Organic Cropping System A ﬀ ects Grain Chemical Composition, Rheological and Agronomic Performance of Durum Wheat

: Durum wheat plays a central role for Italy, a country characterized by an historical tradition of pasta making. This crop is one of the major sources of starch in the Mediterranean diet, also providing substantial amounts of nutrients, either essential or healthy, such as minerals and phytochemicals. In this study, the e ﬀ ect of cropping system (organic (ORG) and conventional (CONV)) on the content of proteins, wet gluten, minerals, and total polyphenols (TPC), as well as on the antioxidant activity (AA) and yield characteristics were evaluated in nine genotypes of whole durum wheat. The analysis of variance conﬁrmed the key role of cropping system, which explained more than 50% of the total variation for some minerals (Na, K, and Fe), and more than 80% for proteins, wet gluten, and TPC. The CONV reported the highest levels of proteins, wet gluten, and ash, whereas the TPC and AA were statistically higher in the ORG ones. The quality traits under study were also genotype dependent. For instance, ‘Ramirez’ had a higher ash content under ORG than CONV, while ‘Core’ had an opposite behavior. This study provides a better understanding about the e ﬀ ect of both cropping system (ORG vs. CONV) and genotype on the yield and qualitative traits of the whole grain durum wheat.


Introduction
Wheat is ranked fourth in the top 50 commodities of FAO list (http://faostat3.fao.org/browse/ rankings/commodities_by_regions/E). Durum wheat (Triticum turgidum L. ssp. durum) represents about 8% of the total world wheat production [1]. It is one of the most widely grown food crops in the Mediterranean area: it can be used as it is (as in the case of cuccìa), or processed as semolina for the production of pasta, couscous, bulgur, and bread [2]. In the last 10 years in Italy, where there is an historical tradition of pasta making and the crop has a central role in the human diet and economy [3], 38,585 tonnes of wheat were produced [4].
Wheat not only is a major source of starch and energy in the diet, but also provides substantial amount of several compounds which are essential or beneficial for health, notably proteins, vitamins (mainly B ones), dietary fiber, minerals and phytochemicals, that are concentrated in the bran and germ components of whole grains [5,6]. In particular, whole grain products are recommended for healthy diets being recognized sources of dietary fiber and antioxidant compounds [7]. The former, especially phenolic compounds, are located in the outer layer of cereal seeds, both in free and insoluble forms. Actually, the health properties of whole grain appear to be dependent on a synergism among its
The local climate is semiarid-Mediterranean with mild winter and hot, rainless summer. We used two adjoining fields, one for each cropping system under study, in the same area to guarantee the same pedoclimatic conditions for both types of cropping systems. These experimental fields were separated by rows of trees, resulting in a total distance of 50 m. In addition, they had the same crop rotation, involving durum wheat and filed bean. In each cropping system (CONV and ORG), a randomized block design with four replications was adopted. Each block included the nine genotypes reported in Table 1. In both cropping systems, seeds were sown at a density of 350 viable seeds m −2 in December 2015. The size of single plot was 4 × 6 m and consisted of 20 rows, each 6 m long and spared 0.20 m. Tillage consisted of a 40 cm depth ploughing by harrowing in October. The CONV cropping system was managed in accordance with the local farming practices, while the crop management under ORG cropping system followed current EU regulations (889/2008). The same fertilizer doses were applied under CONV and ORG cropping systems, namely a pre-sowing application of 50 kg ha −1 N and 100 kg ha −1 of P 2 O 5 in the form of synthetic fertilizer (CONV) or a mixture of feathers and torrefied bone and meat meal (ORG). A further 30 kg ha −1 N was supplied at the full tillering (stage 25) [21] as urea (CONV) or a mixture of dried manure and hydrolyzed pelt (ORG). In the CONV cropping system the weeding was performed with MPPA at the first detectable node (stage 31) [21], while in ORG the crop was kept free of weeds through hand hoeing when necessary. Harvest was done at full ripeness (stage 92) [21] during the last 10 days of June under both CONV and ORG cropping systems. The grain yield was determined on 11.2 m 2 per plot (four central meters of 14 middle rows) to avoid border effects and was expressed at 13% moisture level.

