The content and quality of protein in winter wheat grains depending on sulphur fertilization

Abstract

On soils lacking in water-soluble sulphur, the sulphur fertilization of winter wheat, in general, increases the yields. There are not sufficient investigations about the influence of sulphur on the quality of yield. The objective of this work was to investigate the content and quality of protein in wheat grain depending on sulphur fertilization. The present study relies on field trials conducted on two different soils during 2004–2009. Sulphur was applied with NS-fertilizer Axan or Axan Super at the rate of S 10 or 13.6 kg ha⁻¹ accompanied by a nitrogen background of N 100 kg ha⁻¹. The rates of N- and NS-fertilizers were divided and applied at the beginning and at the end of tillering. At harvest, the grain samples from trial variants in four replications were taken, and the contents of crude protein, wet gluten, amino acids (lysine, threonine, cysteine, methionine) and gluten index in wheat grain were determined. Besides, the contents of amino acids were recalculated on their concentrations in protein. The protein and wet gluten contents in grain varied significantly depending on weather conditions of the trial years. On break-stony soil, sulphur increased the yield by 1.16 t ha⁻¹ on average, i.e. by 21.7%. With increasing yields the protein and wet gluten concentrations in grain decreased. Under the influence of sulphur, the gluten index increased significantly – from 58 to 74, i.e. by 27.6%. In 2004 and 2005, sulphur increased the cysteine and methionine content in wheat grain. Although sulphur application in many cases decreased the protein and wet gluten contents in wheat grain, it improved the biological quality of protein because the concentrations of above-mentioned amino acids recalculated on their concentrations in protein increased significantly. The sulphur application in pseudopodzolic soil had a weaker effect on the grain quality than in break-stony soil.

Keywords:

• Crude protein
• cysteine
• gluten index
• grain yield
• lysine
• methionine
• threonine
• Triticum aestivum
• wet gluten

Introduction

Sulphur was rarely deficient for agricultural crops until about two decades ago, and has become one of the most limiting nutrients for agricultural production in Europe today (McGrath 2003, Loudet 2008, Reinbold et al. 2008). Since the 1980s, sulphur fertilization and the effect of sulphur deficiency on growing plants, crops, and the properties of dough prepared from wheat flour with sulphur deficiency have attracted considerable interest (Reinbold et al. 2008).

Many investigations have shown a great significance of sulphur for wheat production because the application of sulphur fertilizers has influenced both the yield and quality of wheat grain (Randall et al. 1990, Zhao et al. 1999, McGrath 2003, Järvan et al. 2006, 2009, Weber et al. 2008, Mars 2009).

Several studies (Byers and Bolton 1979, Podlesna and Cacak-Pietrzak 2008, Habtegebrial and Singh 2009, Mars 2009) indicated that the application of sulphur-containing fertilizers increased both the yield and the protein content of grain. However, in some studies (Järvan et al. 2006, Weber et al. 2008) higher grain yields were correlated with lower protein content. Depending on soil and weather conditions, as well as the influence of applied sulphur fertilizers the correlation between the grain yield and protein content may be positive or negative (Järvan et al. 2009). The effect of applied sulphur on the wheat yield and grain quality is closely related to nitrogen supply (Fitzgerald et al. 1999, Flaete et al. 2005, Györi 2005, Thomason et al. 2007, Järvan 2008).

One of the major components of the storage proteins in wheat grain are prolamins, accounting for about 50% of the total grain N (Shewry et al. 1997). Prolamins are further divided into gliadins and glutenins. Gliadins are minomers that are soluble in aqueous alcohol solutions, whereas glutenins are polymers, consisting of high molecular weight (HMW) and low molecular weight (LMW) subunits, and these become extractable with aqueous alcohol solutions only after the inter-chain (between subunits) disulphide bonds are broken with reducing reagents. Prolamin polypeptides differ greatly in the content of cysteine residues, and can therefore be classified into S-poor, S-rich, and HMW glutenin subunits (Shewry et al. 1997). The limited S availability favours the synthesis of low-S gliadin storage proteins and high molecular weight subunits of glutenin at the expense of S-rich proteins (Wrigley et al. 1980).

