Characterization of biochemical and functional properties of water-soluble tempe flour

Tempe is popular fermented food in Ondonesia. The health benefits of tempe have been reported (Astuti et al., 2000). Zhan & Ho (2005) reported that tempe could significantly reduce total cholesterol, LDL, and triglycerides content in blood. Tempe was also reported to have a hepatoprotective effect (Mohd Yusof et al., 2013), ACE-inhibitory and antioxidant activity (Gibbs et al., 2004), and immunological impact on intestinal mucosa (Soka et al., 2014).


Introduction
Tempe is popular fermented food in Ondonesia.The health benefits of tempe have been reported (Astuti et al., 2000).Zhan & Ho (2005) reported that tempe could significantly reduce total cholesterol, LDL, and triglycerides content in blood.Tempe was also reported to have a hepatoprotective effect (Mohd Yusof et al., 2013), ACE-inhibitory and antioxidant activity (Gibbs et al., 2004), and immunological impact on intestinal mucosa (Soka et al., 2014).
Tempe is also known to contain high quality and high digestibility of protein.Mice fed with tempe flour showed significantly higher feed conversion ratio and true protein digestibility than mice fed with soy flour (Astawan et al., 2015).Soybean as the raw material of tempe is rich in essential amino acids compared to other plant protein sources (Soares et al., 2005).During fermentation, the quality of the protein is improved.Partially hydrolyzed protein with high digestibility is produced and allergenicity of the protein is reduced after fermentation (Chang et al., 2009;Wilson et al., 2005).The antinutrient compounds of soybean was also reduced during tempe processing (Haron & Raob, 2014).
Tempe has a short shelf life due to continuous fermentation which may lead to discoloration and production of unpleasant flavor from ammonia (Nout & Kiers, 2005).Several studies related to this problem have been released, such as innovation in extending tempe shelf life by combining steam blanching at 80 °C (three minutes) and vacuum packaging (Astawan et al., 2016a) and tempe flour production.However, the application of tempe flour has been limited.Processing tempe into water-soluble flour for food ingredient can be an alternative approach to expand its utilization.
Production of water-soluble flour based on tempe might improve the utilization of protein in tempe.Water-soluble flour of tempe might be applied widely as a food ingredient.Ot can be used as a protein source or applied to improve the functionality of foods.Zayas (2012) reported that polypeptides in a smaller size can provide better functionality on the food system.Water-soluble flour based on germinated-soy tempe is also an interest to be analyzed.Zieliński (2003) reported that germination increased the protein content of soybean, while carbohydrate and lipid contents were reduced (Shi et al., 2010).Dur previous study also found that germinated-soy flour had better antioxidant activity and functionality than soy flour (Astawan & Hazmi, 2016).This research was aimed to study the characteristics of water-soluble flour of tempe (made of soy and germinated-soy) and to compare them with those of commercial soy protein isolate and water-soluble flour of soy.
was used to compare the products.Commercial powdered tempe starter (RAPROMA, PT Aneka Fermentasi Ondustri, Bandung, Ondonesia) for tempe production was purchased from Rumah Tempe Ondonesia, Bogor, Ondonesia.All other chemicals used in the analyses were of analytical grade.

Preparation of tempe
Production of tempe was carried out in Rumah Tempe Ondonesia, Bogor by the standard method of the company.Soybeans were sorted, soaked for 2 hours in water, boiled for 60 min then soaked for 24 hours, dehulled, poured by water (100 °C ), drained, inoculated by starter (15 g per 10 kg soybeans), packed in perforated polypropylene bags (25 cm x 12 cm), and fermented for 40 hours at 30 °C.For germinated-soy tempe, soybean was germinated before processed into tempe.Soybeans were soaked in water at room temperature (27 °C) for 3 hours.Following the draining process, soybeans were then moved into a perforated container and watered (1:5 w/v) every 4 hours for 20 hours at room temperature to allow the radicle to grow between 0.5 and 2.5 mm.

Preparation of defatted flour
Soybean, soy tempe, and germinated-soy tempe were processed into flour based on the method of Dmosebi & Dtunola (2013) with slight modifications.Soybeans were dried in a cabinet dryer (Engineering & Equipment GmbH, 6072 Dreieich, West Germany) at 60 °C, and milled by a disc mill with 60 mesh sieve.Soy tempe and germinated-soy tempe were initially sliced to a thickness less than 0.5 cm by a slicer (ALEXANDERWERK, UC OO, Montgomeryville, Pennsylvania) and steamed for 2 min before dried in a cabinet dryer at 60 °C.Dried soy tempe and germinated-soy tempe were then milled by a disc mill with 60 mesh sieve.The flours of soy, soy tempe, and germinated-soy tempe were then suspended in n-hexane (1:3 w/v) and stirred for 1 hour at room temperature.The solvent was then separated from the precipitate.Following two times of extraction, the precipitate was left in a fume hood to vaporize the solvent.

