alexa Recovery of and#946;-Amylase by Hybrid UF Membranes from the Soy Whey Wastewater | OMICS International
ISSN: 2155-9821
Journal of Bioprocessing & Biotechniques

Like us on:

Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on Medical, Pharma, Engineering, Science, Technology and Business
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Recovery of β-Amylase by Hybrid UF Membranes from the Soy Whey Wastewater

Liming Zhao1,2*, Chaoqin Chen1,2, Yongjun Qiu1,2, Jiachun Zhou1,2, Lihua Jiang1,2 and Yue Hugh Guan1,2*
1State Key Laboratory of Bioreactor Engineering, R&D Center of Separation and Extraction Technology in Fermentation Industry, East China University of Science and Technology, Shanghai 200237, China
2Shanghai Institute of Manufacturing Technology & Collaborative Innovation Center, Shanghai 200237, China
Corresponding Author's : Liming Zhao
State Key Laboratory of Bioreactor Engineering
R & D Center of Separation and Extraction Technology in Fermentation Industry
East China University of Science and Technology
Shanghai 200237, China
Tel: +86-21-6425-0829
Fax: +86-21-6425-0829
E-mail: [email protected]
  Yue Hugh Guan
State Key Laboratory of Bioreactor Engineering
R & D Center of Separation and Extraction Technology in Fermentation Industry
East China University of Science and Technology
Shanghai 200237, China
E-mail: [email protected]
Received June 10, 2015; Accepted June 29, 2015; Published July 03, 2015
Citation: Zhao L, Chen C, Qiu Y, Zhou J, Jiang L et al., (2015) Recovery of β-Amylase by Hybrid UF Membranes from the Soy Whey Wastewater. J Bioprocess Biotech 5:236. doi: 10.4172/2155-9821.1000236
Copyright: © 2015 Zhao L, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Bioprocessing & Biotechniques


Large-scale manufacturing of soybean proteins results in huge volume of “wastewater” which contains substantial amount of β-amylase, and the recovery of β-amylase has potential high value. In order to evaluate the feasibility and effect of novel hybrid Ultrafiltration (UF) method against the normal method, the pretreated effluent was concentrated by a low molecular weight cut off (MWCO) 3 kDa UF membrane firstly, terminated at Volume Concentration Ratio (VCR) of ca.7. To this UF retentate was then precipitated by adding 1.2M (NH4)2SO4 to remove other proteins. The supernatant obtained following centrifugation was purified through a further higher 30 kDa MWCO UF membrane, which ended at VCR of ca.10. The prepared was purified 4.8 fold, with an enzyme activity concentration ratio of 61, and is associated with 74% combined recovery for the enzyme activity, giving a final specific activity of 14 U/mg (674 U/mL), of which enzyme activity is fully enough for the industrial applications. Preliminary data of this research indicate that this could be an effective method for treating kinds of related protein wastewaters.