Meteorological Trend
The total annual rainfall amounted to 397 mm, with October and May recorded as the driest months (16 and 12 mm, respectively), and December as the wettest (82 mm). The winter was mild, since the minimum temperature never dropped below 5 • C and temperature excursion was 10 • C, on average. In March the total rainfall decreased, and mean temperatures increased improving the grain filling.

Reagents and Solvents
Analytical grade reagents were purchased from Sigma Aldrich (Milan, Italy). K, Na, Ca, Mg, P, Fe, Cu, Mn, and Zn standards were obtained from Perkin Elmer (Norwalk, CT, USA). Milli-Q system (Millipore Corp., Bedford, MA, USA) ultrapure water was used throughout this research.

Grain Quality Evaluation
The wet gluten and deformation energy of dough (W) were determined, according to Approved Methods 38-12.02 and 54-30.02 of American Association of Cereal Chemists (AACC International, 2000) and expressed as J and %, respectively. Nitrogen content was determined following Kjeldahl method using potassium sulfate and selenium as catalyst and sulphuric acid 0.1 N for titration. Factor 5.7 was applied to obtain the content of protein (expressed as % of dry matter, DM).

Total Polyphenols Content and Antioxidant Activity
Total polyphenols content (TPC) was quantified using a modified Folin-Ciocalteu method [22]. About 1 g of dry material was diluted in 4 mL methanol 80% and stirred at room temperature for 1 h, with shaking. The mixture was centrifuged at 5000× g for 5 min at 25 • C; then, a suitably diluted aliquot was purified with a C-18 end-capped cartridge Phenomenex-Strata (Castel Maggiore, Bologna, Italy) in order to avoid interference by other reducing substances in the assay, and then mixed with Folin-Ciocalteu reagent at room temperature for 2 min. Sodium carbonate (5%, w/v) was added and the mixture was allowed to rest at 40 • C for 20 min in a thermostatic bath. The absorbance was read at 725 nm by a spectrophotometer Shimadzu 1601 UV-Visible (Shimadzu Corp., Tokyo, Japan). The content was determined on the basis of a standard calibration curve generated with increasing concentrations of ferulic acid and expressed as g of ferulic acid equivalent kg −1 of DM.
The antioxidant activity (AA) of the extracts was evaluated as percentage inhibition of DPPH radical. An aliquot (0.1 mL) of each extract used for TPC assay, was added to 3.9 mL of freshly prepared methanolic solution containing 0.24 g L −1 DPPH, and held in the dark for 30 min at room temperature. Then, the absorbance was measured at 515 nm in the spectrophotometer. The percentage inhibition of DPPH was calculated according to Brand-Williams et al. [23].

Ash and Mineral Analysis
The ash content was determined by Approved Methods 08-12.01 of American Association of Cereal Chemists (AACC International, 2000) and expressed as % of DM. Mineral content was carried out using a Perkin Elmer AAnalyst 200 flame atomic absorption spectrometer (Norwalk, CT, USA) equipped with a multi-element hollow cathode lamp and a deuterium background correction system. The quantification of each mineral in the sample was performed by calibration curves generated by using the specific standard. All data presented are mean values of two independent experiments and expressed as mg kg −1 of fresh weight (FW).

Statistical Analysis
Data on wet gluten, W index, ash, AA, minerals, and TPC were subjected to a two-way ('genotype × cropping system') analysis of variance (ANOVA). When the F-test from ANOVA was significant, an LSD test was applied to compare the means. The CoHort software package ver. 6.4 (CoHort software, Monterey, CA, USA) was used.