Extraction
Defatted flour was suspended in water (1:10 w/v) and the pH was adjusted to 9 by using 10 N NaDH.The solution was then stirred for 18 hours at 25 °C and centrifuged for 30 min (BECKMAN, J2-MC, Minnesota, rotor JA-14) by 9000 rpm at 4 °C.The supernatant was collected and dried by a vacuum freeze dryer (EYELA,Tokyo Rikakikai Co.,Ltd.,Tokyo,Japan).The flour was stored at 4 °C for analyses.

Proximate composition
Sample chemical compositions were determined by proximate analysis according to the method of ADAC (2012).Moisture and ash contents were determined by the gravimetric method (ADAC 925.09 and ADAC 923.03).Protein content was measured by the Kjeldahl method (ADAC 955.04D).Fat content was determined by Soxhlet method (ADAC 922.06).Carbohydrate content was calculated by difference.

Trypsin inhibitor activity
The trypsin inhibitor activity was analyzed following the method of Hummel (1959).Trypsin (porcine pancreas) and TAME (p-toluenesulfonyl-L-arginine methyl ester) purchased from Wako Pure Chemical Ondustries, Ltd. (Dsaka, Japan) were used as the enzyme and substrate, respectively.The sample was dissolved in a 10 mM Tris-HCl buffer (pH 8.0) at a concentration of 1 mg/mL buffer.The trypsin was also dissolved in the same buffer at a concentration of 1 mg/ml buffer.Before measurement, the sample solution and the trypsin solution were mixed at a ratio of 1:1 (v/v) and incubated for 5 min at 30 °C Substrate solution made of 1.2 ml Tris-HCl buffer, 1.8 mL TAME (1.74 mM), and 3 µl CaCl 2 (1 M) was also incubated for 5 min at 30 °C.For measurement, 5 µL solution of the sample-trypsin solution was added into the substrate solution, and the mixture was kept at 30 °C.The absorbance (247 nm) at just after mixing (0 min) and at 3 min after the mixing was recorded.The trypsin activity of the sample was obtained by the Equation 1: where Δt is the time difference in recording absorbance (3 min).

Phytic acid content
Phytic acid content was measured by colorimetric method based on Lai et al. (2013).Sample (5 g) was extracted in 2.4% HCl (100 mL) and centrifuged (EPPENDDRF, 5810 R, Germany) at 3000 rpm for 30 min at 25 °C.The supernatant (3 ml) was mixed with 1 mL of the Wade reagent (0.03% solution of FeCl 3 ⋅ 6H 2 D containing 0.3% sulfosalicylic acid) and centrifuged at 3000 rpm for 30 min at 25 °C.The absorbance of the mixture was measured with a spectrophotometer (THERMD FOSHER SCOENTOFOC, 4001/4, Waltham, Massachusetts) at 500 nm.The mixture of 1 ml Wade reagent and 3 ml distilled water was used as the blank.The concentration of the phytic acid was calculated by a standard curve method.

Antioxidant activity (DPPH assay)
Based on the method of Barreira et al. (2008), 150 µL sample solution (10 mg/ mL) was mixed with a 1 mL DPPH (Wako Pure Chemical Ondustries, Ltd., Dsaka, Japan) solution (2.7 mg DPPH in 100 mL methanol) and incubated in a dark room for 20 min at room temperature.The mixture was then centrifuged with the speed of 10000 × g for 5 min (4 °C using centrifuge (KDKUSAN,Japan).The absorbance at 517 nm was taken.The antiradical activity to DPPH was calculated using the Equation 2: Abs control -Abs sample Antiradical activity on DPPH (%) = ×100 Abs control (2)

Foaming properties
The foaming capacity and stability were measured by the method of Klompong et al. (2007).The sample solution (60 mL, 3% w/v) was homogenized with a blender (MOYAKD, BL-101 PL, Ondonesia) for 1 min.The solution was then placed in a 100 mL graduated cylinder and the volume of the foam was recorded.The foaming capacity was expressed using Equation 3: where V 0 = initial volume and V 1 = volume after homogenization.On addition, the foam stability was obtained as a percentage of the volume of retained foam after 20 minutes (Yin et al., 2008).