β-amylase; Soy whey wastewater; Concentration; Ultrafiltration; Purification
MWCO: Molecular Weight Cut-Off (Da); VCR: Volume Concentration Ratio; UF: Ultrafiltration; T: Temperature (°C); TMP: Trans-Membrane Pressure (bar); Jv: Permeated volume per unit time and per unit effective filtration area (L m-2 h-1 or LMH); VF: Feed volume (L); VC: Retentate volume (L); VP: Permeate volume (L)
Isolated Soy Protein (ISP) is one of the most important and widely used plant proteins, which in 2013 witnessed a production capacity of ca. 600 k tons in China. Manufacturing one ton of ISP results in ca. 40 tons of whey wastewater, which contains up to 8 g L-1 protein as well as oligosaccharides [1]. Such soybean whey wastewater possesses high Biochemical Oxygen Demand (BOD) values and therefore by legislation cannot be discharged into the environment directly. Presently, there has been no consensus in this food industry on how to effectively deal with such wastewater. Options explored include anaerobic and/or aerobic digestion approaches [2], but the small gain out of such a treatment (e.g. biogas produced out of anaerobic digestion) has been far from offsetting the hefty investment and operating costs.
A substantial proportion of these whey proteins is β-amylase (α-1,4-glucan maltohydrolase; EC., an exo-hydrolase that releases β-maltose from non-reducing ends of α-1,4-linked polyglucans and oligoglucans until encountering the first α-1,6-branching point [3]. While β-amylase mixed with pullulanase has been found to convert starch to high maltose syrup [4].
A sustainable strategy for dealing with such effluent is to regard it as a source of raw materials for further manufacturing processes. The present work intended to develop a proof-of-principle process technology towards manufacturing β-amylase preparation out of the soybean whey wastewater. Whilst the technology developed retained the biological activity of the target enzyme, we have apparently achieved a low-range costing for the conceivable large-volume wastewater. Indeed, sensible use of UF can strike a much sought-after balance between achieving a decent product concentration and purity, handling a large effluent volume, and maintaining a low production cost [5]. For instance, electro-UF has been used successfully for amylase recovery out of food processing wastewater [6] and in isolation of proteases from surimi wash water [7].
Selection of membrane type is crucial to the intended UF centered process application, as well as the membrane chemical properties, configuration and pore size, which affect aspects like protein adsorption on membrane surface and membrane stability [8]. The proteins involved in this application are mostly of hydrophilic nature and include the targeted β-amylase. Our choices for hydrophilic UF membrane included, but not limited to, cellulose acetate, cellulose triacetate, polyethersulfone and nylon M. Among these, polyethersulfone membrane possesses high flux, sound chemical stability (including in the alkaline pH range), and outstanding thermal stability (e.g. the intact membrane element should be autoclavable). In consideration of a number of technical and non-technical facts, a type of polyethersulfone membrane from the designated manufacturer was chosen for application.
Typically, in membrane separation processes, the first stage membranes should be those with highest capacity, i.e., with higher MWCO, and then the membranes with smaller MWCO are usually used to separate the residue liquid in the subsequent steps. Extraordinary, on the basis of the protein molecular weights distribution and other properties of whey effluent and relevant preliminary data (to be discussed in the results), an effective processing was assembled and tested in this study, involving two sequential UF steps and a (NH4)2SO4 salting out step, in which the lower MWCO UF membrane was applied before the one with higher MWCO. The suitable UF membranes were selected, and the befitting operating conditions were selected to recover and purify β-amylase.
Materials and Methods
Pre-treatment of the crude sample
The soybean whey effluent was collected from Shandong Wonderful Industrial Group Co. Ltd (Shandong, China), and then adjusted to pH range from 4 to 9 by 1M HCl or 1M NaOH. After settling for 2 h, a supernatant was obtained following 10-min centrifugation at 3050 g. The procedure of recovery β-amylase from the soy whey wastewater was illustrated in Figure 1.