Yield and Rheological Traits
The reduction of available nitrogen in the soil is notoriously the most critical point of ORG wheat, because of its effects on both the yield and protein content. Here, the cropping system explained more than 80% of total variation for yield, protein content, and wet gluten ( Table 2). As expected, the CONV yields were higher than ORG ones (3.3 vs. 2.1 t ha −1 ; Table 3), as previously observed by Quaranta et al. [24].  Only 'Mongibello' showed the higher yield under the ORG cropping system (3.7 t ha −1 at moisture 13%), since it is probably better adapted to low available soil nitrogen conditions. This genotype also achieved in the ORG cropping system a higher protein content than the others (Table 3), confirming its better ability in nitrogen uptake under stress nutritional conditions. Quaranta et al. [24] made a similar conclusion on three genotypes of durum wheat. 'Claudio' and 'Ramirez' were characterized by lowest protein content (11.1% DM). Overall, the protein content here reported was lower than the values of previous works [3,24]. This discrepancy could be attributable to the different genotypes adopted and pedo-climatic conditions of experimental locations. According to Vaccari et al. [25], an increase of temperatures induced a lower protein concentration in the grains, as observed in our samples. Changes in protein content at various stages of grain development of durum wheat genotypes have been already reported by Galterio et al. [26]. The total protein and wet gluten content play critical roles in determining flour's processing suitability. The wet gluten represents physic-chemical properties of gluten, such as also the gluten index. Here, the wet gluten did not show statistically significant differences among the genotypes, as well the interaction 'cropping system × genotype'. This result could indicate that for the genotypes under study there was no correlation between the total protein content and the wet gluten, in accordance with Kovacs et al. [27]. In addition, the content of pentosanes and hemicelluloses appeared to have a strong effect on the wet gluten [28]. In contrast, Simic et al. [29] considered the genotype as the most important factor influencing the qualitative characteristics of gluten. Nevertheless, these authors did not investigate the effect of cropping system, as done in our study. Along with the wet gluten, the W is an important rheological property for defining the end-use quality of the flour. ANOVA results displayed that both cropping system and genotype had a similar influence on this parameter ( Table 2). The CONV cropping system reported the higher value than the ORG ones (110.3 vs. 87.4 × 10 −4 J) for all the studied genotypes excepted for 'Mongibello' (Table 3). This registered the highest level of W in both cropping systems compared to the other genotypes. It is interesting to note that the genotypes with the lowest values of W also reported the highest level for ash ('Ramirez' and 'Tirex') (Tables 3 and 4). In a previous work, the ash content in wheat was negatively correlated to W [30]. Table 4. Ash (% dry matter) and macro-minerals content (mg kg −1 fresh weight) as affected by 'cropping system (CS) × genotype (G)' interaction. Different letters (a-d or A, B) within each main factor indicate significance differences (LSD test, p ≤ 0.05). All data are expressed as mean ± standard deviation.

