grain yield; TW—test weight; PC—protein content; WG—wet gluten content; GI—gluten index; WA—water absorption; FR—farinograph resistance; SD—softening degree; P—over-pressure; L—extensibility; W—alveograph deformation energy; P/L—over-pressure/extensibility ratio; TPC—total polyphenols content and TFC—total flavonoid content.

). TPC and L were positively correlated with each other but negatively correlated with GI and P/L ratio. Also, W exhibited a positive correlation with FR and TW, while displaying a negative correlation with farinograph SD. Furthermore, TPC and showed no association with W, TW, and FR, whereas GY was unrelated to the P/L ratio.

According to the PCA biplot, two wheat varieties from conventional production (Zvezdana and NS Epoha) clustered closely with GY. In contrast, NS Epoha O and Genotype 2 wheat varieties from organic production demonstrated close associations with GI and P/L ratio. The old Hungarian variety Bánkúti 1201 and the Austrian variety Genotype 3 were strongly associated with PC with WG, WA, and TFC, while Genotype 5 was closely linked to P, TW, and W. Additionally, the PCA biplot revealed that the varieties Genotype 1 and Genotype 4 were characterized by high values of L parameter. Mycotoxin contamination levels in the examined wheat samples are presented in

Table 9. Mycotoxins in wheat cultivars from organic production and conventional production.

Sample	Mycotoxin Content (µg/kg)
	ZEA	T-2 Toxin	HT-2 Toxin	MON	TEN	Ergocryptine
NS Epoha O	ND	ND	ND	7.4	ND	18.8
Genotype 1 O	ND	0.7603	2.26	ND	ND	ND
Bánkúti 1201 O	ND	0.6716	0.17	61.4	ND	ND
Genotype 2 O	ND	0.8289	3.24	37.6	ND	ND
Genotype 3 O	4.78	0.8283	3.23	21.3	25.6	19.0
Genotype 4 O	ND	0.6398	ND	ND	ND	ND
Genotype 5 O	ND	ND	ND	9.47	7.39	18.9
Zvezdana C	ND	ND	ND	ND	15.2	ND
NS Epoha C	ND	ND	ND	65.5	ND	ND

Abbreviations: ZEA—Zearalenone; MON—Moniliformin; TEN—Tentoxin; ND—not detected.

. To minimize matrix effects on the detected mycotoxin concentrations, quantification was performed using an external matrix-matched calibration procedure. Among the 22 mycotoxins analyzed, only six were detected and/or quantified in selected wheat varieties from both organic and conventional farming systems. The warm and dry conditions observed in February and March 2024 (elevated temperatures and below-average precipitationled to wheat flowering (heading) occurring 15–20 days earlier than usual (in the first half of April instead of early May). These weather conditions werelikely unfavourable for the development of toxigenic filamentous fungi in wheat. Consequently, among Fusarium toxins, only ZEA, T-2, HT-2, and MON were detected and/or quantified, while among Alternaria toxins, only TEN was quantified. Regarding ergot alkaloids (Claviceps toxins), only ergocryptine was quantified in wheat from the organic farming system, whereas no ergot alkaloids were detected in conventionally produced wheat. Wheat varieties Zvezdana C and Epoha C from conventional production were contaminated with only one of the examined mycotoxins (TEN and MON, respectively). On the other hand, wheat varieties from the organic farming system were contaminated with at least two of the examined mycotoxins. The results indicate that the Austrian variety Genotype 3 was contaminated with all detected mycotoxins (Table 9), indicating a higher susceptibility to fungal infections under organic farming conditions, leading to increased mycotoxin contamination.

To comprehensively assess mycotoxin content across various wheat cultivars grown under organic and conventional farming systems, the results were subjected to a robust compositional PCA. The first principal component (PC1) accounted for over 43% of the total variability, while the second principal component (PC2) nealry 29% (Figure 3).

T-2 toxin exhibited a positive correlation with HT-2 toxin confirming the values of the F1 factor (

Table 10. Loading values of first two factors from PCA biplot on mycotoxin profiles.

	F1	F2
ZEA	0.9448	−0.0623
T-2 toxin	0.5008	0.7825
HT-2 toxin	0.7382	0.5412
MON	−0.1927	0.4634
TEN	0.8074	−0.3718
Ergocryptine	0.4977	−0.6894

Abbreviations: ZEA—Zearalenone; MON—Moniliformin; TEN—Tentoxin.

