Grain filling duration and glutenin polymerization under variable nitrogen supply and environmental conditions for durum wheat

Field Crops Research 171 (2015) 23–31

Contents lists available at ScienceDirect

Field Crops Research journal homepage: www.elsevier.com/locate/fcr

Grain filling duration and glutenin polymerization under variable nitrogen supply and environmental conditions for durum wheat Roberto Ferrise a , Marco Bindi a , Pierre Martre b,c,∗ a b c

Department of Agri-food Production and Environmental Sciences, University of Florence, Piazzale delle Cascine 18, 50144 Florence, Italy INRA, UMR1095 Genetics, Diversity and Ecophysiology of Cereals, 5 Chemin de Beaulieu, F-63 039 Clermont-Ferrand, Cedex 02, France Blaise Pascal University, UMR1095 Genetics, Diversity and Ecophysiology of Cereals, F-63 177 Aubière, France

a r t i c l e

i n f o

Article history: Received 4 July 2014 Received in revised form 28 October 2014 Accepted 28 October 2014 Available online 29 November 2014 Keywords: Durum wheat Glutenin Polymeric protein Protein composition Size-exclusion HPLC Wheat quality

a b s t r a c t The end-use value of durum wheat (Triticum turgidum L. subsp. durum (Desf.) Husn.) is mainly governed by its grain protein concentration and composition. Adjustment of variables to compensate for inter-annual and location variations in semolina quality leads to high cost for processors in the wheat industry. A better understanding of the mechanisms governing environmental variations of grain protein composition is thus required. Here, a field experiment was setup in a Mediterranean environment with the aim to analyze the effect of sowing date and nitrogen (N) fertilization on the dynamics of grain dry mass, water, protein composition and glutenin polymer size distribution for durum wheat cv. Creso. The results indicated that (1) grain dry mass accumulation was related to grain water concentration and stopped at 44.9% independently of the growing conditions and N supply; (2) during the grain filling period as well as at ripeness maturity, the quantity of the different protein fractions scaled with the quantity of N per grain; (3) SDS-extractable glutenin polymers were produced continuously until the same grain water concentration as the dry mass deposition was reached; (4) SDS-unextractable polymeric proteins were found as early as 7 days after anthesis and their rate of accumulation increased sharply when grain dry mass was 60% of its final value and proceeded at a constant rate until ripeness maturity, thus suggesting that the insolubilization of glutenin polymers is not directly related to the rapid loss of water after physiological maturity, but rather to the continuous dehydration of the grain. © 2014 Elsevier B.V. All rights reserved.

1. Introduction In contrast with other cereals, most of the wheat production is used after processing, mainly by the pasta industry in the case of durum wheat (Triticum turgidum L. subsp. durum (Desf.) Husn.), which require specific functional properties. These properties largely depend on structures and interactions of the grain storage proteins gliadin and glutenin (Gras et al., 2001; Shewry and Halford, 2002). Customers are becoming more discriminating in their quality requirements and inter-annual variability in product quality is becoming less acceptable, particularly for premium

Abbreviations: DAA, days after anthesis; DM, dry mass; EPP, total SDSextractable polymeric proteins; GWM, mass of water per grain; SD, sowing date; SDS, sodium dodecyl sulfate; SPP, small-size polymeric proteins; LPP, large-size polymeric proteins; N, nitrogen; TPP, total polymeric proteins; UPP, SDSunextractable polymeric proteins; %GWC, percent grain water concentration. ∗ Corresponding author at: INRA, UMR1095 Genetics, Diversity and Ecophysiology of Cereals, 5 Chemin de Beaulieu, Cedex 02, F-63 039 Clermont-Ferrand, France. Tel.: +33 473 624 351; fax: +33 473 624 457. E-mail addresses: roberto.ferrise@unifi.it (R. Ferrise), marco.bindi@unifi.it (M. Bindi), [email protected] (P. Martre). http://dx.doi.org/10.1016/j.fcr.2014.10.016 0378-4290/© 2014 Elsevier B.V. All rights reserved.

products (Marchylo et al., 2001). Therefore, adjustment of variables to compensate for inter-annual and location variations in semolina quality lead to high cost for processors in the wheat industry. A better understanding of the mechanisms governing environmental variations of grain protein composition is thus required. The gliadin fraction accounts for 20% to 30% of total grain protein content and consists mainly of single chain polypeptides with molecular mass (Mr ) ranging from 15 to 60 × 103 , but most are in the narrow range 25 to 40 × 103 (Bunce et al., 1985; Wieser, 2007). The glutenin fraction accounts for 30% to 45% of total grain protein content and consists of very high Mr polymers (5 × 105 to more than 1 × 107 ) stabilized by inter-chain disulfide bonds and are partially insoluble in denaturing sodium dodecyl sulfate (SDS) solutions (Wieser et al., 2006). Glutenin polymers are made of low (LMW-GS; Mr 32 to 35 × 103 ; D‘Ovidio and Masci, 2004) and high (HMW-GS; Mr 67 to 88 × 103; Gao et al., 2010) molecular mass subunits and account for 5% to 10% and 20% to 30% of total grain protein content, respectively. The rheological properties of wheat gluten and dough depend on the balance between monomeric gliadins and polymeric glutenins and most importantly on the Mr distribution of the latter (Weegels et al., 1996; MacRitchie, 1999; Don et al., 2003). The ratio of SDS-unextractable polymeric proteins (UPP) to total polymeric

24

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

proteins (TPP) has often been used as a measure of the distribution of the Mr of glutenin polymers (e.g. Gupta et al., 1993; Larroque and Bekes, 2000) and has consistently been linked to dough and gluten rheological properties (Southan and MacRitchie, 1999; Morel et al., 2000; Ohm et al., 2009). Extended periods of high temperature are common in most durum wheat-growing areas of the world and above optimal temperature is one of the major environmental factors affecting durum wheat yield and quality. Weaker dough from grains that experience one or several days of very high temperature (>30 ◦ C) has been related to a marked decrease in the proportion of UPP, independently of changes in total grain protein concentration or proportions of gliadin classes or glutenin subunits (Ciaffi et al., 1996; Corbellini et al., 1998; Wardlaw et al., 2002; Don et al., 2005). Nitrogen (N) supply has strong effect on both grain protein concentration and composition. Higher rate of N application tend to increase the gliadin-to-glutenin ratio and to decrease the proportion of UPP, resulting in higher dough extensibility (e.g. Johansson et al., 2004, 2005; Godfrey et al., 2010). Significant interactions between growing temperature and N supply in grain protein composition and the size distribution of glutenin polymers have been reported (Johansson et al., 2005; Malik et al., 2011), which complicate the interpretation of their own effects under variable environmental conditions. Some studies have suggested that the formation of UPP in both durum and bread wheat (T. aestivum L.) may also be controlled by the water status of critical grain structures and is triggered by the sharp increase in grain water loss at physiological maturity (Stone and Nicolas, 1996; Carceller and Aussenac, 1999, 2001). In contrast, other studies have suggested that the formation of UPP start during the grain filling period (Zhu and Khan, 1999; Panozzo et al., 2001; Johansson et al., 2005; Jia et al., 2012), maybe when a threshold concentration of monomer subunits or small-size polymers is reached (Gupta et al., 1996). Moreover, how growing conditions and N supply affect the dynamics of glutenin polymer formation in relation to the cessation of grain dry mass accumulation and changes in %GWC is largely unknown. These aspects of glutenin polymer formation (i.e. rate and duration of the accumulation) are particularly important to understand how the environment modifies the grain protein composition and to develop a phenomenological model of grain protein accumulation and polymer formation in response to environmental variations. The aim of this work was thus to study the effect of sowing date and N fertilization, two important management practices, on the dynamics of grain dry mass, water mass and concentration, and protein composition and glutenin polymer size distribution for the durum wheat cv. Creso grown in the field in a Mediterranean environment. In particular, we aimed at testing the following hypothesis: (1) the quantity of the different grain protein fractions scales with the quantity of N per grain and the scaling relationships are independent of the growth conditions; (2) the formation of glutenin polymers is related to changes in grain water concentration; (3) the insolubilization of glutenin polymers is not directly related to the rapid loss of water after physiological maturity, but rather to the continuous dehydration of the grain and thus starts early during the grain-filling period.