Total Polyphenols Content and Antioxidant Activity
The TPC and AA of the investigated whole grain genotypes are presented in Table 5. Here, ANOVA revealed that the cropping system explained more than 80% of the total variation (Table 3). In particular, the ORG cropping system recorded a higher TCP than CONV ones (2.6 vs. 1.9 g kg −1 of DM). Additionally, Zuchowsky et al. [31] concluded that ORG wheat contained more polyphenols than CONV ones. Similar results were observed in other crops, such as tomato and potato [32,33]. The results also showed some significant differences in AA, although mean values did not vary much among the genotypes studied ( Table 5). The highest value was found for 'Duilio' (85% inhibition of DPPH), while the lowest for 'Claudio' (69% inhibition of DPPH). As reported for the TPC, also for the AA the ORG cropping system had the highest value (77% inhibition of DPPH). We also observed a strong correlation between AA and TPC (R 2 = 0.97, p ≤ 0.01), which was consistent with results reported by previous works carried out on wheat grain [8,34]. However, our results contrast with those reported by Yu et al. [35], who found no correlation between total polyphenols content and radical scavenging capacity. Regarding TPC, statistically significant differences were observed among genotypes. 'Duilio', followed by 'Ramirez' and 'Saragolla', had the highest amounts (2.9 and 2.3 g kg −1 of DM, respectively), while 'Claudio' and 'Simeto' showed the lowest content (1.7 and 1.8 g kg −1 of DM, respectively) ( Table 5). Our results were in contrast with those reported by previous studies, which found that 'Anco Marzio' showed the highest content [3,22]. On the contrary, Martini et al. [36] observed that such genotype had the lowest content of TPC (as sum of free, conjugated, and bound phenolic acids). These discrepancies could be mainly attributable to the different environmental conditions, that have an important impact on the biosynthesis and accumulation of phenolic compounds. This hypothesis was supported by the consideration that previous findings reported that both cultivation area and season had a relevant role on the fluctuation of TPC [37,38]. Moreover, to our knowledge, previous investigations have not investigated the variation of TPC in relation to the cropping system. This might provide an additional explanation for such differences.

Ash, Macro-, and Micro-Minerals Content
Here, we also report the ash, macro-and micro-minerals content of whole grain in relation to both genotype and cropping system. It is well known that the 'genotype × cropping system' interaction has a significant effect upon the crop yield and quality performances [39][40][41]. In the present work, ANOVA outcomes showed statistical significance for the interaction 'cropping system × genotype', except for K and Na ( Table 6). The cropping system appeared to be the predominant factor influencing the minerals content, because it reported a higher source of variation than the genotype, excluding Ca and K (Table 6). Moreover, the cropping system explained more than 50% of the total variation for Mn (77.6%), K (73.9%), Na (72.9%), Fe (58.7%), and Cu (52.4%) ( Table 6). Thus, the high concentration of mineral elements in the present study might be attributed to the cropping system adopted and reflects the influence of soil supply of nutrients and the different genetic efficiency of nutrients uptake. Nevertheless, because of a poor number of studies, comparison of the mineral composition in CONV and ORG grains with literature data is still difficult. In addition, to our knowledge, all previous investigations concerning the genotypic variation of mineral concentration in relation to the cropping system, except for the work conducted by Ryan et al. [18], have been not carried out on the whole grain. Sometimes literature reports also appeared to be contradictory in relation to the adopted cropping system. For instance, Mäder et al. [16] assessed the lack of significant differences for the content of ash and most minerals for wheat produced in organic and inorganic cropping systems, whereas others found significant differences [18,19].
In particular, Nitika et al. [19] reported that Cu and Mn contents were significantly higher in CONV grown wheat varieties as compared with ORG grown ones. In this work similar results were found, since all the considered micro-elements were significantly highest in the CONV cropping system, even if the response was genotype-dependent (Table 7). For example, 'Claudio' and 'Saragolla' had the higher values of Fe, Mn, and Zn in the ORG cropping system. On the contrary, the macro-elements, except for P, showed an opposite behavior (Tables 4 and 8). This trend was in agreement with the findings of Ryan et al. [18], which reported that CONV grain had higher P and lower Zn and Cu contents than ORG ones. This discrepancy could be explained taking into account that the experiments were undertaken in different environments. The environmental factor could gain an important role on the mineral elements' concentration, as observed in a collection of Italian durum wheat genotypes [42], and in other crops [43][44][45].
Highly significant variation existed among genotypes for the concentration of ash and each mineral, except for Na, indicating the potential for genetic improvement of such qualitative traits (Tables 4 and 8).
In general, high mean levels of most of the minerals were found compared to previous studies on whole grain [12,18]. Overall our data showed that the Na/K ratio was very low (0.02) (Table 4), in agreement with Rodriguez et al. [49]. The Na/K ratio is an important health parameter, since high values are associated with an increased risk of elevated blood pressure and cardiovascular disease [50].