). This relationship is expected, as both toxins are primarily produced by Fusarium sporotrichioides [14]. TEN (an Alternaria toxin) showed a positive correlation with ZEA, that confirm values of F1 factor, (a Fusarium toxin) and ergocryptine (a Claviceps toxin), while ergocryptine displayed a negative correlation with MON (a Fusarium toxin). Furthermore, T-2 toxin was corelated with TEN, while HT-2 toxin showed no association with MON or ergocryptine.

Two wheat varieties from organic production (Genotype 5 and NS Epoha O) demonstrated close associations with ergocryptine, while one conventional variety (Zvezdana C) was closely associated with TEN. Further, the old Hungarian variety Bánkúti 1201, the Austrian variety Genotype 2, and Genotype 4 from organic production, as well as NS Epoha from conventional production, were closely associated with MON. Conversely, the Hungarian variety Genotype 1 and the Austrian variety Genotype 2 demonstrated close associations with T-2 and HT-2 toxins, while Genotype 3 was closely linked to ZEA and TEN.

4. Discussion

Our preliminary survey conducted in Serbia highlights the potential of organic wheat farming in terms of end-use quality. This observation contrasts with the findings of Takač et al. [44], who reported that protein content, wet gluten content, Zeleny sedimentation value, and farinograph parameters of bread wheat cultivars were superior in conventional production compared to organic farming in Hungary. This suggests that the relative performance of organic versus conventional wheat may be strongly influenced by regional agroclimatic conditions and local management practices. Under organic farming conditions in 2024, the Serbian control variety NS Epoha demonstrated broad adaptability in both yield and quality, despite being primarily classified as a milling-quality wheat cultivar. The average GY in 2024 was comparable to the values reported by Földi et al. [33] for the dry season of 2018. This outcome may be explained by significant variability in weather conditions, particularly high temperatures during the growing season, the early onset of key phenological stages (stem elongation and heading), and the late frost events, which collectively resulted in yields lower than the long-term average for the studied agroecological region. These conditions accelerated phenological development and strongly influenced the final grain yield. Elevated temperatures resulted in an unusually early onset of stem elongation (early March instead of early April) and heading (first half of April instead of early May), a shift not previously recorded in recent years. Consequently, heading occurred 15–20 days earlier, and plants displayed greater vigour and more advanced development compared to typical seasons. Between 20 and 26 April, minimum night temperatures dropped below 0 °C on three occasions, creating a substantial risk of partial spike (flower) damage, which may have further contributed to yield reductions. This stress period coincided with a slight soil moisture deficit, compounding the negative effects on reproductive development. Favourable warm and rainy conditions returned in late May and persisted through June, promoting grain filling and ripening; however, these improvements were insufficient to fully offset the earlier cold-induced stress events. Generally, the protein and wet gluten contents from organic production were similar to those reported by Takač et al. [44] for spelt wheat cultivars. The excessively high WG content observed in most wheat varieties from organic production led to a significant shift in the glutenin-to-gliadin ratio, favouring gliadin [48]. Although these varieties exhibited good dough mixing properties according to farinograph measurements, the altered gluten composition negatively affected their alveograph performance. This can be explained by the fact that glutenins primarily contribute to dough elasticity, while gliadins are responsible for extensibility, as previously mentioned. This observation was further confirmed by the PCA biplot (Figure 2), where WG clustered closely with FR and showed a strong negative correlation with the GI, a parameter reflecting the proportion of weak versus strong components of wet gluten. The high WG content can likely be attributed to the type of fertilizer used in organic production, as well as to favourable soil nutrient availability, particularly nitrogen (Table 2). Moreover, the bread-making quality of two wheat varieties from conventional production was below the three-year average reported by Živančev et al. [31]. Contrary to the findings of Wang et al. [22], who reported significant differences in TPC and TFC between organic and conventional wheat in the UK and Germany, no such differences were observed in this study, except in one organic variety. However, the relatively stable antioxidant activity across production systems in our study suggests that environmental factors, such as seasonal fluctuations in temperature and precipitation, may exert a stronger influence on phenolic accumulation than the farming system itself. This observation aligns with conclusions of Baranski et al. [21], who emphasized that year-to-year climatic variability often outweigh production practices in determining polyphenol levels. It is also noteworthy that one organic wheat variety exhibited slightly higher TPC and TFC values, highlighting the influence of genetic background in modulating antioxidant capacity. Nevertheless, our findings do not exclude the potential of organic farming to enhance antioxidant levels, as certain varieties may benefit more from organic practices under specific conditions.

