Amino acid production from a sunflower wholemeal protein concentrate

Available online at www.sciencedirect.com

Bioresource Technology 99 (2008) 4749–4754

Amino acid production from a sunflower wholemeal protein concentrate C. Ordo´n˜ez, C. Benı´tez *, J.L. Gonza´lez Departamento de Quı´mica Agrı´cola y Edafologı´a, Facultad de Ciencias, Universidad de Co´rdoba, Campus de Rabanales, Edificio Marie Curie, Ctra. Nacional IV-a, km. 336, 14014 Co´rdoba, Spain Received 8 February 2007; received in revised form 21 September 2007; accepted 22 September 2007 Available online 5 November 2007

Abstract A study was undertaken to investigate the influence of protein concentration and the addition of different doses of endopeptidase (Alcalase) and exopeptidase (Flavourzyme) on the sequential enzymatic hydrolysis of a protein concentrate obtained from defatted sunflower wholemeal. The results show that the greatest degree of hydrolysis (37.8%) is achieved by hydrolyzing an aqueous substrate with a 5% protein concentrate, and using a 0.02 g Alcalase/g of protein concentrate of the substrate. The aminograms performed reveal that the free amino acid found in the highest proportion in the hydrolysate was aspartic acid, which accounted for over 50% of the free amino acids present, regardless of the substrate concentration and the enzyme dosage used. Finally, the hydrolysate obtained from a substrate containing a 5% protein concentrate and a 0.02 g Alcalase/g of protein concentrate displayed characteristics that indicate its suitability for use as a vegetable-origin plant growth regulator. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Sunflower wholemeal; Enzymatic hydrolysis; Degree of hydrolysis; Free amino acids

1. Introduction Defatted oil seed wholemeals (soy, sunflower, etc.) represent an important source of proteins, and have traditionally been used in animal feed. Sunflower proteins are characterised by a moderately low level of albumins (17–20%) and a high level of globular proteins (55–66%) (Gheyasuddin et al., 1970a) with a balanced amino acid composition (except for a deficiency in lysine (Tkachuk and Irvine, 1968)). These high protein levels as well as their high proportions of essential amino acids and non-toxic elements (Hoernicke et al., 1998) have meant that for quite some years now the sub-product derived from the sunflower seed—meaning once the oil was extracted (sunflower wholemeal)—was used in animal feed. Regarding its use in food suitable for human consumption, which presents disadvantages, such as high fibre con-

*

Corresponding author. Tel.: +34 957218651; fax: +34 957212146. E-mail address: [email protected] (C. Benı´tez).

0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2007.09.061

tent, the presence of husks and polyphenols, and above all chlorogenic acid, giving it a dark green color (Gassman, 1983; Parrado et al., 1993) and low lysine content, it was established that the latter reactions could all be corrected (Rahmna and Narasinga, 1981; Villanueva et al., 1999). In addition to its use as a wholemeal product, since the 1970s, scientists have been looking into taking better advantage of its high protein content for the production of protein concentrates and isolates (Dench et al., 1981; Fujimaki et al., 1977; Gheyasuddin et al., 1970b; Pacheco et al., 1994) which can be used in both animal feed and food suitable for human consumption. However, the presence of polyphenols, especially chlorogenic acid—favoured by the deficiency of boron in sunflowers (Cakmak and Romheld, 1977)—which is often found in lime soils and which gives rise to this typical dark green color, has led to experiment it for plant nutrition, since previous correction is thus avoided. The background to this study can be found in previous publications made by the authors (Ordo´n˜ez et al., 2001, 2002, 2006), leading to the identification and production