The composition of proteins is influenced substantially by the sulphur availability (Zhao et al. 1999, Naeem and MacRitchie 2003) because sulphur is a constituent of several essential compounds such as cysteine, methionine, coenzymes, thioredoxine and sulfolipids (Singh 2003). Sulphur application accelerates the metabolic pathway of protein synthesis in the plant (Aulakh 2003), and serves to increase the content of sulphur-bearing amino acids, methionine and cysteine (Byers and Bolton 1979, Wrigley et al. 1980, Aulakh 2003, Granvogl et al. 2008, Reinbold et al. 2008, Habtegebrial and Singh 2009).

Several studies (Castle and Randall 1987, MacRitchie and Gupta 1993, Wieser et al. 2004) have indicated that sulphur deficiency affects the amount and the proportions of different gluten protein types in wheat flour. Sulphur deficiency caused a significant increase in the amount of S-free ω-gliadins and moderately increased the amount of S-poor high molecular weight (HMW) glutenin subunits. Additionally, it has been shown that sulphur deficiency causes low yield and poor technological properties of wheat, the latter of which results in dough that is less extensible and more resistant to extension and in loaves of smaller volume and poorer texture (MacRitchie and Gupta 1993). The supply of sulphur to the plant is important for the quantitative composition of gluten proteins and, therefore, for the technological properties of wheat flour (Randall and Wrigley 1986, Zhao et al. 1999, Koehler et al. 2004).

The results of field trials conducted in England showed that even though additions of sulphur fertilizer did not affect the total content of crude protein in wheat grain, they tended to increase the amount of gel protein. The gel protein fraction contains predominantly glutenins which are closely linked to bread-making quality (Zhao et al., 1999).

In wheat protein, the essential amino acid in greatest deficit is lysine (McDonald et al. 2002, Shewry 2009). The decrease in the relative lysine content of high protein grain results from proportional increases in the lysine-poor gluten proteins when excess nitrogen is available, for example, when fertilizer is applied to increase grain yield and protein content (Shewry 2009).

In an earlier investigation (Järvan et al. 2011) we observed a significant effect of nitrogen and sulphur fertilization on the yield and yield components of winter wheat in two locations with different soil and climatic conditions in Estonia. Therefore, grain samples from this earlier study were selected to further explore the effects of nitrogen and sulphur fertilization on several grain quality parameters such as protein and wet gluten content, gluten index and some essential amino acids content in grain and protein of wheat.

For the present study we have set the following hypothesis: the sulphur fertilization of winter wheat, with increasing yields may decrease the protein concentration in grain but improve the quality of protein.

Materials and methods

The present investigation is based on the field trials conducted in 2004, 2005, 2007 and 2008 on break-stony soil at Saku in Northern Estonia (59°18′N, 24°39′E) and in 2005 and 2009 on pseudopodzolic soil at Auksi in Southern Estonia (58°27′ N, 25°36′E). The agrochemical properties of these soils have been characterized in detail in our earlier study (Järvan et al. 2011). Incidentally, the content of water-soluble S (ISO 11048) at the beginning of the growing season determined by using ICP (wave length 181.975 nm) was the following: at Saku S 8–10 mg kg⁻¹, and at Auksi S 6–12 mg kg⁻¹. In the previous autumn under wheat sowing with the complex fertilizer the plant nutrients were applied at the following rates: in 2007 and 2008 at Saku – N₁₂ P₂₆ K₅₀ S₁₅ kg ha⁻¹; in 2009 at Auksi – N₁₂ P₂₆ K₇₅ S₉ kg ha⁻¹. In the field trials of 2004 and 2005, wheat was not given any mineral fertilizers in the previous autumn, because the phosphorus and potassium contents in soil were sufficient.

The trials were performed with winter wheat (Triticum aestivum L.) variety ‘Lars’, except in 2009 when the variety ‘Ada’ was sown at Auksi. Red clover as preceding crop and green manure to wheat was grown in trials at Saku and, in 2005, at Auksi. In 2009 at Auksi, wheat was preceded by oilseed rape.