Studies comparing mycotoxin levels in organic and conventional wheat have shown mixed results, although most findings are favourable for organic production. Several investigations have indicated that organic wheat generally contains considerably lower levels of Fusarium mycotoxins compared to conventional wheat. Remža et al. [49] found that organic wheat samples in Slovakia contained substantially lower concentration of DON and ZEA than conventional samples. Similarly, Schneweis et al. [50] reported that conventionally cultivated wheat in Germany exhibited a higher frequency of Fusarium contamination and elevated concentrations of ZEA and DON compared to organic wheat. Rossi et al. [51] likewise confirmed that organic wheat produced in Italy had lower DON levels than conventional wheat, while ochratoxin A concentration were low and did not differ significantly between production systems. Edwards [52] observed no significant disparity in DON and ZEA concentrations between organic and conventional wheat samples in the UK; however, organic samples exhibited markedly lower levels of HT-2 and T-2 toxins. On the other hand, Pussemier et al. [53] investigated the DON, OTA, and ZEN content in organic and conventional Belgian wheat samples, as well as wholemeal wheat flour samples from Belgian retail shops. The authors emphasized that the conventional wheat tended to contain higher levels of the examined mycotoxins compared to organic wheat, while the opposite trend was observed for wholemeal wheat flour, where organic samples were more frequently contaminated with DON and OTA than conventional ones. A review by Bernhoft et al. [54] examines the effects of organic versus conventional cereal production methods on Fusarium head blight and mycotoxin contamination. The authors reported that organic cereals contained lower levels of Fusarium mycotoxins, with DON, ZEA, and T-2/HT-2 concentration being 62%, 110%, and 180% higher, respectively, in conventional cereals. Wang et al. [22] reported that organic flour exhibited ZEA concentrations about 9% higher than conventional flour, while the Fusarium mycotoxin levels were roughly ten-fold lower than the EU maximum limits, implying that both production systems meet acceptable safety standards. In our study, wheat varieties from organic farming systems were more frequently contaminated with various mycotoxins than those from conventional farming systems; however, the detected concentrations of mycotoxins remained relatively low (Table 8). These findings likely reflect the combined effects of the wheat varietal characteristics, the implemented agricultural practice, and, most importantly, the weather conditions during the growing season. In addition to the temperature variation described above, an analysis of precipitation during May—the critical period for wheat anthesis and potential Fusarium infection—showed that in several seasons (2022–2025) rainfall remained below or close to the long-term average (Figure S1). No excessive or prolonged wet periods were recorded during flowering, which substantially reduces the likelihood of fungal infection in these years. Although intra-seasonal variability was pronounced, the absence of high humidity during the sensitive phenological stage suggests that natural infection pressure was relatively low. Future multi-year trials including seasons with wetter flowering periods will allow a more comprehensive assessment of genotype × environment interactions related to fungal infection. A major limitation of previously published data has been related to the difficulty of isolating the influence of the production system from other confounding factors, such as genotype, climate, and storage conditions. Consequently, the reported results are often contradictory or lack general applicability. Moreover, a truly interdisciplinary approach is still lacking-agronomy, microbiology, toxicology, and food technology need to be more closely integrated to enable a comprehensive assessment of mycotoxin occurrence and risk in cereal production.

5. Conclusions

The results indicate that the weather conditions were a decisive factor influencing wheat development and yield. Organic production resulted in lower yields but superior technological and grain quality characteristics. Yields in organic production were approximately two-fold lower than in conventional production, but the quality parameters of five organically grown varieties (PC > 17%, WG > 49%, WA > 62%) were significantly higher than the standard for organic wheat. The elevated WG content in organic varieties was accompanied by lower GI, indicating a higher proportion of the gliadin fraction in gluten. No consistent differences in TPC and TFC were observed between organic and conventional systems, except for one organic variety. This suggests that seasonal climatic factors (e.g., temperature and precipitation) exert a stronger influence on the accumulation of phenolic compounds than the production system itself. Contamination with mycotoxins was more frequent in organic production but remained at low concentrations. The absence of DON and the generally low mycotoxin levels were most likely due to unfavourable conditions for Fusarium development during early spring (warm and dry during weather coinciding with wheat anthesis).