4750

C. Ordo´n˜ez et al. / Bioresource Technology 99 (2008) 4749–4754

of a protein concentrate and other products that can be used as fertilizers. Amino acid fertilizers, applied via the root network or to the leaves, have favourable effects on both soils and plants (Aktas et al., 1993; Aylgworth, 1996; Ga´miz et al., 1997; Gu¨nesß et al., 1994; Martı´nez et al., 2001; Tejada and Gonza´lez, 2003; Yan et al., 1996). The most widely-used type of protein hydrolysis for the production of amino acids used in plant nutrition is enzymatic hydrolysis. Nowadays, bacterial or fungal origin proteolytic enzymes are used if possible (Guadix et al., 2000). The preference for this type of hydrolysis over acid and basic hydrolysis is rooted in the fact that in the two latter cases substances, such as HCl or NaOH, are used, which can significantly affect plant nutrition. Furthermore, these processes require heating, which cause protein denaturalisation. The aim of this study is to obtain amino acid hydrolysates, from a protein concentrate, itself from defatted sunflower wholemeal (Ordo´n˜ez et al., 2001), determining its free amino acid composition and its possible use in agriculture. 2. Methods 2.1. Characterisation of the sunflower wholemeal and the process for obtaining the protein concentrate The initial experimental sample, provided by the company Oleı´cola El Tejar (Cordoba, Spain), was obtained by extracting the sunflower seed oil using the system proposed by Bernardini (1981). Once the sample was ground and homogenised to a particle size of under 1 mm, it was subjected to the following series of analytical determinations: humidity and organic material (MAPA, 1994); determination of nitrogen (Duchafour, 1975); total protein (MAPA, 1994); and determination of total organic carbon (Sims and Haby, 1971). In order to determine the mineral elements, mineralization was carried out following the standard methodology (C.I.I.T.D.F., 1969) and subsequently the concentrations of sodium and potassium were determined by emission spectrophotometry, while calcium and magnesium were determined by absorption spectrophotometry. Phosphorus was assessed according to the Williams and Stewart Method, (Guitian and Carballas, 1976). The results obtained are shown in Table 1. The defatted sunflower wholemeal was used to prepare a protein concentrate using potassium hydroxide and phosphoric acid, following the criteria provided by Ordo´n˜ez et al. (2001). An outline of this process is shown in Fig. 1. 2.2. Obtaining and characterising the hydrolysates In order to prepare the hydrolysates from the protein concentrate, two enzymes were used, both supplied by Novo Nordisk A/S (Bagsvaerd, Denmark) and with the following characteristics:

Table 1 Characteristics (refereed dry matter) of the sunflower flour Moisture Protein Organic carbon Fatty matter Sodium Potassium Calcium Magnesium Phosphorus

7.70% ± 0.3 27.0% ± 1.0 36.3% ± 0.8 1.39% ± 0.12 111 mg/kg ± 16 0.983% ± 0.110 0.210% ± 0.045 0.182% ± 0.018 0.589% ± 0.043

Mean values of the triplicates ± standard error.

Solid Integral defatted sunflower flour

Extraction T=40 ºC, pH=10.5 KOH 0.5N

Liquid Extract T=25 ºC pH=4.5

Protein Precipitation

H3PO4 0.5N

Supernatant

Centrifugation and washing T=25 ºC Drying t=24 hours Protein concentrate

Fig. 1. Global scheme of the process to obtain the protein concentrate (from Ordo´n˜ez et al., 2001).

Alcalase 2.4 L, an endoprotease, produced from Bacilus licheniformis stock. Flavourzyme 1000 L, an enzyme complex produced by fermentation, from selected Aspergillus oryzae stocks. To carry out the different enzyme hydrolyzes, the pHstat technique was used (Parrado et al., 1991). The initial step, prior to adding the hydrolytic enzymes, was to bring the protein concentrate in distilled water to a maximum solubilisation, adjusting both the temperature (55 °C) and the pH (7) of the hydrolysis reaction. The reaction was initiated following the addition of the Alcalase 2.4 L enzyme, keeping the process at a constant pH level by adding KOH 0.1 N. After this endopeptidase had been left to react for 2 h and 30 min, the Flavourzyme 1000 L enzyme, an exopeptidase that acts as a catalyst, was added and left to react for another 2 h. The enzyme reaction was stopped irreversibly by thermal treatment, subjecting the hydrolysate to a temperate of 75 °C for 15 min. The protein hydrolysates were centrifuged at 8000 rpm for 30 min, obtaining a supernatant that was cold-stored until use and analysis.