UF by the centrifuge
For determining the UF operating temperature and optimum pH for preparing Sample 2# (Table 1) for the first UF stage, centrifugal UF experiments were conducted on an Eppendorf 5810R centrifuge at 3050 × g for 20 min at 4°C using Amicon Ultra-15 polyethersulfone membranes of either 3 or 10 kDa MWCO (Millipore, USA), with sample loading volume being 15 mL. The UF operation temperature was adjusted according to the centrifuge set up, and the retentate collected was a counterpart to Sample 2#.
UF by pilot-plant device
Pilot-scale experiments were run with a tangential-flow pilot-plant UF system made of Qiyu Biotechnology Co. Ltd. (Shanghai, China). The set-up contains a 10 L tank and a spiral-wound UF membrane vessel, and the membrane has an effective filtration area of 0.24 m2 (termed as the “1812” module). Polyethersulfone membranes of 3 kDa, 10 kDa and 30 kDa MWCOs (with catalogue numbers QY-UF-3-T-1812, QYUF- 10-T-1812, and QY-UF-30-T-1812 respectively) from the same supplier were used. The TMP was in the 1-3 bar range, and the crossflow rate was controlled at 120 L h-1. The UF flux (as L m-2 h-1 or LMH) is defined as the permeated volume at unit time and for a unit effective filtration area. This permeate flux was measured by timed collection using a graduated cylinder. When needed, samples for permeate and retentate were collected for protein and β-amylase assays. This spiralwound pilot-plant set-up was used for both the first and the second UF operations. After each batch of experiment, the device should be cleaned with 0.04% (w/w) NaOH, 0.02% (w/w) EDTA, 0.05% (w/w) SDS, and de-ionized water respectively, to recover the UF membrane performance. Volume concentration ratio for the UF is defined as Equation (1):
(NH4)2SO4 salting-out
A specified amount of NH4)2SO4 (Sinopharm Chemical Reagent Beijing Co. Ltd, Beijing, China) was added into the first stage UF retentate at a suitably concentration. The pH of this mixture was then adjusted to the range of 4-6.5. After settling for 3 h at 4°C, a supernatant and precipitate were obtained following 10 min centrifugation at ambient temperature and 3050 × g. Volume of the supernatant was then measured in a measuring cylinder following decanting. The precipitate was completely re-dissolved in a suitable volume of deionized water. Enzyme specific activity (as U /mg protein) and the enzyme recovery (in %) in the supernatant were obtained following measuring total protein and the β-amylase activity. This retentate was then adjusted to pH 7 and pumped into the second stage UF.
Further analytical purification
In order to confirm the abundant presence of the target enzyme in the final preparation, further chromatographic purification at analytical scale was operated using a Sephadex G-75 60 × 16 mm gel filtration column (Pharmacia). Sodium acetate buffer (pH 6.0), as mobile phase, was run at 0.5 ml/min flow rate and the eluent monitored at 280 nm for proteins [9].
β-amylase activity and protein concentration measurements
The β-amylase activity was measured following the method described earlier [10], which uses soluble starch as the substrate. One unit of β-amylase activity is defined as the amount of enzyme that liberates 1 μmol min-1 of maltose from soluble starch following 3-min incubation in 25 mM sodium acetate buffer at pH 4.8 and at 25°C. For this assay, 3,5-dinitrosalicyclic acid and soluble starch was from Sinopharm Chemical Reagent Beijing Co. Ltd (Beijing, China). Protein content was measured by the Kjeldahl method through nitrogen determination [11].
Sodium dodecyl sulfate-PAGE (SDS-PAGE)
The purity of samples for β-amylase was further assessed using SDS-PAGE on a 12% polyacrylamide gel under the reducing condition in a PhastSystem unit, with the gel being stained using Coomassie brilliant blue (Amersham Pharmacia Biotech). A number of “standard” proteins with molecular weights spanning from 14.3 to 97.2 kDa were used as calibrants [9]. Acrylamide, bis-acrylamide, low molecular weight protein markers, and the Bradford reagent for this assay were from BioRad (USA).
Statistical analysis
Throughout this work, all experiments were repeated for at least three times. The data were analyzed using Student’s t-test and the 95% confidence intervals were obtained. SPSS 18.0 for Windows was used in the statistical analysis.
Results and Discussion