2. Materials and methods 2.1. Experiment set up, plant sampling and grain dry mass, water mass and nitrogen determination Rain-fed field experiments were carried out at the University of Florence, Italy (11◦ 13 E, 43◦ 46 N; 42 m elevation) during the 2002–2003 and 2004–2005 growing seasons (hereafter referred

as 2003 and 2005, respectively). Seeds of the durum wheat (T. turgidum L. subsp. durum (Desf.) Husn.) cultivar Creso were sown on 11 December 2002 and 05 November 2004 (normal sowing, treatments termed 03SD1 and 05SD1, respectively), and 27 January 2003 and 18 January 2005 (late sowing, hereafter 03SD2 and 05SD2, respectively). Creso is a semi-dwarf Italian durum wheat cultivar, characterised by very limited cold requirements and a high sensitivity to photoperiod (Motzo and Giunta, 2007). Due to its moderate but constant yield and its great adaptability to Italian environmental conditions, it has been widely cultivated during past decades and is still in use (Arduini et al., 2006). Three N treatments were applied with a total of 0, 6, and 18 g N m−2 (hereafter N0, N6, and N18, respectively). The treatments were arranged in a split-plot design with three blocks, where the main plots corresponded to the sowing dates. For a detailed description of the experiment set up and data collection the reader is referred to Ferrise et al. (2010). Daily weather data were recorded on a weather station adjacent to the field plots. Thermal time was calculated by summing daily degree-days, which were calculated as the daily mean air temperatures above a base temperature of 0 ◦ C (Cao and Moss, 1989). Within each subplot, 20 plants were randomly tagged and their phenological development determined as described in Tottman (1987) by daily inspections in the field. For each treatment, 16 mainstems were randomly collected at 5–7 days intervals starting at growth stage (GS) 71 (grain water ripe) in 2003 and at GS 65 (anthesis) in 2005. Spikes were hand threshed and grains were counted. Grain dry mass (DM) was determined on subsamples (ca. one-third of the whole sample weight) after oven drying at 80 ◦ C for 48 h. Average single grain water mass (GWM) was calculated as the difference between fresh mass and DM divided by the number of grains of the subsample. The remaining grains were freeze-dried and milled to whole flour using a rotor mill (Cyclomill, PBI, UK). Grain total N concentration was determined with the Dumas method (AOAC method no. 7.024) using a FlashEA 1112 NC Analyzer (Thermo Electron Corp., Waltham, MA, USA) and was expressed on a DM basis. Grain protein concentration was calculated by multiplying grain N concentration by 5.62 (Tkachuk, 1966; Mossé et al., 1985). 2.2. Protein extraction Whole flour samples (75 mg) were stirred for 2 h at 60 ◦ C in the presence of 7.5 mL of a 0.1 M sodium phosphate buffer (pH 6.9) containing 2% (v/v) sodium dodecyl sulfate (SDS), and were then centrifuged for 30 min at 37.5 × 103 × g at 20 ◦ C (Dachkevitch and Autran, 1989) to obtain a supernatant (SDS-extractable protein fractions). The Pellets were then stirred for 10 min at room temperature in the presence of 7.5 mL of the same extractant, and the resulting dispersion was sonicated for 30 s at 10 W and 22.5 kHz (Daniel and Triboi, 2002), using a 3-mm diameter probe mounted on a XL—MicrosonTM ultrasonic cell disruptor (Misonic Inc., NY, USA). After centrifugation (30 min, 37.5 × 103 × g, 20 ◦ C), the supernatants (SDS-unextractable protein fractions) were collected. Both extracts were filtered through 0.20 ␮m regenerated cellulose filters (Titan2 HPLC filtration systems, Sun SRi, TN, USA) and stored at −20 ◦ C in sealed HPLC vials until analysis. 2.3. Protein fractionation and quantification by size-exclusion high performance liquid chromatography Proteins in the SDS-extractable and -unextractable fractions were fractionated by size-exclusion high performance liquid chromatography (SE-HPLC) using a Bio-Tek Kontron HPLC (Bio-Tek Instruments, Inc., VT, USA). A Kroma System 2000 V1.83 was used to control the pump and for acquisition and

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

400

A

12 400

300

Absorbance at 214 nm (mV)

2 000 000

29 000 66 000

200

97 000

100

0 40

1

2

3

B

4

5

2 DAA 13 DAA 21 DAA 35 DAA

30 20 10 0 20

30

40

50

60

Retention time (min) Fig. 1. Size-exclusion high-performance liquid chromatography elution profiles of unreduced proteins extracted with 0.1 M sodium-phosphate buffer (pH 6.9) containing 2% (v: v) sodium dodecyl phosphate (SDS) without ((A) SDS-unextractable protein fraction) and with ((B) SDS-unextractable protein fraction) sonication from grains of durum wheat (Triticum turgidum var. durum (Desf.) cv. Creso harvested 7, 13, 21, and 35 days after anthesis (DAA). Crops were sown on 18 January 2005 and received 18 g N m−2 (treatment 05SD2N18). The elution profiles were divided into five peaks as indicated by the vertical lines under the peaks. Vertical arrows with Mr indicate the elution times of the standard proteins used to calibrate the column (see Section 2).