The effect of sulphur as a plant nutrient on wheat grain quality was investigated on the nitrogen background of N 100 kg ha⁻¹ that was applied broadcast as solid topdressing divided into two portions: N 60 kg ha⁻¹ at the beginning of tillering and N 40 kg ha⁻¹ at the end of tillering. As detailed in Järvan et al. (2011), the dates of fertilizer application within years somewhat differed because due to weather conditions the wheat plants passed the growth stages at different times. Two fertilizer variants were compared – N (control) and NS. In the N-treatment ammonium nitrate at the rate N100 was used. In the NS-treatment the same nitrogen rate was applied with Axan or Axan Super. These granulated fertilizers contained N 27% (N-NO₃ 13.5% and N-NH₄ 13.5%) and water-soluble sulphate-S 2.7 or 3.7%. In the NS-treatment sulphur (S) was given at the rate 10–13.6 kg ha⁻¹. All trials included also a non-fertilized 0-treatment – a so-called field background. The field trials were performed on 25 m² trial plots in four replications. Weather conditions during the growing period in both locations are shown in Table I.

Table I. Mean air temperature and precipitation for growing seasons 2004–2009 at Saku and Auksi, and long-term average (1922–2009) at Jõgeva.

	At Saku				At Auksi
Month	2004	2005	2007	2008	2005	2009	Long term average
Air temperature, °C
May	9.5	9.7	10.8	8.8	10.7	11.5	10.2
June	12.4	13.0	15.1	13.2	14.7	13.9	14.4
July	15.7	17.2	16.2	15.3	18.5	17.1	16.7
01–15 August	16.4	15.9	17.3	14.3	16.3	15.7	15.3
01 May–15 August	13.5	14.0	14.8	12.9	15.0	14.5	14.2
Precipitation, mm
May	35	47	27	10	67	17	50
June	75	39	8	96	43	85	66
July	259	82	42	53	48	136	81
01–15 August	25	92	18	111	96	68	40
01 May–15 August	394	260	95	270	254	306	237

At the maturity stage, wheat grain yields from trial plots were harvested with a combine harvester, dried, sorted and calculated to 86% dry matter. The grain samples of 1 kg weight were taken from the yields of trial variants in four replications. These samples were presented in the laboratory, where the content of crude protein and wet gluten, and gluten index of wheat grain were determined. Later, for each trial treatment the grain samples of four replications were gathered, mixed carefully and the average grain samples were taken for amino acids analysis. In these samples the contents of lysine, threonine, cysteine and methionine were determined in three replications.

The quality analyses were performed in the plant production laboratory of the Agricultural Research Centre. Protein was determined according to AOAC Official Method 2001.11. The procedure is applicable to the determination of Kjeldahl N using digestion block and steam distillation and autotitration unit. A ground laboratory sample is digested in H₂SO₄ at boiling point, using catalyst K₂SO₄+CuSO₄ in tablets form. Using a steam distillation unit acid is neutralized with NaOH and the liberated NH₃ is distilled into boric acid solution and titrated with standardized sulphuric acid to a colorimetric endpoint.

Wet gluten content and quality were determined according to ICC Standard No. 155:1994. Gluten separated from whole wheat meal by Glutomatic equipment is centrifuged to force wet gluten through a specially constructed sieve under standardized conditions. The total weight is defined as gluten quantity. The percentage of wet gluten remaining on the sieve after centrifugation is defined as the gluten index.

The determination of amino acids was performed according to the following methods: EVS-EN ISO 13903:2005 and HPLC UV (Perkin Elmer LC System). Cysteine and methionine are extracted with an oxidation mixture: performic acid (HCO₂OH)+ phenole (C₆H₅OH), then lysine and threonine are extracted with a hydrolysis mixture: hydrochloric acid (HCl)+phenole (C₆H₅OH). The amino acids are separated by reversed-phase high-performance liquid chromatography (HPLC) with precolumn derivatization (FOMOC-Cl, ADAM) and are detected at 263 nm. Column: RP C18 (150×4 mm, 5 mm i.d.) at 24°C; eluent A: acetate buffer 50 mM pH=4.2; eluent B: acetonitrile; gradient elution, eluted within 45 min.

All results were based on four or three replicates. The means were calculated for each variant and the Tukey–Kramer honestly significant difference (HSD) test was used to determine the differences between the means (JMP 5.0.1 software; SAS Institute, Cary, NC).

Results and discussion

The results of field trials conducted in 2004–2008 on the break-stony soil at Saku had shown a great difference in the quality of wheat grains. The contents of protein and wet gluten in grain varied significantly depending on trial years (Table II). Comparing these parameters, for example, in the case of nitrogen fertilizer treatment at the rate of N 100 kg ha⁻¹ it was evident that in 2004 and 2005 the grain protein content was about 14%, whereas in 2007 and 2008 the protein content resulted only in about 10.5%. Also the wet gluten contents of wheat grain in 2004 and 2005 were significantly higher than in 2007 and 2008.