Emulsifying properties
The emulsifying activity index (EAO) and the emulsion stability index (ESO) were measured by the method of Klompong et al. (2007).As much as 300 mg sample was solved in 30-mL distilled water.The solution was then mixed with olive oil (10 mL) and homogenized by using blender with the speed of 20000 rpm for 1 min.Ommediately after homogenization, 50 µL emulsion in the bottom of the container was taken and mixed with 5 mL sodium dodecyl sulfate (SDS) 0.1%.The absorbance at 500 nm was measured by a spectrophotometer (THERMD FOSHER SCOENTOFOC, 4001/4, Waltham, Massachusetts).The EAO was calculated using Equation 4: while the ESO using Equation 5: where A 0 = absorbance after homogenization; A 10 = absorbance at 10 min after homogenization; Δt = 10 min; and ΔA = A 0 -A 10 .

Protein digestibility (in vitro)
Determination of protein digestibility (in vitro) was conducted by using pepsin and pancreatin from Sigma-Aldrich Co.LLC Abdel-Aal (2008).Sample solution (1.5 g/30 mL) was mixed with pepsin solution (pH 1.9) and incubated for 30 min at 37 °C.Following the incubation, pH was altered to 7.5 by using NaDH.Pancreatin solution was then added and solution was incubated for 6 h at 37 o C. The mixture was then mixed (1:1) with TCA solution (20 g/100 mL) and centrifuged.The soluble protein in obtained the supernatant was then measured by using Lowry method.BSA was used as standard.Soluble protein was compared to total protein.

Statistical analysis
Data were analyzed by ANDVA and differences between means by Duncan test using SPSS (Ver.22,Chicago, OL).Significance was considered at the level of 5%.

Proximate composition
Table 1 shows the chemical composition of SPO and water-soluble flour from soy, soy tempe, and germinated-soy tempe.There is no significant difference in moisture content among samples.Meanwhile, ash, protein, and carbohydrate content of water-soluble flour were significantly different (p < 0.05) from SPO. Ash as water-soluble component was solved during extraction which then resulted in water-soluble flour with high ash content.The fat content of the SPO was like SF, but both values were significantly lower (p < 0.05) compared to those of STF and GTF.

Molecular weight of protein
The electrophoresis patterns are shown in Figure 1.On Figure 1a, the presence of high molecular weight bands is the distinguishing characteristics of non-fermented samples (lane A and B).The fermentation during tempe preparation results in partially hydrolyzed proteins (Bavia et al., 2012).Thus, higher molecular weight polypeptides of STF and GTF (lane C and D) are suggested to be hydrolyzed into lower molecular weight polypeptides as shown in Figure 1b (lane C and D).

Trypsin inhibitor activity
Trypsin inhibitors are antinutrient compounds that interfere with the activity of trypsin.The presence of trypsin inhibitors was shown in Figure 2.
Figure 2 shows lower trypsin activity in SPO, compared to water-soluble flours.Thermal treatment was expected as contributor to the reduction of trypsin inhibitor activity.Radha et al. (2008) reported that thermal process correlated with the reduction of trypsin inhibitor activity.
The high trypsin activity indicated low or no trypsin inhibitor.Several treatments in tempe processing resulted in STF and GTF with no trypsin inhibitor.Egounlety & Aworh (2003) explained that soaking, dehulling and thermal treatment during tempe processing reduced trypsin inhibitors.About 80% of trypsin inhibitor activity from soybean was reduced after tempe processing (Bavia et al. 2012).

Phytic acid
Phytic acid reduces the bioavailability of protein and minerals.Therefore, the presence of phytic acid in soy-based products is undesirable.Figure 3 provides phytic acid contents of SPO and water-soluble flour.The result showed that phytic acid contents among all samples were significantly different (p < 0.05) to each other.The phytic acid content of SF was significantly higher (p < 0.05) than others, while that of GTF was significantly lower (p < 0.05) than others.A reduction of the phytic acid content during the germination has been reported.Rusydi & Azrina (2012) explained that endogenous phytase activity and the leaching out process during soaking could be the reason of the phytic acid reduction during germination.
The fermentation process also had significant contribution to the reduction of phytic acid contents.Water-soluble flour from fermented soy (STF and GTF) showed significantly lower (p < 0.05) phytic acid content to a water-soluble flour from unfermented soy (SF).On accordance with our result, Haron & Raob (2014) reported that tempe processing significantly reduced phytic acid contained in soybean.