reprocessing of data from the detector. A SuperoseTM 6 10/300 GL TricornTM column (General Electric Healthcare, Bucks, UK; manufacture’s claimed Mr separation range of 5 × 103 to 5 × 106 ) was used. The column was eluted isocratically at ambient temperature with a 0.1 M sodium phosphate buffer (pH 6.9) containing 0.1% (v/v) SDS (Bietz, 1984). Each sample (75 ␮L) was run for 125 min at a flow rate of 0.3 mL min−1 . Protein elution was detected at 214 nm. The column was calibrated by running standard proteins (Dextran blue [Mr = 20 × 106 ], phosphorylase B [Mr = 97.4 × 103 ], bovine serum albumin [Mr = 66 × 103 ], anhydrase carbonic [Mr = 29 × 103 ], and cytochrom C [Mr = 12.4 × 103 ]) under the same conditions as the wheat samples. The calibration curve was Mr = (2.699 × 1012 ) × (F × tr )–6.926 (r2 = 0.994, d.f. = 9, P < 0.0001), where F (mL min−1 ) is the flow rate and tr is the retention time (min). Similar to previous studies (Bénétrix et al., 1994; Cornec et al., 1994), the elution profiles were divided into five peaks (Fig. 1) exhibiting Mr of 0.8 to 2.7 × 106 for peak 1; 1 to 8 × 105 for peak 2, 44 to 100 × 103 for peak 3, 20 to 44 × 103 for peak 4, 6 to 20 × 103 for peak 5. Based on the Mr limits of the peaks and previous studies it can be deduced that Peak 1 consisted of large-size glutenin polymers, peak 2 of small-size glutenin polymers, peak 3 to 4 were mostly gliadins, and peak 5 albumins–globulins (Dachkevitch and Autran, 1989; Bénétrix et al., 1994; van Herpen et al., 2008; Belitz et al., 2009). The composition of each of these five peaks for immature and mature grains was confirmed by SDS-polyacrylamide gel electrophoresis analysis of collected peaks (Supplementary Fig. S1). The results largely confirmed the protein composition of the five peaks, in particular the predominance of ω-gliadins in peak 3 was confirmed. Therefore, peaks 1 and 2 of the SDS-extractable protein

25

fraction are herein referred to as large-(LPP) and small-(SPP) size polymeric proteins, respectively, and their sum as SDS-extractable polymeric proteins (EPP). Peaks 1 and 2 of the SDS-unextractable protein fractions are referred to as SDS-unextractable polymeric proteins (UPP), peaks 3 and 4 of the SDS-extractable plus the SDSunextractable protein fractions as gliadins, and peaks 5 and 6 of the SDS-extractable plus the SDS-unextractable protein fractions as albumins–globulins. The sum of the EPP and UPP fractions represents the total polymeric proteins (TPP). To quantify the proteins in each of the SE-HPLC peak, SDSextractable and -unextractable protein fractions were extracted as described above from 3.5 g of whole meal flour obtained from mature grains. The extracts were dialyzed against water for 24 h at room temperature, freeze-dried, and total N concentration was determined on five subsamples by the Kjeldahl method using a Kjeltec 2300 analyzer (Foss Tecator AB, Hoeganaes, Sweden). Eighty eight and 24 ␮g of freeze-dried SDS-extractable and SDS-unextractable proteins were dissolved in 75 ␮L of the extraction buffer containing 2% (v/v) SDS, respectively. These solutions were then diluted at 1:1, 2:3, 1:2, 2:5, 1:3, 1:4, and 1:5 with the same buffer and subjected twice to SE-HPLC fractionation as described above. The calibration equation were y = (5.08 × 10−3 ) × surface area + 36.13 × 10−3 (r2 = 1.000, d.f. = 13, P < 0.0001) for the SDS-extractable protein fraction and y = (6.74 × 10−3 ) × surface area + 0.1298 (r2 = 0.999, d.f. = 8, P < 0.0001) for the SDS-unextractable protein fraction, where y is the quantity of N (␮g) determined by the Kjeldahl method. A highly significant correlation was found between the sum of the quantity of N per grain of the six SE-HPLC peaks for the SDS-extractable and SDS-unextractable fractions and the total quantity of N per grain determined independently with the Dumas method (r = 0.990, d.f. = 200, P < 0.001). The slope of the reduced major axis regression (Warton et al., 2006) between these two dependent variables was not different from one (0.936 ± 0.009; P < 0.001) and the intercept was not different from zero (P < 0.001), indicating that after sonication ca. 95% of the total proteins had been extracted. 2.4. Data analysis and statistics All statistical analyses were done using SPSS for Windows 18.0.0 (SPSS Inc. Chicago, IL, USA). Differences in the investigated parameters, were analyzed by performing an analysis of variance (˛ = 0.05) using a year-combined split-plot design. The main effect of each factor and their interactions were tested using the appropriate expected mean square values as reported in McIntosh (1983). Significant differences among N fertilization treatments were calculated using the Duncan’s test. All regression analysis were done with the software package R-2.12.2 for windows (R Development Core Team, 2007). To determine the rate and duration of accumulation of grain dry mass, water mass, total N and protein fractions, data were fitted with a 3-parameter logistic function equation (Triboi et al., 2003): Q (t) =



Qmax

1 + 0.05 exp (−4r (t − t95 )) /Qmax



(1)

where Q is the quantity of dry mass or N, t is the number of days or degree-days (base 0 ◦ C) after anthesis, and Qmax is the final value of Q approached as t→ ∞, r is the maximum rate of accumulation defined as the derivative of the point of inflexion, and t95 is the duration of accumulation defined as the duration, from anthesis, in which 95% of Qmax is accumulated. The accumulation of the different grain components were analyzed simultaneously for all year/SD/N treatment combinations with a parallel curve analysis procedure (Ross, 1984) using the nls() function of the R/stats package. Statistical differences in Qmax , r and t95 among years, SD

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

The two seasons differed greatly in the amount and distribution of rainfall and temperature (Supplementaty Table S1 available at JXB online; Ferrise et al., 2010). Weather in 2005 was more similar to the climatology of the area, while 2003 was drier and warmer than usual. During the wheat growing season, rainfalls in 2003 were halved compared to 2005 (205 mm in 2003 vs 417 mm in 2005). In particular during grain filling, rainfalls in 2003 were 25 mm for 03SD1 and only 8 mm for 03SD2, whilst in 2005 rainfalls were 75 and 30 mm for 05SD1 and 05SD2, respectively (Supplementary Table S1). Average daily temperature was ca. 1 ◦ C higher in 2003 than in 2005 and for the late (SD2) than for the normal (SD1) sowing date. During the post-anthesis period, the number of days with maximum daily temperature (Tmax ) above 30 ◦ C ranged from 17 days in 05SD1 to 30 days in 03SD2. During the same period, on average, Tmax exceeded 35 ◦ C in 10 days during 2003 and in 2 days in 2005. 3.2. Dynamics of single grain dry mass, nitrogen and water Year and SD affected the rate but not the duration of grain DM accumulation (Fig. 2A). The differences were almost entirely ascribed to the mean daily maximum temperature calculated for the 15 days prior to anthesis (r = −0.99; P = 0.002; Ferrise et al., 2010). The kinetics of grain DM was not modified by N treatments. Although more evident in 2005 than in 2003, SD and N treatments affected both the rate and the duration of grain N accumulation leading to higher quantity of N per grain for SD2 and with higher N fertilization rates (Fig. 2B; Supplementary Table S2). The mass of water per grain (GWM) increased gradually from anthesis until 21 days later and then remained constant until 30 days after anthesis (DAA; Fig. 2C). GWM decreased sharply after 30 DAA. Although statistically significant, maximum GWM was only 2% and 6% lower for N6 and N18 compared with N0, respectively (Supplementary Table S2). Plotting grain DM normalized by final grain DM versus %GWC, revealed a unique relationship (Fig. 3A). Grain DM increased linearly with respect to %GWC and regardless of the year, SD and N treatment the maximum grain DM was reached at 44.9% (CI95% = 43.3–46.4%). In Fig. 3B the GWM normalized by the maximum GWM was plotted versus the percentage of final grain DM. The results indicated that, regardless of the year, SD or N treatment, grain DM accumulation continued up to the beginning of the grain dehydration phase, suggesting that the end of grain filling is synchronous with the end of the water plateau. 3.3. Dynamics of grain protein fractions In both years, the final quantity per grain of the different protein fractions was strongly modified by the SD and N treatments