An outline for the extraction and purification process developed in the present work can be grasped from the first and the second columns of Table 1. The thrust of this work was not on selecting a particular unit operation for achieving a desired purity, but on achieving an effective process technology using robust and preferably simple techniques.
The crude effluent unfortunately is turbid (Table 1), resulting from causes like, (a) prior to the generation of this wastewater, the soybean protein isolate is precipitated at pH 4.5, a typical PI value for the major soybean proteins, and so is prone to further precipitation for slightly shifted solubility of the same proteins; and (b) the total solid material is as high as 2.5% (Table 1). This situation means that, before feeding such effluent into the first UF device, a low-speed centrifugation was desired for removing solid materials and the UF feed pH is better off to be kept away from the pH 4.5 vicinity (Figure 2). Curve a of Figure 2A shows that, the recovered enzyme activity out of this centrifugal supernatant is invariably high irrespective of this effluent pH. So, a befit operation pH depended on the UF performance given below.
Effect of the first stage UF
The starting material (i.e. Sample 1#) effectively is the wastewater, and is a mixture of (a) the depleted soybean whey where the majority of proteins have been precipitated and, to a less extent, of (b) the aqueous washings out of the precedent soybean protein manufacturing process. Overall, components in this effluent are highly diluted. Either for enriching β-amylase or for removing the interfering impurity proteins or both, straightforward use of just one unit operation (such as precipitation) was perceived to be unrealistic. For this reason, our idea was to instate a concentration step for removing a proportionate amount of water, and consequently concentrate both the target enzyme and the impurity proteins.
To follow on introduction part above, UF operation appears to be the most appropriate for this type of applications. Notwithstanding of this, we faced a dilemma in adopting high or low MWCO UF membranes for this, mainly, concentration step. Conceivably, a high MWCO membrane offers advantages including high filtration flux and removal of certain impurity proteins. However, this conventional idea is not suitable for this pretreated clear soybean whey effluent. The target enzyme has a molecular weight ranging from 53-64 kDa [12], as to get exactly what MWCO membrane should be adopted for the first UF, the 3 kDa, 10 kDa and 30 kDa MWCO polyethersulfone membranes were prepared, the 40 kDa membrane was not suit for the retention of β-amylase, also it is less common. Curves b-d of Figure 3 show the instantaneous filtration flux with the 3 kDa, 10 kDa, and 30 kDa MWCO membranes in response to varied VCRs over the first UF operation, what surprised us was that the higher MWCO membranes did not offer higher flux but even worse, instead, the flux of 3 kDa membrane was the highest one. The reason is that, the protein molecular weight of the soybean whey effluent ranges from 14 to 97 kDa (Figure 4, lane 5), the main interfering impurity proteins have molecular weight less than 30 kDa, the key of which is Kunitz trypsin inhibitor, with a molecular weight of 21 kDa [1]. Hence, the usage of high MWCO UF membranes (i.e. 10-30 kDa, which were close to the impurity protein molecular weight) by such effluent was doomed to a terrible membrane fouling. For the enzyme activity recovery in the retentate, three sizes of UF membranes all had been invariably more than 90% even the VCR exceeded 9 (data not given), and the 3 kDa membrane showed a better enzyme activity recovery behavior. Obviously, the low MWCO (3 kDa) UF membrane was preferred for the first filtration.
The subsequent 3 kDa UF flux was very much dependent on the pH (Figure 2) [13]. At pH 5, 6, 7, 8 and 9, the stable permeate fluxes were 18, 32, 51, 30 and 29 LMH, respectively. These results unambiguously show that the feed pH for the first UF ought to be kept at a high value and certainly pH 7 was the one. Furthermore, Curve b of Figure 2A shows that, within the pH 4.5-9 range, the enzyme recovery in the retentate following the 3 kDa MWCO UF was more than 90% only when pH was around 6-7. Combining all these results, the crude soybean whey wastewater was best adjusted to pH 7 before the centrifugal supernatant was being fed to the first UF device.