Table II. The effect of N and NS fertilization on the yield and some biological quality parameters of winter wheat grains in the field trials on the break-stony soil at Saku.

	Year
Parameter/fertilizer treatment*	2004	2005	2007	2008	Average 2004–2008
Grain yield, t ha ^−1 (n = 4)
Without fertilizer	3.15^b	4.58^c	3.47^b	3.44^c	3.66^c
N	3.44^b	5.08^b	5.66^a	7.20^b	5.34^b
NS	4.92^a	5.88^a	5.92^a	9.26^a	6.50^a
Protein content, % (n = 4)
Without fertilizer	10.6^b	11.1^c	7.9^c	7.8^b	9.4^c
N	14.1^a	13.9^a	10.6^a	10.4^a	12.2^a
NS	11.6^b	13.4^b	10.1^b	10.2^a	11.3^b
Wet gluten content, % (n = 4)
Without fertilizer	23.0^b	21.9^b	11.2^b	11.7^b	17.0^b
N	31.9^a	31.6^a	22.8^a	19.4^a	26.4^a
NS	25.1^b	30.2^a	21.0^a	18.6^a	23.7^a
Gluten index, % (n = 4)
Without fertilizer	77^a	79^a	98^a	90^a	86^a
N	45^b	32^c	75^b	79^b	58^c
NS	74^a	51^b	90^a	80^b	74^a
Cystein, g kg ^−1 (n = 3)
Without fertilizer	2.60^a	2.35^b	1.47^b	1.25^b	1.92^a
N	2.34^b	2.23^b	1.74^a	1.45^a	1.94^a
NS	2.53^a	2.91^a	1.76^a	1.48^ab	2.17^a
Threonin, g kg ^−1 (n = 3)
Without fertilizer	3.17^a	2.82^ab	3.09^b	2.01^b	2.77^a
N	3.73^a	2.74^b	4.03^a	2.43^a	3.23^a
NS	3.75^a	3.40^a	3.90^a	2.41^a	3.36^a
Methionin, g kg ^−1 (n = 3)
Without fertilizer	2.04^b	1.77^a	1.12^b	0.95^b	1.47^ab
N	1.86^c	1.26^b	1.44^a	1.11^a	1.39^b
NS	2.14^a	1.98^a	1.40^ab	1.15^a	1.67^a
Lysin, g kg ^−1 (n = 3)
Without fertilizer	4.88^b	3.28^a	2.56^a	2.68^b	3.35^a
N	5.24^a	3.13^a	2.77^a	3.17^ab	3.58^a
NS	4.66^c	3.65^a	2.53^a	3.44^a	3.57^a
*Fertilizer treatments: N: ammonium nitrate at the rate N 60 + 40 kg ha^−1; NS: Axan at the rate N60 S6 + N40 S4 kg ha^−1 in 2004 and 2005 or Axan Super at the rate N60 S 8.2 + N40 S5.4 kg ha^−1 in 2007 and 2008. n: number of replicates.
Different letters in the same column indicate significant difference at p < 0.05.

Although Thomason et al. (2007) have concluded that protein level is typically greater in wheat grown under stress of water and temperature this conclusion in the full extent is not applicable for the results of field trials conducted in 2004–2008 at Saku. Comparing the weather conditions of the years, it became evident that the amounts of precipitation during the main growth period, from May to the middle of August when the wheat was harvested, were as follows: in 2004: 394 mm, 2005: 260 mm, 2007: 95 mm, and in 2008: 270 mm. Therefore, the growth period in 2005 and 2008 was with normal precipitation, 2004 with high rainfall but 2007 was extraordinary dry. The springs of 2004 and 2005 were with sufficient and regular precipitation which favoured the uptake of fertilizer nutrients applied as topdressing. The high availability of nitrogen enhances growth and biomass formation in wheat plants as well as the transfer of nitrogen in the form of amino acids to the growing grains (Wrigley et al. 1980, Zörb et al. 2010). Developing wheat grains require nitrogen and sulphur to synthesize storage proteins. During generative growth, either the external medium or reserves in vegetative tissues must provide nitrogen and sulphur for protein synthesis in the grains (Fitzgerald et al. 1999). According to Timms et al. (1981) and Grove et al. (2009) fertilization regimes of high nitrogen without giving any extra sulphur may induce a sulphur deficiency in the grains that will highly influence their quality.