-1

Grain dry mass (mg DM grain )

60

A 50 40 30 20 10 0 1.5

B

-1

3.1. Weather conditions

2005 SD1 SD2

N0 N6 N18

Grain total N (mg N grain )

3. Results

2003 SD1 SD2

-1

and N treatments were judged by comparing the eight models produced by the parallel curve analysis by ANOVA analyses and by calculating the second-order corrected Akaike’s information criterion using the ACIc() function of the R/qpcR package. Piecewise linear models were fitted using the R/segmented package (Muggeo, 2003). Differences in grain water concentration at which grain filling ceased and the rate of polymeric protein fractions changed were considered as significantly different when the 95% confidence interval (CI95% ) of their difference did not contain zero. Allometric grain N allocation relationships were fitted by reduced major axis regression using the R/smatr package (Warton et al., 2012).

Grain water (mg H2O grain )

26

1.2 0.9 0.6 0.3 0.0 50

C

40 30 20 10 0 0

10

20

30

40

50

Days after anthesis Fig. 2. Grain dry mass (A), total nitrogen (B), and water content (C) versus days after anthesis for durum wheat cv. Creso grown in the field at Florence, Italy, during the 2002–2003 (03) and 2004–2005 (05) growing seasons. The crops were sown in October/November (normal sowing, SD1) or in January (late sowing, SD2) and received either 0, 6, or 18 g N m−2 (N0, N6, and N18, respectively). Data are means ±1 s.e. for n = 3 independent replicates.

(Fig. 4 and Supplementary Table S3). Year × SD and year × N interactions were significant for all the protein fractions except for albumin–globulin, for which only the year × SD interaction was significant. Higher final quantities per grain of the three protein fractions were found in 2005 than in 2003, for SD2 compared to SD1 and at higher N fertilization rates. Due to the interactions between the main factors, the effects of SD and N were larger in 2005 than in 2003. The kinetics of the different protein fractions reflected the dynamics of total grain N, but the rate of accumulation of albumin–globulin increased before that of gliadin and TPP and the rate of accumulation of TPP increased after that of gliadin (Fig. 4). When averaged over the two years and SD and N treatments, the maximum rate of accumulation was similar for the albumins– globulins (0.020 ± 0.001 mg N grain−1 d−1 ) and the gliadin (0.021 ± 0.001 mg N grain−1 d−1 ) but was slightly higher for TPP (0.023 ± 0.001 mg N grain−1 d−1 ). On average, albumin–globulin

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

2003 SD1 SD2

27

2003 SD1 SD2

2005 SD1 SD2

N0 N6 N18

2005 SD1 SD2

N0 N6 N18

A

100

A

0.5

Albumins-globulins -1 (mg N grain )

% final grain dry mass

120

80 60 40 20

0.4 0.3 0.2 0.1

0 80

60

40

20

0.0 0.5

0

B

Gliadins -1 (mg N grain )

B

100 80

0.4 0.3 0.2

60

0.1

40

0.0 0.5

20 0 0

20

40

60

80

100

% final grain dry mass Fig. 3. Percent of final grain dry mass versus grain water concentration (A) and percent of maximum grain water versus percent of final grain dry mass (B) for durum wheat cv. Creso grown in the field during two seasons, with two sowing dates and three rates of N application. Symbols are as in Fig. 2. In (A) the solid line is piecewise regression; the vertical and horizontal bars above the data indicate the break point and its 95% confidence interval, respectively. Data are means ±1 s.e. for n = 3 independent replicates.

accumulated up to ca. 32.7 ± 0.8 DAA, while gliadin stopped at 28.4 ± 0.8 DAA and TPP continued up to 37.0 ± 0.8 DAA. Although SD and N treatments modified the kinetics of the three protein fractions in a complex way, there were unique relationships between the quantity of albumin–globulin, gliadin and TPP and the total quantity of N per grain (Fig. 5). When an allometric power relationship was fitted to the mature grains only, the scaling exponent of the regression was significantly different from 1 only for albumin–globulin (inset Fig. 5A; P = 0.026). Accordingly, the final protein composition reflected the scaling coefficient of the allometric relationships and on average albumin–globulin, gliadin, and TPP accounted for 28.2%, 25.8%, and 35.8% of the total grain protein content (Supplementary Table S4). N treatments did not modify the proportions of these three protein fractions and the year × N and year × SD × N interactions were significant only for gliadin. The proportions of albumin–globulin were modified by the year and the SD and the proportion of TPP was modified only by the SD and the year × SD interaction. The gliadin to glutenin ratio was modified neither by N fertilization nor by SD but changed in response to the year, with higher ratio in 2005 (Supplementary Table S4). However, although statistically significant, the differences among treatments were very small.

Total polymeric proteins -1 (mg N grain )

% maximum grain water

Grain water concentraton (% of FM)

C

0.4 0.3 0.2 0.1 0.0 0

10

20

30

40

50

Days after anthesis Fig. 4. Quantity of albumin–globulin (A), gliadin (B), and total polymeric proteins (C) versus days after anthesis for grains of durum wheat cv. Creso grown in the field during two seasons, with two sowing dates and three rates of N application. Symbols are as in Fig. 2. Data are means ±1 s.e. for n = 3 independent replicates.

