Abstract

Bioconversion (e.g. anaerobic fermentation and compost) is the common recycling method of waste activated sludge (WAS) and its hydrolysis, as the rate-limiting step of fermentation, could be accelerated by protease. However, the commercial protease was unstable in a sludge environment, which increased the cost. An endogenous alkaline protease stable in sludge environment was screened in this study and its suitability for treating the sludge was analyzed. The optimal production medium was determined by Response Surface Methodology as starch 20 g/L, KH2PO4 4 g/L, MgSO4·7H2O 1 g/L, sodium carboxy-methyl-cellulose 4 g/L, casein 4 g/L and initial pH 11.3, which elevated the yield of protease by up to 15 times (713.46 U/mL) compared with the basal medium. The obtained protease was active and stable at 35 °C–50 °C and pH 7.0–11.0. Furthermore, it was highly tolerant to sludge environment and maintained high efficiency of sludge hydrolysis for a long time. Thus, the obtained protease significantly hydrolyzed WAS and improved its bioavailability. Overall, this work provided a new insight for enzymatic treatment of WAS by isolating the endogenous and stable protease in a sludge environment, which would promote the resource utilization of WAS by further bioconversion.

HIGHLIGHTS

  • A novel protease-producing bacterium was isolated from the sludge.

  • The screened protease was highly tolerant to the sludge environment.

  • Treatment of the sludge by the screened protease elevated its bioavailability.

  • The screened protease was a promising biocatalyst for sludge treatment.

Graphical Abstract

Graphical Abstract
Graphical Abstract

INTRODUCTION

As industrialization and urbanization developed rapidly, the amount of wastewater substantially increased significantly (Yang et al. 2010). The activated sludge process was widely used for wastewater treatment, which generated a vast amount of waste activated sludge (WAS). This was enriched with harmful substances (e.g. organic matters, heavy metals and pathogens), which could seriously threat the ecological environment if not handled properly (Rai et al. 2004). In addition to the characteristics of pollution, WAS contains abundant biomass (e.g. protein and polysaccharide), which showed the potential for resource and energy recovery.

Bioconversion or microbial utilization processes are typical sludge recycling methods (e.g. anaerobic fermentation and compost). However, these processes are restrained by the hydrolysis of WAS, which is known to be the rate-limiting step. Generally, complex biomacromolecules (e.g. protein and polysaccharide) and microbial cell walls hinder the release of organics (Lee et al. 2009). Therefore, a pretreatment for disintegrating sludge is needed. Currently, many physical and chemical pretreatment methods have been reported, including ultrasound (Erden & Filibeli 2010), microwaves (Lee et al. 2015), oxidative treatment (Hu et al. 2018), alkaline treatment (Li et al. 2008) and acid treatment (Devlin et al. 2011). Among them, alkaline treatment is a preferable method for its simple device, high efficiency and easy operation (Li et al. 2013). However, it suffers from some limitations, such as high cost, secondary pollution and extreme operational conditions. To diminish the limitations described above, there is a prerequisite to couple alkaline treatment with other methods (e.g. the use of surfactants or enzymes) (Chen et al. 2013; Pang et al. 2020b).

Protein is the predominant organic component in WAS (Yu et al. 2008) and the degradation of protein plays a crucial role in fermentation of WAS, which raised the hypothesis that protease could effectively enhance the biodegradability of WAS. Hence, alkali–protease coupling treatment has drawn much attention for its minimal byproducts and mild operation conditions (Hasegawa et al. 2000; Guo et al. 2016). Gessesse et al. (2003) found that enzymes were abundant in in-situ sludge. Furthermore, Yang et al. (2010) investigated the treatment of WAS using a commercial protease and concluded that the hydrolysis could be significantly enhanced by protease. However, the easy inactivation of the above commercial protease during enzymatic treatment made it economically unfeasible (Odnell et al. 2016). Commercially available enzymes are not derived from bacteria naturally existing in the WAS, thus it made sense that these enzymes were markedly vulnerable to lose activity in the sludge environment (Speda et al. 2017). Consequently, the alternative way to alleviate these limitations has been to isolate endogenous protease producers from in-situ sludge, in order to obtain protease with high activity and stability.

Given the facts above, the present research aimed to provide a robust endogenous protease for sludge pretreatment by isolating indigenous protease-producing bacteria from the sludge. Moreover, the optimal medium for protease production was determined by Response Surface Methodology (RSM). To comprehensively analyze the process of enzymatic treatment of WAS, it was necessary to determine the dynamics of protease activity over time. To the best of our knowledge, there have been few studies considering the dynamics of enzyme activity during its application in WAS. Therefore, we investigated the stability of the obtained protease in the sludge environment and the efficiency of sludge solubilization, which would manifest the application potential of the screened protease in WAS treatment.