In our trials conducted in 2004 and 2005 at Saku, the fertilization of winter wheat with nitrogen at the rate of N 100 kg ha⁻¹ had an insignificant effect on the grain yield, but it significantly affected the protein content. In these trials under the influence of nitrogen, the protein content in wheat grain increased as follows: in 2004, from 10.6 to 14.1, i.e. by 33.0%, and in 2005, from 11.1 to 13.9, i.e. by 25.2%. Also the wet gluten content rose from 23.0 to 31.9 and from 21.9 to 31.6, i.e. by 38.7 and 44.3%, respectively. However, at the same time the one-sided fertilization with nitrogen significantly decreased the quality of grain protein. The first sign of protein quality degradation was a great decline in gluten index value, which decreased from 77 to 45, i.e. by 41.6%, and from 79 to 32, i.e. by 59.5% as compared with the unfertilized variant. There were also some significant changes in the contents of some amino acids under the influence of nitrogen application.

In 2007, there was a great lack of precipitation during growth from May to the middle of July. Therefore, the uptake and moving of topdressed nutrients to the wheat plants were inhibited; for this reason, the plants at the maturing stage had nitrogen and sulphur available in too small quantities to synthesize grain protein in sufficient amounts. In that year, the grain protein content in N-treatment reached only 10.6% and in NS-treatment 10.1%. The wet gluten content in wheat grain by these treatments was 22.8 and 21.0%, respectively.

In the field trial of 2008, the protein and wet gluten contents of wheat grain in treatments were more or less on the same level as in 2007. Their low content might be caused, at first, from weather conditions in the growing season. In 2008, the spring until to the middle of June was relatively dry, and the applied fertilizers, probably, could not take a significant effect. Besides, rainy and cool days prevailed at the maturity stage of wheat which affected the protein synthesis negatively. Another reason for low protein content might be the protein dilution effect, because much higher yields were reached in fertilized treatments this year than in previous years. By Thomason et al. (2007), wheat yield and protein concentration are inversely related due to dilution effects within the plant. Also in the study of Weber et al. (2008) higher grain yields were correlated with lower protein concentration. The authors consider the dilution effect to be a possible reason for the lower protein concentration.

The sulphur application at the rate of S 10 to 13.6 kg ha⁻¹ on the nitrogen background of N 100 kg ha⁻¹ increased the wheat grain yield in all trials conducted in 2004–2008 on break-stony soil at Saku. As an average of four years, the grain yield increased by 1.16 t ha⁻¹, i.e. 21.7%. At the same time, with the increasing yields the protein concentration in grains decreased significantly in 2004, 2005 and 2007. As an average of these three trials, the protein concentration decreased by 9.1%. Also, the sulphur application tended to decrease the wet gluten content; a significant decrease (21.3%) was revealed in the trial conditions of 2004.

In trials conducted in 2005 and 2009 on the pseudopodzolic soil at Auksi (Table III), the application of nitrogen fertilizer at the rate of N 100 kg ha⁻¹ increased the winter wheat yield by 81.7 and 57.6% as compared with the unfertilized treatment. The nitrogen application had a significant effect on the quality of wheat grains because in both years the contents of protein and wet gluten were increased. The application of sulphur in addition to nitrogen on pseudopodzolic soil did not have any positive effect either on grain yield or on protein and wet gluten content. The gluten index remained practically unaffected by treatments, except in 2005 when in N-treatment the gluten index value decreased by 29.1% as compared with the unfertilized variant or by 25.6% when compared with the NS-treatment.

Table III. The effect of N and NS fertilization on the yield and some quality parameters of winter wheat grains in the field trials on the pseudopodzolic soil at Auksi (n = 4).