3.4. Dynamics of glutenin polymer assembly To investigate the assembly of glutenin polymers, TPP were split into small-size (SPP) and large-size (LPP) SDS-extractable and SDS-unextractable (UPP) polymeric proteins. The patterns of accumulation of the different size-classes of polymers were affected in a complex way by year, SD and N treatments (Fig. 6). Year × SD and year × N interactions for the final quantities of SPP and LPP per grain were significant. For these two polymer fractions, no differences were found in 2003 in response to SD and N treatments, while in 2005 their final quantities were higher for SD2 than for SD1 and at higher N fertilization rates. The final quantity of UPP was not modified by SD, but was 24% higher in 2003 than in 2005 and for both years was 20% higher for N18 compared with N0. The UPP to TPP ratio was not modified by SD and N treatments, but the year × SD interaction was significant. The UPP to TPP ratio was higher in 2003 (0.24 ± 0.003) than in the 2005 (0.17 ± 0.005). The final quantities of SPP and LPP were closely correlated to the final quantity of N grain (r = 0.91 for SPP and r = 0.85 for LPP, P < 0.001,

28

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

2003 SD2 SD1

2005 SD1 SD2

2003 SD1 SD2

2005 SD1 SD2

N0 N6 N18

N0 N6 N18

0.3

A

A -1

SPP (mg N grain )

Albumins-globulins -1 (mg N grain )

0.5 0.4 0.3 0.2

1.25 Y = 0.27 X

0.2

0.1

0.1 0.0

B -1

B

LPP (mg N grain )

Gliadins -1 (mg N grain )

0.0 0.5 0.4 0.3 0.2

1.06 Y = 0.25 X

-1

0.0 0.5

C

0.4 0.3 0.2

0.05

C

0.09

0.06

0.03

1.01 Y = 0.35 X

0.00

0.1 0.0 0.0

0.10

0.00

UPP (mg N grain )

Total polymeric proteins -1 (mg N grain )

0.1

0.15

0

10

20

30

40

50

Days after anthesis

0.3

0.6

0.9

1.2

1.5 -1

Grain total N (mg N grain ) Fig. 5. Quantity of albumin–globulin (A), gliadin (B), and total polymeric proteins (C) versus total grain nitrogen for developing and mature grains of durum wheat cv. Creso grown in the field during two seasons and with two sowing dates and three rates of N applications. Symbols are as in Fig. 2. Inset show mature grain only and solid lines are power regressions fitted on Log transformed data (all r2 < 0.918; d.f. = 10; and P < 0.001), the scaling exponent was significantly different from one (P = 0.026) only for the albumins–globulins—total nitrogen relationship. Data are means ±1 s.e. for n = 3 independent replicates.

Fig. 6. Quantity of small-size (A) and large-size (B) SDS-extractable polymeric proteins and SDS-unextractable polymeric proteins (C) versus days after anthesis for grains of durum wheat cv. Creso grown in the field during two seasons and with two sowing dates and three rates of N applications. Symbols are as in Fig. 2. Data are means ±1 s.e. for n = 3 independent replicates.

59.5% (CI95% = 52.6–66.4%), then continued linearly with a marked increase of the rate of accumulation until grain harvest ripeness (%GWC < 15%). 4. Discussion

d.f. = 11), while no correlation were found for the final quantity of UPP (r = 0.26, P = 0.42, d.f. = 11). The accumulation of the polymeric protein fractions (normalized by their maximum value) was analyzed with respect to %GWC. Since no significant differences were found among the treatments, piecewise regression analyses were performed pooling the data from all the treatments. The normalized quantity of SPP increased linearly up to 47.2% (CI95% = 45.4–49.0%) %GWC and then decreased slightly (Fig. 7A). For LPP, two distinct break-points were clearly detectable (Fig. 7B). The first break-point, at 55.0% (CI95% = 50.3–59.8%), marked a sharp increase of the accumulation of LPP, the second one, at 43.0% (CI95% = 39.4–46.7%), marked the end of LPP accumulation. Therefore, the end of the accumulation of both SPP and LPP was concomitant with the end of the grain filling. The accumulation of the normalized quantity of UPP started as early as the first sampling date (7 DAA), progressed slowly up to

In good agreement with previous studies on maize (Saini and Westgate, 2000; Borras and Wesgate, 2006), sorghum (Gambin and Borras, 2005), soybean (Swank et al., 1987), sunflower (Rondanini et al., 2007), castor bean (Vallejos et al., 2011) and wheat (Schnyder and Baum, 1992; Calderini et al., 2000; Ferreira et al., 2012a) we found close synchronisms between the dynamics of grain DM and water during grain development. Regardless of the year, sowing date and N fertilization rate, the accumulation of DM in the grain proceeded linearly as the %GWC decreased up to a critical value, suggesting the presence of regulatory mechanisms internal to the grain independent on the environmental conditions. Slafer et al. (2009) proposed an explanation of such patterns according to which reserves replace water into the grain until a critical minimum content is reached. In our study, the end of grain filling occurred at 44.8% of grain moisture. This value is in line with those

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

2003 SD2 SD1

2005 SD1 SD2

N0 N6 N18 160 140

A

% final SPP

120 100 80 60 40 20 0

% final LPP

120

B

100 80 60 40 20 0

% final UPP

120

C

100 80 60 40 20 0 80 70 60 50 40 30 20 10 0 Grain water concentration (% of FM)

Fig. 7. Percent of the final quantity of small-size (A) and large-size (B) SDSextractable polymeric proteins and SDS-unextractable polymeric proteins (C) versus grain water concentration for grains of durum wheat cv. Creso grown in the field during two seasons and with two sowing dates and three rates of N applications. Symbols are as in Fig. 2. Solids lines are piecewise regressions; the horizontal bars above the regressions indicate the 95% confidence intervals of the break points. The vertical dotted lines indicate the 95% confidence interval of the end of grain filling as estimated in Fig. 3A. Data are means ±1 SE for n = 3 independent replicates.

reported by Schnyder and Baum (1992) and Gooding et al. (2003) for bread wheat and is fairly close to the value of 45.9% found by Ferreira et al. (2012a) in a study on the effect of heat stress during grain filling on durum wheat cv. Dakter. In that study, the authors suggested that the end of grain filling is marked by the physical packing limit reached by the starch granules included in the endosperm. Once this limit is approached, the starch granules cause the irreversible damage to the cell organelles, thus limiting any further dry matter deposition. Accordingly, the final grain DM may be restricted by the maximum volume of the grain, as also reported in Hasan et al. (2011). In our study, although the environmental conditions during grain development as well as N availability changed the rate and the duration of accumulation of the different protein fractions, unique

