The composition of TS was reported separately (Ratanapariyanuch et al. 2011
). In brief, stillage contained a number of organic and inorganic constituents that constituted a solution with about 3% dissolved matter. Our original hypothesis was that some of the dissolved constituents might either alter the efficiency of protein extraction or affect the quality of the extracted protein. Neither the efficiency of protein extraction nor the quality of protein was affected by the whole stillage. Therefore, we did not have reported the effect of individual components of the stillage on protein yield and quality. In this study, the relative efficiencies of protein extraction using TS and NaCl solution were used to determine the effect of these solutions. In addition, SDS-PAGE of extracted protein, amino acid sequences of tryptic peptide fragments of extracted protein, digestibility, and lysine availability of extracted protein were compared for protein extracted using TS and NaCl solution.
High pH and salt concentrations are not necessarily practical if they are not cost effective even if they increase protein extraction efficiency. At alkaline pH, most proteins have a net negative charge, which results in strong intramolecular electrostatic repulsion. This would cause swelling and unfolding of protein molecules (Damodaran 1996
) and possible loss of functionality. Similarly, when pH is above the isoelectric pH, protein solubility increases. Typically, the maximum solubility of protein occurs in alkaline solutions. Damodaran (1996)
noted that when ionic strength is low (NaCl concentration < 0.5 M) the solubility of proteins that contain polar surface domains typically increases. The effects of pH and salt concentration demonstrated in this study are in agreement with the literature. Lindeboom and Wanasundara (2007)
extracted protein from yellow mustard (Sinapis alba
) using water at different pHs (3.5-10.0). They discovered that the protein content of the extracts increased when the pH was above 7.5, and was as high as 25 mg/ml at pH 10.0.
According to the molecular weights of the proteins extracted by TS and NaCl solution (Figure ), Aluko and McIntosh (2001)
reported that a 52 kDa polypeptide was present in a purified 12S globulin storage protein (cruciferin) from Brassica napus
seed. Aluko et al. (2004)
stated that in S
protein isolates, a 2S albumin storage protein (napin) band appeared at 5 kDa and cruciferin bands at 22, 28, and 35 kDa. Aluko and McIntosh (2004)
demonstrated that 12 and 13 kDa polypeptides were subunits of the napin of mustard seed. Shim and Wanasundara (2008)
reported that a single protein band of 14.5 kDa polypeptides were two polypeptide chains of 4.5 and 10 kDa linked by disulfide bonds. From the above information, it was concluded that the bands found in SDS-PAGE were cruciferin and napin, and that they could be extracted with either TS or NaCl solution. These results were then confirmed by peptide sequencing.
A combination of in-gel trypsin digestion of protein separated by SDS-PAGE followed by MALDI-TOF MS of the digests produced the masses used for searching peptide-mass databases. Table shows the search results, which identifies peptide fragments of B. juncea
. These results are in agreement with those of Aluko and McIntosh (2001)
,Aluko et al. (2004)
,Aluko and McIntosh (2004)
, and Shim and Wanasundara (2008)
, as described above. In addition, using fragment exact mass, the same peptide sequences of cruciferin were separated into different bands by gel electrophoresis. This can be explained by: (1) possible degradation of the extracted protein to smaller molecules by enzyme, pH or hydrolysis during processing and (2) the cruciferin present in rapeseed is a member of the 12S globulins which are hexameric molecules consisting of homologous but non-identical subunits (Tandang et al. 2004
). Surprisingly, no peptides arising from yeast, bacteria or wheat were found. Only napin and cruciferin were identified in the extracted protein. These proteins isolates are, therefore, similar to those prepared from related Brassica
species. The potential exists to process these isolates using hydrolytic enzymes to produce bioactive peptides and antioxidants that may be added to feed and food (Xue et al. 2009
The amino acid composition is comparable to the amino acid composition of proteins isolated from B. juncea
analyzed by Alireza-Sadeghi et al. (2006)
. In addition, the quantity of essential amino acids extracted is sufficient to meet Food and Agriculture Organization (FAO) standards (2002) (Table ). Lysine is frequently the factor limiting the protein quality of mixed diets for human food and animal feed. When the total lysine content was compared with the available lysine content, it was found that approximately 75% of the lysine in the extracted protein would be available in feed. These results agree with those of Larbier et al. (1991)
who found that lysine digestibility of whole rapeseed meal, dehulled rapeseed meal, and soybean meal for cockerels were 80.1, 86.0, and 88.9%, respectively. The available lysine values for chicks were 72.8, 78.3, and 85.5%, respectively. The digestibility of the isolates produced in this study was below that reported by Alireza-Sadeghi et al. (2006)
. The higher reported digestibility may be the result of charcoal adsorption treatment of the mustard protein isolates that was used in that study.
The darker color of protein extracted with TS may be due to the inclusion of colored compounds with the protein or reactions between compounds in the stillage and protein to produce color. In addition, protein extracted with TS could have absorbed colored materials from the alkaline glycerol or TS. Therefore, the compound present in TS may affect the other protein properties such as in vivo digestibility, which were not examined in this study. Consequently, other qualities of the extracted protein should be tested in future studies.
The efficiency of protein extraction is affected by the ratio of meal to solvent, where a higher ratio leads to lower efficiencies. However, the energy required to evaporate water from the protein solution in the final processing step would make the overall process inefficient at low meal to solvent ratios. The ratio of ground defatted meal:solvent (1:30, w/v) used in the preliminary experiments would not be practical for industrial application, thus the use of a higher ratio (1:5, w/v) was evaluated. As expected, the results showed that when the meal: solvent ratio used for protein extraction was increased from 1:30 to 1:5, protein extraction efficiency decreased from 80% to 60%. The efficiency of the protein extraction process developed in this study was compared with that of a published protocol (Milanova et al. 2006
). In the published protocol, the cold-water treatment caused the protein to salt out in micelle form. The percent recovery from the protein micelle was only 7.6%. This protein recovery was significantly lower than the 80% achieved with extractions at pH 10.0 and a NaCl concentration of 1.0 M. It can be concluded that the process developed in this research was more efficient in terms of protein extraction than the published protocol.
In conclusions, a biorefinery process was developed that linked coproducts of bio-ethanol and biodiesel production. TS was used for protein extraction from defatted B. juncea meal, a coproduct of biodiesel production from oilseed. In addition, biodiesel plants can provide alkali to increase pH and protein solubility. Therefore, ethanol, biodiesel, and protein industries benefit from process integration. TS did not affect the efficiency of protein extraction or nutritional qualities of the protein extracts. The use of a byproduct, TS, as a part of a protein extraction process would increase the viability of the linked industrial processes. The current work demonstrates that the protein products of stillage-based extractions are of acceptable quality for use in feeds.