Antioxidant activity
Food ingredient with antioxidant activity is suggested to have beneficial effect on health and preservation impact on food system.Figure 4 shows that the DPPH scavenging activities were significantly different (p < 0.05) among all groups.The SPO showed the lowest antiradical activity compared to the water-soluble flours.Meanwhile, the STF and GTF had significantly greater (p < 0.05) antioxidant activity than the SF.During the fermentation of tempe, bioactive peptides were produced (Gibbs et al., 2004).Moreover, according to Chang et al. (2009), tempe had higher antioxidant activity than unfermented soybeans due to isoflavones-derived compound and hydrolyzed peptides produced during fermentation.Production of hydrolyzed peptides in STF and GTF was also shown in this study as evidenced in the SDS-PAGE profiles (Figure 1).2008) also found that germinated soybeans had greater antioxidant activity than non-germinated ones due to bioactive compounds produced during the germination.Germination enhanced the content of ascorbic acid, phenolic compounds, and isoflavones, which were responsible for improving the antioxidant capacity of soybeans (Huang et al. 2014).

Foaming capacity and stability
The foamability of protein is an ability of the protein to trap gas by forming a thin liquid film.Foamability is becoming important for several food systems such as ice cream, cake, and confectionery products.To provide good foaming properties, proteins must be capable of diffusing in an air-water interface.Foaming properties are usually described by a foaming capacity and a foam stability.The foaming capacity (FC) expresses volume of formed foam after homogenization, while the foam stability (FS) expresses volume of remained foam after a specific time.
The FC and FS of samples are indicated in Table 2, respectively.As expected (Kaur & Singh, 2007;Eltayeb et al., 2011), the protein concentration had a correlation with foam properties.The SPO, which is high in protein, showed significantly higher (p < 0.05) FC and FS compared to water-soluble flours.
The fermentation had a positive effect on foam formation.The STF and GTF with hydrolyzed protein performed significantly higher (p < 0.05) FC than SF.Molina Drtiz & Wagner (2002) explained that protein with low molecular size was easily migrated and remained in air-water interface, similar with the current results.However, there is no impact of fermentation on the stability of a foam.The FS of STF and GTF were not significantly different with SF.Previous reports showed that the high FC did not always result in high FS, and vice versa.Jitngarmkusol et al. (2008) explained that big bubble with less flexible surface protein could easily collapse.High surface hydrophobicity is needed to allow the formation of stable foam (Molina Drtiz & Wagner, 2002).Moreover, lipid might disturb the stability of the film which is formed by proteins (Kinsella, 1979).Therefore, defatting process is important for the foam properties.These results suggest that the STF and GTF had a chance to be developed as food ingredient for food system that requires foaming properties.

Activity and stability of emulsion
High activity and stability emulsion is required for water-oil food system.On this study, the emulsion properties were described by the emulsifying activity index (EAO) and emulsion stability index (ESO) in Table 2.The SPO showed significantly higher EAO and ESO compared to water-soluble flours.High concentration of protein supports the reduction of surface tension (Kinsella, 1979).
The STF and GTF showed significantly higher (p < 0.05) EAO and ESO than SF.The result revealed that the fermentation during the tempe processing improved emulsion properties.A similar result in fermented peanut flour was reported by Yu et al. (2007).They stated that the protein degradation by proteases improved the solubility of proteins and resulted in the exposure of hydrophobic groups.There is no significant difference between STF and GTF, suggesting no effects of germination of soybeans on the emulsion properties.Dverall, the STF and GTF might be developed for water-oil food system such as salad dressing and sausage.

Protein digestibility (in vitro)
The protein digestibility of water-soluble flours and the SPO is shown in Figure 5.According to Sarwar Gilani et al. ( 2012), the protein digestibility is correlated with the amount of trypsin inhibitors and phytic acid in the sample, and the structure of the proteins.On the current study, trypsin inhibitor may have low impacts on protein digestibility, since trypsin inhibitor activity in SPO and all water-soluble flours are low (Figure 2).However, current results have proved that phytic acid and the hydrolyzed form of protein influenced the profile of protein digestibility.Sample with higher phytic acid content showed lower protein digestibility.Partial hydrolysis during tempe preparation might provide peptides which were more accessible to digestive enzymes.Based on the SDS-PAGE profile, the STF and GTF contained hydrolyzed peptides with lower molecular weight compared to SF and SPO.

Conclusion
Dverall, the water-soluble tempe flours showed better functional characteristics and nutritional value than the water-soluble soybean flour.Fermentation during tempe processing resulted in water-soluble flours with smaller protein size, higher antioxidant activity, and lower phytic acid content than the soybean based flour.Water-soluble flour had lower functional properties than soy protein isolate due to low protein content.A significant correlation of fermentation to foaming and emulsion properties was observed.The STF and GTF had a potential to use as food ingredient for food system that requires foaming and emulsion properties.However, optimization of the extraction process is needed to improve the protein content.

Table 1 .
Chemical composition of water-soluble flour and soy protein isolate 1 .

Table 2 .
Foaming properties of water-soluble flour and soy protein isolate 1 .