29

relationships were found between each fraction and the total quantity of N accumulated in the grain. These results are in agreement with findings of Triboi et al. (2003) and Saint Pierre et al. (2008) who reported the process of N partitioning in bread wheat to be substantially not affected by environmental conditions during grain filling, thus leading to final protein composition mainly depending on the final quantity of grain N. Our results reveal that also in durum wheat variations in protein composition at maturity, due to different environmental and nutritional conditions, are ascribed to differences in the total quantity of nitrogen accumulated during the grain filling and simply reflect the coefficient of N partitioning among the different protein fractions. However, in contrast with previous studies on bread wheat (Triboi et al., 2003; Plessis et al., 2013), here we found that the scaling exponent for gliadin and glutenin were not different from unity and therefore the gliadin-to-glutenin ratio was independent of the growing conditions and N availability. According to our results, on average the albumin–globulin fraction accounted for 28.2% of total grain protein, which is substantially higher than values usually reported (10–22%; DuPont and Altenbach, 2003). Moreover, the duration of accumulation of albumin–globulin was longer than that of gliadin. This is in contrast to previous reports (e.g. Carceller and Aussenac, 1999; Triboi et al., 2003; Naeem et al., 2012). Likely, our results may be ascribed to the limitations of the SE-HPLC to separate albumin–globulin and gliadin proteins as the size of these proteins largely over-lap. Triboi et al. (2003) found synchronized changes in the N partitioning coefficients for the different protein fractions in correspondence to the transition between the cell division and the effective grain-filling phases of grain development. More specifically, they found a decrease in the N partitioning coefficients for the albumin–globulin and a corresponding increase in the partitioning coefficients for both gliadin and glutenin. Here, we found that the coefficient of N partitioning for albumin–globulin remained constant during grain development, while the coefficient for gliadin decreased and that of glutenin increased at circa 0.6 mg N grain−1 that was reached between 15 and 25 DAA. These changes in the protein partitioning coefficients most likely indicate the involvement of gliadin proteins in glutenin polymers assembly (Ferreira et al., 2012b). Due to this uncertainty, we cannot take the scaling coefficient of the allometric relationships found in this study (insets of Fig. 5) as those reflecting the real N partitioning among the albumin–globulin and gliadin fractions. Nevertheless, this does not alter the finding that the quantity of different protein fractions (actually the quantity of the SE-HPLC peaks) in durum wheat grains scales with the quantity of N accumulated in the grain. The accumulation of SPP and LPP started as early as 7 DAA, the first sampling dates analyzed here, and stopped in correspondence to the end of dry mass deposition. Studies on the expression of glutenin genes in bread wheat, indicate that the synthesis of LMW-GS and HMW-GS starts at early stages of grain development and proceeds until the end of the grain filling. Dupont et al. (2006) found the transcripts began accumulating at circa 6 DAA and continued until the end of the protein accumulation. Similarly, Shewry et al. (2009) found transcript abundances increasing steadily from 6 to 14 DAA and then stabilize until the end of the grain-filling period. The formation of glutenin polymers is, then, an active process involving cell structures and organelles such as the endoplasmic reticulum and Golgi apparatus (reviewed by Tosi, 2012). Based on these premises, it is likely to suppose that the damage of cell organelles, possibly when the packing limit of the starch granules in the endosperm was approached, hampered any further deposition of SPP and LPP protein fractions, and therefore the end of DM, SSP and LPP accumulation are synchronized. In contrast, the accumulation of UPP continued until harvest ripeness, when

30

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31

%GWC was close to 10%, well after the end of grain filling and of any cellular activity. This suggests that the assembly of UPP is a spontaneous process that can take place even when the “cell machinery” is not functioning. In this study, UPP accumulation increased markedly at 59.5% %GWC, when grain dry mass was only 60% of its final value. The relative rate of UPP accumulation was constant and independent of the growing conditions and N supply until harvest maturity. This is in contrast with findings of Carceller and Aussenac (1999, 2001) on bread wheat, who reported that the rapid increase of the UPP fraction is concomitant with the rapid water loss during grain desiccation. They attributed this dynamic to the increased protein interactions consequent to the disruption of protein bodies that would increase the molecular density and facilitate the formation of inter-chain disulphide bonds. Likewise, Daniel and Triboi (2002) found similar dynamics and reported of earlier insolubilization under drought conditions. They related the insolubilization of proteins to the polimerization of smaller size proteins into larger ones and suggested that this process is related to the changes in grain water and redox status. Based on these findings, Shewry et al. (2009) implied a possible influence of grain dehydration on protein polymerization. On the contrary, in durum wheat, Ferreira et al. (2012a) demonstrated the glutenin polymer assembly to be independent on the loss of integrity of protein bodies. Furthermore, they report accumulation UPP from the early phase of grain filling and did not record any change in the rate of accumulation during the desiccation phase. They ascribed the formation of UPP to the progressive polymerization of EPP and related this dynamic to the continuous decrease of the cellular redox status observed during the grain filling. Our results reinforce this hypothesis and suggest that the insolubilization of glutenin polymers is not directly related to the rapid loss of water after physiological maturity, but rather to the continuous dehydration of the grain and the change in the redox conditions.

5. Conclusion With the aim to explore the kinetics of grain dry mass and protein accumulation and glutenin polymer assembly as affected by crop management practices, crops of durum wheat cv. Creso were sown at two different sowing dates and fertilized with increasing doses of N. We show that (1) the accumulation of dry mass in the grain is associated by %GWC and stopped at 44.9% independently of the growing conditions and N supply; (2) the quantity of albumin–globulin, gliadin and glutenin per grain in immature and mature grains scales with the quantity of N per grain; (3) the formation of SPP and LPP is a continuous process that ceases at the same %GWC as dry mass deposition; and (4) the rate of UPP accumulation increases sharply at 59.5% %GWC, when grain dry mass is only 60% of its final value and the relative rate of UPP accumulation is constant and independent of the growing conditions and N supply until harvest ripeness when %GWC is close to 10%. Therefore, in contrast with previous findings, our results suggest that the insolubilization of glutenin polymers starts as early as 7 DAA and is not directly related to the rapid loss of water after physiological maturity, but rather to the continuous dehydration of the grain. The findings discussed in this paper are relevant to a single durum wheat cultivar. Consequently, any generalization might need caution. In particular, thresholds values found in this study (e.g. the %GWC at which UPP start to accumulate rapidly) may change depending on the cultivar considered. Accordingly, further studies analyzing the genetic variability of these parameters are required.