C. Ordo´n˜ez et al. / Bioresource Technology 99 (2008) 4749–4754

Four different protein hydrolysates were produced depending on the substrate concentration and the enzyme dosage used:

4751

summary table which included the F-value and the degree of significance or probability. 3. Results and discussion

Hydrolysate 1 (H-1): Substrate concentration 5% protein concentrate/water Enzyme/substrate ratio: 0.01 g Alcalase/g protein concentrate 0.02 g Flavourzyme/g protein concentrate Hydrolysate 2 (H-2): Substrate concentration 8% protein concentrate/water Enzyme/substrate ratio: 0.01 g Alcalase/g protein concentrate 0.02 g Flavourzyme/g protein concentrate Hydrolysate 3 (H-3): Substrate concentration 5% protein concentrate/water Enzyme/substrate ratio: 0.02 g Alcalase/g protein concentrate 0.025 g Flavourzyme/g protein concentrate Hydrolysate 4 (H-4): Substrate concentration 8% protein concentrate/water Enzyme/substrate ratio: 0.02 g Alcalase/g protein concentrate 0.025 g Flavourzyme/g protein concentrate Each of these processes was carried out in triplicate. The degree of hydrolysis was determined following the method described by Adler-Nissen (1986). The next step was to characterise the free amino acids in the hydrolysates. For this the amino acids were determined via absorbency of their dansylated derivatives at 254 nm (Tapuhi et al., 1991), following separation by reversedphase HPLC columns (Fallon, 1990), using a Beckman 126 high resolution liquid chromatography, with a double-channel UV/V diode-array detector and a Supelcosil LC-18 column thermostatized at 25 °C. Finally, the molecular weight of the samples was investigated by gel filtration using an FPLC system (Pharmacia) equipped with a Superose 12 HR 10/30 column. The injection volume was 100 lL, and the elution buffer was 0.02 M sodium phosphate (pH 7) and 0.02% NaN3. Elution was carried out at a flux of 0.75 mL/min, using a UV detector at 214 nm. The approximate molecular weight of the SEP was determined using pig heart lactate dehydrogenase (145 900), hen ovotransferrin (78 000), bovine erythrocyte carbonic anhydrase (30 000), aprotinin (6500), and adrenocorticotropic hormone fragment 1–14 (1681 Da), as molecular weight standards (Parrado et al., 1993). 2.3. Statistical treatment The software package Statgraphics Plus 3.1 (Statistical Graphics Corporation, 1994) was used for statistical analysis. Single factor and multi-factor analysis of variance were applied to the data populations involved, according to the LSD criteria with a 95% confidence level. Where significant differences appeared, the results were presented in a

Table 2 shows the degree of hydrolysis accumulated in the four hydrolysates using endoprotease (Alcalase) and the analysis of variance performed using the type of hydrolysate as a variation factor. Greater degrees of hydrolysis were found in hydrolysates H-1 and H-3 which were obtained from substrate concentrations of 5% while substrates with a concentration of 8% had a lower degree of hydrolysis. The fact that higher substrate concentrations gave rise to a lower degree of hydrolysis could either be the result of the higher substrate concentration itself or of the existence of an irreversible endoprotease inhibitor in the substrate (Weber and Nielsen, 1991). This inhibitor acts by bonding irreversibly with an active enzyme fraction in a much shorter time than necessary for hydrolysis (Guadix et al., 2000). Higher enzyme concentration seems to have a marked influence particularly when lower substrate concentrations are used (H-1 and H-3), which corroborates our previous observations on the inhibitor effect. With regards to the duration, hydrolysis was at least 90% complete after just 1 h. There were no significant differences between the various hydrolysates in terms of the degree of hydrolysis at the 2 h point. If we only consider the hydrolysis carried out using exopeptidase, then there were no significant differences in intensity between the four types of hydrolysates. The reaction of the exopeptidase (flavourzyme) is of little importance and is carried out in 30 min. If it is compared with the endopeptidase (Alcalase), its effect is very small in that in each of the cases, the degree of hydrolysis increases less than 1%. Only in the case of H-1 does the degree of hydrolysis increase between 4% and 5%. The final data reflect lower degrees of hydrolysis than those obtained by Pacheco et al. (1994) and Villanueva et al. (1999); however, it should be noted that while Pacheco and Villanueva used a protein isolate, we used a protein concentrate with a lower protein concentration. According to the results obtained for the degree of hydrolysis, it seemed logical to use 5% substrates and an enzyme/substrate ratio of 0.02 g/g concentrate in the case of Alcalase (endopeptidase), whereas for the Flavourzyme (exopeptidase), the ratio seemed not to have a noticeable effect. Tables 3 and 4 show the free amino acid composition of the four hydrolysates obtained, expressed in mmoles of A.A./L of hydrolysate and in mg of A.A./L of hydrolysate, respectively. Twenty amino acids were quantified: two sulphurous (methionine and cysteine), eight hydrophobic (alanine, valine, leucine, isoleucine, proline, methionine, phenylalanine and tryptophan), seven polar (glycine, serine, threonine, cysteine, asparagine, glutamine and tyrosine), two acidic (aspartic and glutamatic acid) and three basic (lysine, arginine and histidine). With the data