Year/fertilizer treatment*	Yield, t ha^−1	Protein, %	Wet gluten, %	Gluten index, %
2005
Without fertilizer	3.38^b	11.5^b	20.8^b	86^a
N	6.14^a	13.2^a	27.6^a	61^b
NS	6.63^a	13.7^a	29.3^a	82^a
2009
Without fertilizer	5.43^b	11.8^b	22.9^b	86^a
N	8.56^a	12.6^a	27.0^a	77^a
NS	8.65^a	12.4^ab	26.3^ab	80^a
Average 2005 and 2009
Without fertilizer	4.40^b	11.5^b	21.8^b	86^a
N	7.35^a	12.9^a	27.3^a	69^c
NS	7.64^a	13.1^a	27.8^a	81^b
*Fertilizer treatments: N: ammonium nitrate at the rate N 60+ 40 kg ha^−1; NS: Axan at the rate N60 S6+N40 S4 kg ha^−1 in 2005 or Axan Super at the rate N60 S 8.2+N40 S5.4 kg ha^−1 in 2009.
n: number of replicates.
Different letters in the same column indicate significant difference at p<0.05.

It is generally accepted that the values of the gluten index range 60–90% in trade for bread-making (Tayyar 2010). It has been found that the optimum gluten index values for central European wheat cultivars range between 75 and 90% (Curic et al. 2001).

In our research the results on gluten index of wheat grain have given evidence for a great variation, depending on the year and the treatment, from 32% to 98% (Table II). The lowest gluten indexes were found in 2004 and 2005 in treatment whereby nitrogen fertilizer at the rate N 100 kg ha⁻¹ without sulphur was applied. Although the contents of protein and wet gluten in wheat grain increased considerably under the influence of nitrogen, the gluten index values reached only 45% and 32%, which cannot be regarded as sufficient for good bread-making quality of wheat. As our further research (Järvan et al. 2006) with wheat flours from different treatments had shown, several parameters of baking quality (stability and quality number of dough, loaf's volume, height to diameter, and porosity) decreased significantly when the gluten index of wheat grain was below 50%. In comparison with the unfertilized treatment the fertilization with nitrogen at the rate N 100 kg ha⁻¹ decreased considerably the gluten index value of wheat grain grown on the break-stony soil at Saku in 2004–2008 (Table II) and on the pseudopodzolic soil at Auksi in 2005 (Table III). Also the investigations of Garrido-Lestache et al. (2004) revealed that increased nitrogen rates had a negative effect on the gluten index of wheat. In another study, Ames et al. (2003) found out that increased protein content as a result of nitrogen fertilization has no significant influence on gluten strength as measured by gluten index.

Compared with N-fertilized treatment, sulphur supply at the rate S 10 or 13.6 kg ha⁻¹ increased significantly the gluten index of wheat, which rose by 27.6% as an average of four year trials conducted at Saku. In the field trials conducted at Auksi, the application of sulphur increased the gluten index of wheat grain by 34.4% in weather conditions of 2005, but had no effect in 2009. Also investigations conducted in Spain by Garrido-Lestache et al. (2004) have shown that sulphur application had no significant effect on the protein quality indices of wheat.

The biological quality of wheat as feed and as food is highly dependent on the content and composition of protein. With increasing protein content, the percentage of some essential amino acids (lysine, threonine and cystine among them) in wheat grain will generally decrease (Maner 1987). Of the 20 amino acids commonly present in proteins, 10 can be considered to be essential in that they cannot be synthesized by animals and must be provided in the diet (Shewry 2009). In wheat protein, the essential amino acid in greatest deficit is lysine (McDonald et al. 2002, Shewry 2009). Lysine is very important in feeding animals and poultry. For pigs, lysine has been shown to be the first limiting amino acid, followed by threonine, valine and methionine (Maner 1987).

The amino acid composition of wheat grain is susceptible to fertilizer treatments (Byers and Bolton 1979, Mortensen et al. 1992, Zhao et al. 1999, Granvogl et al. 2008, Zörb et al. 2010).

Byers and Bolton (1979) showed that sulphur deficiency decreased the concentration of cysteine and methionine (expressed as a percentage of total recovered amino acids) in grain or flour markedly, with the effect being more pronounced on cysteine than on methionine. Because the methionine is not the most limiting essential amino acid in wheat, the effect of sulphur availability on the nutritional value of wheat proteins would be less important if the concentrations of other more limiting amino acids, such as lysine and threonine, were not affected (Byers and Bolton 1979). However, they also found that both lysine and threonine were decreased as a result of sulphur deficiency, particularly when the nitrogen supply was high. Wrigley et al. (1980) found that sulphur deficiency had a minor effect on lysine concentration, but on the other hand it caused a noticeable decrease in the concentration of threonine.