Acknowledgements The authors thank Jean-Luc Ollier and Vitalie Samoil for their assistance with SE-HPLC analysis and gratefully acknowledge the INRA Environment & Agronomy division for supporting VS through a post-doctoral fellowship. This work was supported by a Galileo French/Italian bilateral project (grant no. 08418ZJ). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.fcr.2014.10.016. References Arduini, I., Masoni, A., Ercoli, L., Mariotti, M., 2006. Grain yield, and dry matter and nitrogen accumulation and remobilization in durum wheat as affected by variety and seeding rate. Eur. J. Agron. 25, 309–318. Belitz, H.D., Grosch, W., Schieberle, P., 2009. Cereals and Cereal Products. Food Chemistry. Springer-Verlag, Berlin, pp. 670–745 (4th revised and extended edition). Bénétrix, F., Kaan, F., Autran, J.-C., 1994. Changes in protein complexes of durum wheat in developing seed. Crop Sci. 34, 462–468. Bietz, J.A., 1984. Analysis of wheat gluten proteins by high-performance liquid chromatography. Baker’s Dig. 58, 2. Borras, L., Wesgate, M.E., 2006. Predicting maize kernel sink capacity early in development. Field Crops Res. 95, 223–233. Bunce, N.A.C., White, R.P., Shewry, P.R., 1985. Variation in estimates of molecular weights of cereal prolamins by SDS-PAGE. J. Cereal Sci. 3, 131–142. Calderini, D.F., Abeledo, L.G., Slafer, G.A., 2000. Physiological maturity in wheat based on kernel water and dry matter. Agron. J. 92, 895–901. Cao, W.X., Moss, D.N., 1989. Temperature effect on leaf emergence and phyllochron in wheat and barley. Crop Sci. 29, 1018–1021. Carceller, J.L., Aussenac, T., 1999. Accumulation and changes in molecular size distribution of polymeric proteins in developing grains of hexaploid wheats: role of the desiccation phase. Aust. J. Plant Physiol. 26, 301–310. Carceller, J.L., Aussenac, T., 2001. SDS-insoluble glutenin polymer formation in developing grains of hexaploid wheat: the role of the ratio of high to low molecular weight glutenin subunits and drying rate during ripening. Aust. J. Plant Physiol. 28, 193–201. Ciaffi, M., Tozzi, L., Borghi, B., Corbellini, M., Lafiandra, D., 1996. Effect of heat shock during grain filling on the gluten protein composition of bread wheat. J. Cereal Sci. 24, 91–100. Corbellini, M., Mazza, L., Ciaffi, M., Lafiandra, D., Borghi, B., 1998. Effect of heat shock during grain filling on protein composition and technological quality of wheats. Euphytica 100, 147–154. Cornec, M., Popineau, Y., Lefebvre, J., 1994. Characterization of gluten subfractions by SE-HPLC and dynamic rheological analysis in shear. J. Cereal Sci. 19, 131–139. D‘Ovidio, R., Masci, S., 2004. The low-molecular-weight glutenin subunits of wheat gluten. J. Cereal Sci. 39, 321–339. Dachkevitch, T., Autran, J.C., 1989. Prediction of baking quality of bread wheats in breeding programs by size-exclusion high-performance liquid chromatography. Cereal Chem. 66, 448–456. Daniel, C., Triboi, E., 2002. Changes in wheat protein aggregation during grain development: effects of temperatures and water stress. Eur. J. Agron. 16, 1–12. Don, C., Lichtendonk, W.J., Plijter, J.J., Hamer, R.J., 2003. Understanding the link between GMP and dough: from glutenin particles in flour towards developed dough. J. Cereal Sci. 38, 157–165. Don, C., Lookhart, G., Naeem, H., MacRitchie, F., Hamer, R.J., 2005. Heat stress and genotype affect the glutenin particles of the glutenin macropolymer-gel fraction. J. Cereal Sci. 42, 69–80. DuPont, F.M., Altenbach, S.B., 2003. Molecular and biochemical impacts of environmental factors on wheat grain development and protein synthesis. J. Cereal Sci. 38, 133–146. Dupont, F.M., Hurkman, W.J., Vensel, W.H., Tanaka, C., Kothari, K.M., Chung, O.K., Altenbach, S.B., 2006. Protein accumulation and composition in wheat grains: effects of mineral nutrients and high temperature. Eur. J. Agron. 25, 96–107. Ferreira, M.S.L., Martre, P., Mangavel, C., Girousse, C., Rosa, N.N., Samson, M.-F., Morel, M.-H., 2012a. Physicochemical control of durum wheat grain filling and glutenin polymer assembly under different temperature regimes. J. Cereal Sci. 56, 58–66. Ferreira, M.S.L., Samson, M.-F., Bonicel, J., Morel, M.-H., 2012b. Relationship between endosperm cells redox homeostasis and glutenin polymers assembly in developing durum wheat grain. Plant Physiol. Biochem. 61, 36–45. Ferrise, R., Triossi, A., Stratonovitch, P., Bindi, M., Martre, P., 2010. Sowing date and nitrogen fertilisation effects on dry matter and nitrogen dynamics for durum wheat: an experimental and simulation study. Field Crops Res. 117, 245–257. Gambin, B.L., Borras, L., 2005. Sorghum kernel weight: growth patterns from different positions within the panicle. Crop Sci. 45, 553–561. Gao, L., Ma, W., Chen, J., Wang, K., Li, J., Wang, S., Bekes, F., Appels, R., Yan, Y., 2010. Characterization and comparative analysis of wheat high molecular weight glutenin subunits by SDS-PAGE, RP-HPLC, HPCE, and MALDI–TOF–MS. J. Agric. Food Chem. 58, 2777–2786.

R. Ferrise et al. / Field Crops Research 171 (2015) 23–31 Godfrey, D., Hawkesford, M.J., Powers, S.J., Millar, S., Shewry, P.R., 2010. Effects of crop nutrition on wheat grain composition and end use quality. J. Agric. Food Chem. 58, 3012–3021. Gooding, M.J., Ellis, R.H., Shewry, P.R., Schofield, J.D., 2003. Effects of restricted water availability and increased temperature on the grain filling, drying and quality of winter wheat. J. Cereal Sci. 37, 295–309. Gras, P.W., Anderssen, R.S., Keentok, M., Békés, F., Appels, R., 2001. Gluten protein functionality in wheat flour processing: a review. Aust. J. Agric. Res. 52, 1311–1323. Gupta, R.B., Khan, K., MacRitchie, F., 1993. Biochemical basis of flour properties in bread wheats. I. Effects of variation in the quantity and size distribution of polymeric protein. J. Cereal Sci. 18, 23–41. Gupta, R.B., Masci, S., Lafiandra, D., Bariana, H.S., MacRitchie, F., 1996. Accumulation of protein subunits and their polymers in developing grains of hexaploid wheats. J. Exp. Bot. 47, 1377–1385. Hasan, A.K., Herrera, J., Lizana, C., Calderini, D.F., 2011. Carpel weight, grain length and stabilized grain water content are physiological drivers of grain weight determination of wheat. Field Crops Res. 123, 241–247. Jia, D., Dai, X., He, M., 2012. Polymerization of glutenin during grain development and quality expression in winter wheat in response to irrigation levels. Crop Sci. 52, 1816–1827. Johansson, E., Kuktaite, R., Andersson, A., Prieto-Linde, M.L., 2005. Protein polymer build-up during wheat grain development: influences of temperature and nitrogen timing. J. Sci. Food Agric. 85, 473–479. Johansson, E., Prieto-Linde, M.L., Svensson, G., 2004. Influence of nitrogen application rate and timing on grain protein composition and gluten strength in Swedish wheat cultivars. J. Plant Nutr. Soil Sci. 167, 345–350. Larroque, O.R., Bekes, F., 2000. Rapid size-exclusion chromatography analysis of molecular zie distribution for wheat endosperm protein. Cereal Chem. 77, 451–453. MacRitchie, F., 1999. Wheat proteins: characterization and role in flour functionality. Cereal Foods World 44, 188–193. Malik, A.H., Prieto-Linde, M.L., Kuktaite, R., Andersson, A., Johansson, E., 2011. Individual and interactive effects of cultivar maturation time, nitrogen regime and temperature level on accumulation of wheat grain proteins. J. Sci. Food Agric. 91, 2192–2200. Marchylo, B.A., Dexter, J.E., Clarke, F.R., Clarke, J.M., Preston, K.R., 2001. Relationships among bread-making quality, gluten strength, physical dough properties, and pasta cooking quality for some Canadian durum wheat genotypes. Can. J. Plant Sci. 81, 611–620. McIntosh, M.S., 1983. Analysis of combined experiments. Agron. J. 153–155, 153–155. Morel, M.-H., Dehlon, P., Autran, J.C., Leygue, J.P., L.H.h., C.B., 2000. Effects of temperature, sonication time, and power settings on size distribution and extractability of total wheat flour proteins as determined by size-exclusion high-performance liquid chromatography. Cereal Chem. 77, 685–691. Mossé, J., Huet, J.C., Baudet, J., 1985. The amino acid composition of wheat grain as a function of nitrogen content. J. Cereal Sci. 3, 115–130. Motzo, R., Giunta, F., 2007. The effect of breeding on the phenology of Italian durum wheats: from landraces to modern cultivars. Eur. J. Agron. 26, 462–470. Muggeo, V.M.R., 2003. Estimating regression models with unknown break-points. Stat. Med., 22. Naeem, H.A., Paulon, D., Irmak, S., MacRitchie, F., 2012. Developmental and environmental effects on the assembly of glutenin polymers and the impact on grain quality of wheat. J. Cereal Sci. 56, 51–57. Ohm, J.B., Hareland, G., Simsek, S., Seabourn, B., 2009. Size-exclusion HPLC of protein using a narrow-bore column for evaluation of breadmaking quality of hard spring wheat flours. Cereal Chem. 86, 463–469. Panozzo, J.F., Eagles, H.A., Wootton, M., 2001. Changes in protein composition during grain development in wheat. Aust. J. Agric. Res. 52, 485–493. Plessis, A., Ravel, C., Bordes, J., Balfourier, F., Martre, P., 2013. Association study of wheat grain protein composition reveals that gliadin and glutenin composition are trans-regulated by different chromosome regions. J. Exp. Bot. 64, 3627–3644.