C. Ordo´n˜ez et al. / Bioresource Technology 99 (2008) 4749–4754

4752

Table 2 Accumulated hydrolysis degree (%) for Alcalase action 30 min

1h

2h

H-1 23.6 ± 0.8 b H-2 23.0 ± 0.36 a b H-3 30.4 ± 0.26 e H-4 22.8 ± 0.36 a Source of variation

30.4 ± 0.17 e 26.4 ± 0.26 c 34.6 ± 0.28 g 26.6 ± 0.31 c Factor of variation

Hydrolysis degree

Hydrolysate type

33.8 ± 0.26 28.8 ± 0.88 37.0 ± 0.75 28.6 ± 0.52

2 h 15 min f d h d

33.8 ± 0.56 29.0 ± 0.30 37.4 ± 0.60 28.8 ± 0.27

2 h 30 min f d h d

F

– 29.1 ± 0.42 d 37.5 ± 0.38 h 28.8 ± 0.32 d Significant level

356.30

0.0001 (***)

Mean values of the triplicates ± standard errors. Different letters indicate a significant difference at p < 0.05. Hydrolysate type: H-1, H-2, H-3 and H-4 (***) p < 0,001; (**) p < 0,01; (*) p < 0,05.

Table 3 Free aminoacids content (mM) of protein hydrolysated

Asn Gln Asp Ser Glu Gly Thr Ala Arg Pro Met Val Trp Ile + Phe Leu Cys Lys His Tyr

H-1

H-2

H-3

H-4

0.019 ± 0.002 0.190 ± 0.015 52.8 ± 4.96 0.518 ± 0.038 1.35 ± 0.094 3.70 ± 0.124 1.72 ± 0.120 4.41 ± 0.350 0.150 ± 0.005 2.68 ± 0.0.208 1.33 ± 0.093 4.90 ± 0.231 0.954 ± 0.083 5.3 ± 0.330 4.34 ± 0.316 0.762 ± 0.041 0.588 ± 0.038 0.385 ± 0.026 0.717 ± 0.025

0.150 ± 0.008 0.147 ± 0.009 68.6 ± 4.33 0.705 ± 0.015 2.65 ± 0.096 3.19 ± 0.111 2.20 ± 0.106 3.17 ± 0.086 0.150 ± 0.011 4.37 ± 0.350 0.498 ± 0.040 2.73 ± 0.033 1.40 ± 0.041 6.5 ± 0.155 2.65 ± 0.065 1.52 ± 0.063 0.405 ± 0.030 0.383 ± 0.028 0.265 ± 0.015

0.165 ± 0.006 0.229 ± 0.010 60.6 ± 5.18 1.92 ± 0.063 3.43 ± 0.112 2.97 ± 0.036 2.67 ± 0.106 8.98 ± 0.368 0.396 ± 0.022 4.06 ± 0.138 0.700 ± 0.061 1.87 ± 0.082 1.70 ± 0.096 11.9 ± 0.312 1.23 ± 0.061 0.682 ± 0.027 1.47 ± 0.077 2.96 ± 0.106 1.75 ± 0.042