MATERIALS AND METHODS

Sludge characteristics

The WAS samples were collected from a local Wastewater Treatment Plant in Chengdu which adopted conventional activated sludge processes to treat 100,000 m3/day of municipal wastewater. Prior to use, the samples were stored at 4 °C. The characteristics of WAS were: pH = 6.79 ± 0.50, total solids (TS) = 12,720 ± 109 mg/L, volatile solids (VS) = 7,130 ± 114 mg/L, TCOD = 11,403.00 ± 450.00 mg/L, SCOD = 50.13 ± 4.80 mg/L.

According to the previous publications, enzymes are inherent in the sludge (Yu et al. 2008). To eliminate the interference of these endogenous enzymes, the sludge samples for enzymatic treatment were heated at 121 °C for 30 min, then centrifuged at 10,000 × g for 10 min. The sediment was re-suspended to the original volume in deionized water, which was then adjusted to pH 10.5 according to properties of the protease.

Isolation and identification of alkaline protease-secreting bacteria

Briefly, 1 mL of the sludge was serially diluted and streaked on a plate containing selective medium (pH 10.0) consisting of 10 g/L casein and 20 g/L agar, followed by culture at 30°C for 48 h. The colonies with a clearance zone were identified as protease-producing ones, which were sub-cultured until a uniform colony was obtained. The isolate was identified by morphology and biochemistry analyses on the basis of Bergey's manual (Holt et al. 1994). Analysis of 16S rRNA gene sequences was further conducted to identify bacterial species (Karan et al. 2011; Sonali et al. 2020).

Preparation of inoculum and protease production

The isolate Bacillus amyloliquefaciens HJ1 was maintained and preserved in selective medium at 4 °C before the inoculation. Protease production was carried out in two steps. First, seed culture (i.e. inoculum) was prepared by inoculating a loop of the isolate into 50 mL of YEPD medium (yeast extract 10 g/L, peptone 20 g/L, dextrose 20 g/L), and then incubated at 40 °C and 180 rpm for 12 h. Second, the seed culture was inoculated (10%, V/V) into 50 mL of basal medium (peptone 10 g/L, glucose 5 g/L, K2HPO4 2.5 g/L, MgSO4·7H2O 0.35 g/L). After incubation at 40 °C and 180 rpm for 48 h, the sample was centrifuged (10,000 × g, 10 min, 4 °C) and the supernatant was recovered as crude protease for further study.

Optimization of the medium for protease production

Selection of suitable nutritional factors by single-factor experiment

Initially, the effects of various carbon sources (lactose, starch and sucrose) and nitrogen sources (yeast extract, casein and beef extract) were evaluated by replacing peptone and glucose in the basal medium. In the investigation of surfactant, sodium dodecyl sulfate (SDS), Tween 80 and sodium carboxymethylcellulose (CMC) were separately supplemented at the concentration of 2.5 g/L into the basal medium based on the previous studies (Liu et al. 2004; Yao et al. 2009; Amore et al. 2015; Baweja et al. 2016). Next, the effect of pH on protease production was investigated by adjusting the basal medium to different initial pH (7.0–12.0) using 2 M NaOH or HCl.

Selection of the significant factors by Plackett–Burman design (PBD)

Following the single-factor experiment, PBD was adopted to rapidly select the significant factors regulating protease production (Liu et al. 2010). The experiment was conducted according to the design shown (Table 2). Analysis of variance (ANOVA) was applied to assess the significance of these variables.

Table 1

Morphological and biochemical characteristics of the isolate HJ1

TestResultTestResult
Shape Rod D-tagatose − 
Gram stain N-acetyl glucose − 
Spore formation Fructose 
Starch hydrolysis Mannitol 
Casein hydrolysis Galactose − 
Amylase Gentiobiose 
Catalase activity Mycose − 
Arginine Arabinose − 
Sodium Citrate Sucrose 
2-Nitrophenyl β-D-galactopyranoside Cellobiose 
 Ribose Phaseomannite 
Sorbitol Soya peptone − 
TestResultTestResult
Shape Rod D-tagatose − 
Gram stain N-acetyl glucose − 
Spore formation Fructose 
Starch hydrolysis Mannitol 
Casein hydrolysis Galactose − 
Amylase Gentiobiose 
Catalase activity Mycose − 
Arginine Arabinose − 
Sodium Citrate Sucrose 
2-Nitrophenyl β-D-galactopyranoside Cellobiose 
 Ribose Phaseomannite 
Sorbitol Soya peptone − 

+, positive results; −, negative results.