Similarly, Figure 2C shows that the temperature impacts enzyme recovery significantly. This outcome verified that the 3 kDa MWCO membrane could be operated at ambient temperature.
(NH4)2SO4 salting-out
The main purpose of the first UF operation was to concentrate the crude effluent while the second UF operation was for both concentration and purification. In view of the anticipated scale of the intended process, the membrane fouling problem of the second UF system had to be alleviated by removing the major interfering proteins through imposing an (NH4)2SO4 salting-out step between the two UF operations.
The VCR for the first UF operation had been rationalized to the range of 3.5-10 (Figure 3). Figure 5A further illustrates that this ratio also affects the present (NH4)2SO4 salting-out outcome. On one hand, for keeping the β-amylase recovery >90% in the centrifugal supernatant of the (NH4)2SO4 precipitation step, the first UF operation should be controlled at VCR <9. On the other hand, an increment of this VCR invariably improved the enzyme purity in the (NH4)2SO4-derived supernatant, showing removal of impurity proteins. This improvement is significant with VCR <6 and less significant in the 6-10 VCR range. Considering compositional variation of the crude effluent, VCR ≅7 for the first UF operation was regarded as being appropriate for maximizing the salting-out efficiency. At this terminal point, the total protein concentration in the first UF retentate was 25 ± 3 g/ L and the β-amylase specific activity ca. 3.53 U/mg protein.
The Salting-out processing consists of, (a) adding (NH4)2SO4 up to an appropriate concentration, (b) adjusting the phase system to an appropriate pH, and (c) allowing a settling time period ensued by the low-speed centrifugation. Figure 5B shows that the (NH4)2SO4 concentration had two opposite effects: whilst increments of this concentration always reduce the enzyme recovery from the centrifugal supernatant, it also benefits from an improved β-amylase purity to some extent. In detail, when the (NH4)2SO4 concentration exceeded 1.2M, the enzyme recovery out of the (NH4)2SO4 centrifugal supernatant would drop lower than 90%. The effect for improving the enzyme purity (i.e. the enzyme specific activity) both in the <0.8M and >1.2M (NH4)2SO4 concentration ranges are less significant compared to the 0.8-1.2M (NH4)2SO4 concentration range. Consequently, selection of the (NH4)2SO4 concentration at 1.2M was upheld.
The retentate of the first UF had a value of pH 7 and the subsequent addition of (NH4)2SO4 reduced pH value to ca. pH 6. As PI values of the major interfering impurity proteins are in the pH 4-5 range (e.g. PI of Kunitz trypsin inhibitor is at pH 4.5 and that of lipid oxidase at pH 5.4) [1]), selection of an appropriate pH could preferentially precipitate impurity proteins and then enhance this salting-out effect. As shown in Figure 5C, pH had a trade-off effect between removal of impurity proteins and recovery of the enzyme. For instance, a low pH value favored removal of impurity proteins [as shown by increased β-amylase specific activity in Figurer 5C], but brought down the enzyme recovery in the subsequent centrifugal supernatant. Maintaining the enzyme recovery >90% in the supernatant would require controlling pH >4. In the pH 4-6.5 range, its beneficial effect on the enzyme purity can be divided into 2 sub-ranges: reduction of pH had a more marked effect in the pH 4-5 range, as compared with that in the pH 5-6.5 range. As an optimization, the pH of this slurry was adjusted to pH 4.
To summarize, with the first UF being terminated at ca. VCR=7, the 1.2M (NH4)2SO4 salting-out step and the subsequent pH adjustment of this suspension to 4, our results gave the β-amylase activity recovery of 91% and the removal of ca. 60% impurity proteins.
The second UF operation
Before the (NH4)2SO4 centrifugal supernatant was subjected to the 30 kDa UF (i.e. the second UF operation), its pH needed to be adjusted back from 4 to 7. The reason for this third pH adjustment was at least two-fold. First, the targeted β-amylase is most stable at ca. pH 7 and hence the final preparation ought to be kept at pH 7. Second, there had been a clear trend that UF flux tended to be higher at pH 7 as compared with other pH (Figure 2B).