In our studies conducted on break-stony soil in 2004–2008, it became evident that the application of nitrogen and sulphur had a different effect on some amino acids content in winter wheat grain (Table II). In 2004, the fertilization with nitrogen at the rate of N 100 kg ha⁻¹ increased the lysine content by 7.4%, but decreased the cysteine and methionine content by 10.0 and 8.8%, respectively. Both of these sulphur-containing amino acids are very important for the development of good bread-making quality of wheat flour (Timms et al. 1981). According to Tea et al. (2007), the current practice of applying large amounts of nitrogen fertilizers to cereal crops without considering sulphur requirements is becoming a concern for crop quality. In 2004, under the influence of sulphur added at the rate of S 10 kg ha⁻¹ to nitrogen, the cysteine and methionine contents increased by 8.1 and 15.1%, respectively. In 2005, the differences in amino acid contents depending on fertilizer treatments were more notable than in 2004. A major decline (28.8%) in methionine content in the case of one-sided nitrogen supply was observed. Under the influence of sulphur applied to nitrogen, the content of several amino acids increased significantly as follows: cysteine – 30.5%, methionine – 57.1%, and threonine – 24.1%. In this respect, our results are analogous to the findings of several other researchers (Aulakh 2003, Koehler et al. 2004, Granvogl et al. 2008) who have asserted that the applied sulphur increased the sulphur-bearing amino acids methionine and cysteine. Also, the investigation of Habtegebrial and Singh (2009) had shown a significant positive effect of sulphur fertilization on both cysteine and methionine content in wheat grain, and, in addition, applied sulphur also improved the nitrogen use efficiency.

In 2007 and 2008 when the weather conditions were not favourable for protein synthesis, the fertilizer application practically did not affect the content of amino acids in wheat grain. Due to very different weather conditions in 2004–2008, the values of the amino acid contents varied highly, therefore their four-years’ average values do not shown significant differences. An exception was in the case of methionine in which content in wheat grain increased under applied sulphur by 20.1% as an average of four-year trials.

In trials conducted in 2005 and 2009 on the pseudopodzolic soil, the nitrogen fertilization at the rate of N 100 kg ha⁻¹ in both years significantly increased the content of all amino acids determined (Table V). In 2005, sulphur applied with nitrogen at the rate of S 10 kg ha⁻¹ increased the content of sulphur-containing amino acids in wheat grain as follows: cysteine 12.7%, and methionine 8.5%. In the weather conditions of 2009, sulphur did not increase the amino acid content in wheat grain.

Although in the field trials conducted on break-stony soil at Saku the sulphur application – as compared with the one-sided nitrogen fertilization – in many cases significantly decreased the protein and wet gluten content in wheat grain, it improved the biological quality of protein. Recalculating the concentrations of amino acids in wheat grain on its content in wheat protein it became evident that in the case of N-treatment the concentrations of all amino acids in protein were lower than without fertilizer (Table IV). This result coincides with the results of other researchers (Timms et al. 1981, Shewry 2009, Zörb et al. 2010) stating a loss of nutritional and bread-making quality of wheat protein when excess nitrogen without sulphur is available.

Table IV. The effect of N and NS fertilization on the amino acid content in the protein of winter wheat grown on the break-stony soil at Saku.

	Year
Amino acids in wheat protein/fertilizer treatment*	2004	2005	2007	2008	Average 2004–2008
Cystein, g kg ^−1
Without fertilizer	24.5	21.2	18.6	16.2	20.1
N	16.6	16.0	16.4	13.9	15.7
NS	21.8	21.7	17.4	14.6	18.9
Threonin, g kg ^−1
Without fertilizer	29.9	25.4	39.1	25.8	30.0
N	26.7	19.7	38.0	23.4	27.0
NS	32.3	25.4	38.6	23.6	30.0
Methionin, g kg ^−1
Without fertilizer	19.2	15.9	14.2	12.2	15.4
N	13.2	9.1	13.6	10.7	11.4
NS	18.4	14.8	13.9	11.3	14.6
Lysin, g kg ^−1
Without fertilizer	46.0	29.5	32.4	34.5	35.6
N	37.2	22.5	26.1	30.5	29.1
NS	40.2	27.2	25.0	33.7	31.5
*Fertilizer treatments: N: ammonium nitrate at the rate N 60 + 40 kg ha^-1; NS: Axan at the rate N60 S6 + N40 S4 kg ha^-1 in 2004 and 2005 or Axan Super at the rate N60 S 8.2 + N40 S5.4 kg ha^−1 in 2007 and 2008.