3.27 ± 0.174 0.445 ± 0.022 79.1 ± 6.16 2.97 ± 0.108 3.71 ± 0.132 5.70 ± 0.486 7.74 ± 0.262 13.22 ± 0.315 0.837 ± 0.041 3.00 ± 0.080 4.15 ± 0.192 11.82 ± 1.16 4.04 ± 0.180 23.3 ± 1.45 3.24 ± 0.075 1.97 ± 0.096 3.80 ± 0.085 1.90 ± 0.055 3.18 ± 0.128

Mean values of the triplicates ± standard errors.

Table 4 Free aminoacids content (mM) of protein hydrolysated (mg/L)

Asn Gln Asp Ser Glu Gly Thr Ala Arg Pro Met Val Trp Ile + Phe Leu Cys Lys His Tyr Total content

H-1

H-2

H-3

H-4

2.19 24.3 5957 44.9 174 211 173 313 23.4 260 175 486 178 696 491 78.6 75.5 52.8 117 9532.69

18.45 18.8 7829 61.4 342 182 223 225 23.4 424 65.3 271 260 848 300 157 51.9 52.5 43.2 11 396

19.02 29.3 6911 167 443 169.1 270 637 61.78 395 91.8 186 317 1548 139 70.3 189 406 285 12 234

377 57.0 9025 258 479 324 783 940 131 291 545 1172 752 3034 367 203 486 260 519 20 003

Mean values of the triplicates.

obtained, a multi-factor analysis of variance was carried out taking into account the different concentrations of each amino acid as a source of variation, and the different substrate concentrations and enzyme/substrate ratios as variation factors. None of the samples produced a variation with a significance level over 95%. Degrees of significance of over 90% were only found for the aspartic acid related to the substrate concentration (p = 94.96%) and the enzyme/substrate ratio (p = 90.57%) and serine (p = 94.85%), the isoleucine + phenylalanine mixture (p = 94.36%) and arginine (p = 93.28%) in relation to the enzyme/substrate ratio. The fact that a marked, although not significant, increase was found in the levels of serine when the enzyme/substrate ratio was increased seems to suggest that the aforementioned inhibitor effect occurs on serine-proteinases, in line with the conclusions drawn by Weber and Nielsen (1991) and Guadix et al. (2000). The amino acids that were present in the highest proportion were the aspartic acid (over 50% in the four hydrolysates), followed by the isoleucine + phenylalanine mixture,

C. Ordo´n˜ez et al. / Bioresource Technology 99 (2008) 4749–4754

which is more striking when expressed in mg/L (Table 4). In contrast, the lowest proportions found were for glutamine in hydrolysates H-2 and H-4, while in hydrolysates H-1 and H-3 asparagine had a strong presence. Owing to their aspartic acid, alanine and proline content, these hydrolysates could be useful for the protection of plants against low temperatures (Naidu, 1991). Their isoleucine + phenylalanine content means they could be useful as iron chelating agents or synthetic iron chelating agents (Mehrotra, 1993). If we observe, on the other hand, the total free amino acids, expressed in mg/L, the greatest concentration occurs in H-4, followed by H-2 and H-3 with similar values, and finally H-1, owing to the higher substrate concentration used in hydrolysis. With regards to the presence of low molecular weight amino acids, H-4 contained the highest percentage of amino acids with a weight molecular of less than 100 (serine, glycine, alanine, proline and valine), followed by H-3, H-2 and H-1. This indicates that higher enzyme concentrations favour the presence of these low molecular weight amino acids, and therefore provide greater potential for agricultural use on foliage. Finally, Fig. 2 shows the results obtained using FPLC gel filtration. We can see high molecular weight compounds were not found in any of the hydrolysates and that the main low molecular weight protein components (30 000– 13 000) were found in H-1 and H-3 and (30 000–7000) in H-2 and H-4. H-1 and H-3 contained the highest proportion of peptides of a molecular weight