For the subsequent 30 kDa MWCO membrane (i.e. the second UF), Figure 6 shows that the β-amylase recovery in the retentate had no longer imposed any restrictions on this second VCR termination point. Under this situation, this VCR termination point was determined by considering the effect of the second VCR on the UF flux and the enzyme purity in the second UF retentate. Figure 6 shows that, in the VCR 3-6 range, an increment of this ratio almost had no effect on the enzyme purity, but when VCR >6, an increasing of VCR almost linearly augmented the enzyme purity. We had no intention to go over any VCR >10 termination points, as our intended product for this VCR value had already exceeded the specification for such a product, with an enzyme activity of ca. 674 U/mL, the enzyme specific activity of 14 U/ mg and recovery for the enzyme activity of 92%. These results compare favorably with a leading product by Genencor, OPTIMALT® BBA, and the measured corresponding results being 237 U/mL for an enzyme activity and 2.75 U/mg for an enzyme specific activity. Using filtration flux of the second UF as a check, from VCR 8 to 10, the flux reduced by only 15 LMH whereas from VCR 10 to 12 this flux declined by 41 LMH (Curve a of Figure 3). Combining all these results, selection of VCR 10 for the second UF operation not only exceeded the specification for the intended product, but also was fully justifiable on the bioprocess ground.
The associated biochemical analyses
SDS-PAGE on samples 1, 2 and 3 (Table 1) shows that the soybean whey proteins have molecular weights in the range from 14 to 97 kDa (Figure 6). Cereal β-amylases are monomeric proteins with molecular weights in the range of 53-64 kDa [12]. As illustrated in Figure 4, the single band at about 50 kDa in lane 2 could be confirmed to be the target β-amylase.
The maximal activity of this target β-amylase appears to be at ca. pH 5.5, and this compares consistently with those in the literature [14]. For instance, the maximal activity for a purified soybean amylase is at pH 6.0 [15], that for a β-amylase from the bulbs of G. klattianus at pH 5.5 [9], that for a β-amylase from malted African finger millet (Eleusine coracana) seed at pH 5.0 [16].
The maximal enzyme activity for the presently target β-amylase appears to be at 70°C and is also broadly in agreement with the published results. A soybean β-amylase displays highest enzyme activity at 70°C [15]. A microbial β-amylase extracted from Bacillus polymyxa 26-1 possesses maximal β-amylase activity at 45°C and that from Clostridium thermocellum SS 8 at 60°C [17].
Furthermore, the results demonstrate that the presently targeted soybean amylase is stable in the pH range within 5-7.5, and it has a higher enzyme activity at the temperature range of 55-70°C. However, this β-amylase may well be thermally stable up to 60°C, but over this temperature its activity tends to become permanently lost in a rather sharp fashion.
Compared to the present dilemmas on the soybean protein manufacturing wastewater and/or to the reported technologies on separating β-amylase or comparable products from soybean, the presently developed process entails a combination of two UF operations and (NH4)2SO4 precipitation to extract and partially purify the β-amylase. This proof-of-principle process technology can potentially turn the troublesome large-volume soybean whey wastewater into a profitable product, β-amylase.
In summarize this novel application of hybrid UF membrane process technology comprises of two UF operations and an (NH4)2SO4 saltingout step. Apart from UF, which requires only simple equipment and only low cost chemicals? The arrangement for the two UF operations has been such that membrane-fouling problem is considerably minimized. It will be surprising if a more competitive process technology exists at present time. It is therefore our adamant belief that this new process technology can not only positively impact on this sector of food industry, but also establish itself as a model approach for panoply of other protein products in industry.
This research was supported by the National Natural Science Foundation of China (NO. 31101381) and National High Technology Research and Development Program of China (2014AA021202).

Tables and Figures at a glance


Table icon
Table 1


Figures at a glance


Figure Figure Figure
Figure 1 Figure 2 Figure 3


Figure Figure Figure
Figure 4 Figure 5 Figure 6
Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Article Usage

  • Total views: 12039
  • [From(publication date):
    July-2015 - Sep 19, 2019]
  • Breakdown by view type
  • HTML page views : 8232
  • PDF downloads : 3807