Table V. The effect of N and NS fertilization on the amino acid content in the grains and protein of winter wheat grown on the pseudopodzolic soil at Auksi.

	Content in grains, g kg^−1 (n = 3)			Content in protein, g kg^−1
Amino acid/ fertilizer treatment*	2005	2009	Average 2005–2009	2005	2009	Average 2005–2009
Cystein
Without fertilizer	2.42^c	2.18^c	2.30^b	21.0	18.5	19.8
N	2.75^b	2.47^a	2.61^ab	20.8	19.6	20.2
NS	3.10^a	2.36^b	2.73^a	22.6	19.0	20.8
Threonin
Without fertilizer	2.95^b	3.17^b	3.06^b	25.6	26.9	26.2
N	3.63^a	3.49^a	3.56^a	27.5	27.7	27.6
NS	3.53^a	3.46^a	3.50^a	25.8	27.9	26.8
Methionin
Without fertilizer	1.82^c	1.55^b	1.68^b	15.8	13.1	14.4
N	2.11^b	1.77^a	1.94^ab	16.0	14.0	15.0
NS	2.29^a	1.78^a	2.04^a	16.7	14.4	15.6
Lysin
Without fertilizer	4.55^b	3.07^b	3.81^a	39.6	26.0	32.8
N	5.11^a	3.27^a	4.19^a	38.7	26.0	32.4
NS	5.30^a	3.08^b	4.19^a	38.7	24.8	31.8
*Fertilizer treatments: N: ammonium nitrate at the rate N 60 + 40 kg ha^−1; NS: Axan at the rate N60 S6 + N40 S4 kg ha^-1 in 2005 or Axan Super at the rate N60 S 8.2 + N40 S5.4 kg ha^-1 in 2009.
n: number of replicates.
Different letters in the same column indicate significant difference at p < 0.05.

When comparing the amino acid concentrations recalculated in wheat protein of N- and NS-treatments, it appeared that the sulphur application might increase the amino acids content as an average of four years as follows: methionine 28.1%, cysteine 20.4%, threonine 11.1% and lysine 8.2%. From the viewpoint of animal feeding, how much protein and essential amino acids can be obtained from a unit area is very important. Regardless of the decreased protein content in wheat grain, but thanks to the increased yields, the application of sulphur allowed an extra income of 83.0 kg protein to be obtained from a hectare area. The total amounts of amino acids in the yield from a hectare area under the influence of sulphur were much higher than without sulphur. As an average of four years, the amino acids' yields calculated on unit area (kg ha⁻¹) increased under influence of sulphur as follows: methionine 46.4%, cysteine 36.1%, threonine 21.8% and lysine 21.3%.

In conclusion, the hypothesis which was set up at the start of the present study proved correct. On the break-stony soil in 2004–2008, the sulphur application at the rate of S 10 to 13.6 kg ha⁻¹ on the nitrogen background of N 100 kg ha⁻¹ increased the winter wheat yield by 1.16 t ha⁻¹ on average, i.e. by 21.7%. With increasing yields the protein and wet gluten concentrations in grain decreased significantly but the biological quality of protein improved under the influence of sulphur because the gluten index value and the contents of amino acids in protein increased. On the pseudopodzolic soil, the application of sulphur in addition to nitrogen did not have any positive effect either on grain yield or on protein and wet gluten content. However, in weather conditions of 2005, fertilization with sulphur increased the content of sulphur-containing amino acids cysteine and methionine in wheat grain.

Acknowledgements

Financial support from the Estonian Ministry of Agriculture through the project ‘Improving the food and feed quality of cereals, grain legumes and oil crops by implementing economically effective and environmentally sustainable agrotechnical methods’ (2006–2010) is much appreciated. We thank Mr Jaanus Rebane and Mrs Ann Akk for the description of analysis methods. We thank also the staff of the plant production laboratory of the Agricultural Research Centre for their accurate analytical work. The authors are grateful to Mrs Helena Pärenson for her linguistic consultations.