31

R Development Core Team, 2007. A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rondanini, D.P., Savin, R., Hall, A.J., 2007. Estimation of physiological maturity in sunflower as a function of fruit water concentration. Eur. J. Agron. 26, 295–309. Ross, G.J.S., 1984. Parallel model analysis: fitting non-linear models to several sets of data. In: COMPSTAT 1984—Sixth Symposium on Computational Statistics, pp. 458–463. Saini, H.S., Westgate, M.E., 2000. Reproductive development in grain crops during drought. Adv. Agron. 68, 59–96. Saint Pierre, C., Peterson, C.J., Ross, A.S., Ohm, J.B., Verhoeven, M.C., Larson, M., Hoefer, B., 2008. Winter wheat genotypes under different levels of nitrogen and water stress: changes in grain protein composition. J. Cereal Sci. 47, 407–416. Schnyder, H., Baum, U., 1992. Growth of the grain of wheat (Triticum aestivum L.). The relationship between water content and dry matter accumulation. Eur. J. Agron. 1, 51–57. Shewry, P.R., Halford, N.G., 2002. Cereal seed storage proteins: structures, properties and role in grain utilization. J. Exp. Bot. 53, 947–958. Shewry, P.R., Underwood, C., Wan, Y., Lovegrove, A., Bhandari, D., Toole, G., Mills, E.N.C., Denyer, K., Mitchell, R.A.C., 2009. Storage product synthesis and accumulation in developing grains of wheat. J. Cereal Sci. 50, 106–112. Slafer, G.A., Kantolic, A.G., Appendino, M.L., Miralles, D.J., Savin, R., 2009. Crop development: genetic control, environmental modulation and relevance for genetic improvement of crop yield. In: Sadras, V.O., Miralles, D.J. (Eds.), Crop Physiology. Applications for Genetic Improvement and Agronomy Academic Press, San Diego, CA, pp. 277–308. Southan, M., MacRitchie, F., 1999. Molecular weight distribution of wheat proteins. Cereal Chem. 76, 827–836. Stone, P.J., Nicolas, M.E., 1996. Varietal differences in mature protein composition of wheat resulted from different rates of polymer accumulation during grain filling. Aust. J. Plant Physiol. 23, 727–737. Swank, J.C., Egli, D.B., Pfeiffer, T.W., 1987. Seed growth characterisitcs of soybean genotypes differing in duration of seed fill. Crop Sci. 27, 85–889. Tkachuk, R., 1966. Note on the nitrogen-to-protein conversion factor for wheat flour. Cereal Chem. 43, 223–225. Tosi, P., 2012. Trafficking and deposition of prolamins in wheat. J. Cereal Sci. 56, 81–90. Tottman, D.R., 1987. The decimal code for the growth stages of cereals, with illustrations. Ann. Appl. Biol. 110, 441–454. Triboi, E., Martre, P., Triboi-Blondel, A.-M., 2003. Environmentally-induced changes of protein composition for developing grains of wheat are related to changes in total protein content. J. Exp. Bot. 54, 1731–1742. Vallejos, M., Rondanini, D., Wassner, D.F., 2011. Water relationships of castor bean (Ricinus communis L.) seeds related to final seed dry weight and physiological maturity. Eur. J. Agron. 35, 93–101. van Herpen, T.W.J.M., Cordewener, J.H.G., Klok, H.J., Freeman, J., America, A.H.P., Bosch, D., Smulders, M.J.M., Gilissen, L.J.W.J., Shewry, P.R., Hamer, R.J., 2008. The origin and early development of wheat glutenin particles. J. Cereal Sci. 48, 870–877. Wardlaw, I.F., Blumenthal, C., Larroque, O., Wrigley, C.W., 2002. Contrasting effects of chronic heat stress and heat shock on kernel weight and flour quality in wheat. Funct. Plant Biol. 29, 25–34. Warton, D.I., Duursma, R.A., Falster, D.S., Taskinen, S., 2012. Smatr 3- an R package for estimation and inference about allometric lines. Methods Ecol. Evol. 3, 257–259. Warton, D.I., Wright, I.J., Falster, D.S., Westoby, M., 2006. Bivariate line-fitting methods for allometry. Biol. Rev. 81, 259–291. Weegels, P.L., Hamer, R.J., Schofield, J.D., 1996. Critical review: functional properties of wheat glutenin. J. Cereal Sci. 23, 1–18. Wieser, H., 2007. Chemistry of gluten proteins. Food Microbiol. 24, 115–119. Wieser, H., Bushuk, W., MacRitchie, F., 2006. The polymeric glutenins. In: Wrigley, C., Bekes, F., Bushuk, W. (Eds.), Gliadin and Glutenin: the Unique Balance of Wheat Quality. American Association of Cereal Chemistry, St Paul, MN, pp. 213–240. Zhu, J., Khan, K., 1999. Characterization of monomeric and glutenin polymeric proteins of hard red spring wheats during grain development by multistacking SDS-PAGE and capillary zone electrophoresis. Cereal Chem. 76, 261–269.