Table 2

Plackett–Burman design matrix for evaluating the variables on protease production

RunCasein (g/L)Starch (g/L)pHCMC (g/L)KH2PO4 (g/L)MgSO4·7H2O (g/L)Protease activity (U/ml)
12.5 20 11.0 0.5 707.74 ± 11.43 
10 25 11.0 0.5 676.19 ± 59.54 
10 25 11.0 649.54 ± 14.34 
12.5 20 8.0 0.5 523.87 ± 15.74 
10 20 11.0 705.02 ± 25.91 
10 20 8.0 0.5 439.28 ± 39.18 
12.5 20 11.0 637.02 ± 17.01 
10 25 8.0 0.5 631.86 ± 52.9 
12.5 25 8.0 424.05 ± 37.79 
10 12.5 25 8.0 467.30 ± 27.37 
11 12.5 25 11.0 0.5 471.92 ± 53.29 
12 10 20 8.0 420.24 ± 15.5 
RunCasein (g/L)Starch (g/L)pHCMC (g/L)KH2PO4 (g/L)MgSO4·7H2O (g/L)Protease activity (U/ml)
12.5 20 11.0 0.5 707.74 ± 11.43 
10 25 11.0 0.5 676.19 ± 59.54 
10 25 11.0 649.54 ± 14.34 
12.5 20 8.0 0.5 523.87 ± 15.74 
10 20 11.0 705.02 ± 25.91 
10 20 8.0 0.5 439.28 ± 39.18 
12.5 20 11.0 637.02 ± 17.01 
10 25 8.0 0.5 631.86 ± 52.9 
12.5 25 8.0 424.05 ± 37.79 
10 12.5 25 8.0 467.30 ± 27.37 
11 12.5 25 11.0 0.5 471.92 ± 53.29 
12 10 20 8.0 420.24 ± 15.5 

Optimization of significant factor using central composite design (CCD)

CCD was performed to investigate the simultaneous effects and to optimize the concentration of the significant variables. A five-level-three-factor experiment was performed according to the design shown (Table 4). The experimental design of PBD and CCD, as well as data analysis were carried out using the software Design-Expert Trial (Stat-Ease Inc., USA). To study the relationship between bacterial growth and protease production, the dynamics of cell growth and protease production of the isolate were studied under the optimized conditions. Briefly, batch fermentation was conducted in 1 L flasks containing 250 mL of the optimized medium. The isolate was incubated at 30 °C and 180 rpm. The sample was extracted every 4 h to measure cell density (OD600) and protease activity.

Table 3

Analysis of variance (ANOVA) for Plackett–Burman design

EffectSum of squaresF-valueP-value
Casein 48.33066667 7007.560021 8.8361601 0.0589 
Starch 18.68266667 1047.126101 1.3203703 0.3338 
pH 156.848 73803.88531 93.062771 0.0024 
CMC 39.98933333 4797.440341 6.0493169 0.0909 
KH2PO4 88.39466667 23440.85129 29.557666 0.0122 
MgSO4·7H2−24.656 1823.755008 2.299658 0.2267 
EffectSum of squaresF-valueP-value
Casein 48.33066667 7007.560021 8.8361601 0.0589 
Starch 18.68266667 1047.126101 1.3203703 0.3338 
pH 156.848 73803.88531 93.062771 0.0024 
CMC 39.98933333 4797.440341 6.0493169 0.0909 
KH2PO4 88.39466667 23440.85129 29.557666 0.0122 
MgSO4·7H2−24.656 1823.755008 2.299658 0.2267 
Table 4

Design and the corresponding results of central composite design

Casein (g/L)CMC (g/L)pHObserved (U/ml)Predicted (U/ml)
6.00 2.00 7.00 470.56 ± 111.51 498.05 
6.00 6.00 11.00 716.99 ± 40.18 754.66 
10.00 4.00 12.36 703.94 ± 51.26 737.88 
10.00 4.00 5.64 402.30 ± 151.90 406.89 
10.00 0.64 9.00 675.92 ± 55.54 636.50 
10.00 7.36 9.00 636.75 ± 47.59 629.25 
6.00 2.00 11.00 703.94 ± 51.26 652.42 
14.00 6.00 7.00 443.09 ± 32.85 442.31 
10.00 4.00 9.00 686.26 ± 66.75 655.00 
10 16.73 4.00 9.00 396.58 ± 65.75 390.77 
11 14.00 2.00 7.00 472.19 ± 55.23 467.70 
12 3.27 4.00 9.00 676.19 ± 66.92 635.08 
13 10.00 4.00 9.00 620.43 ± 36.62 655.00 
14 6.00 6.00 7.00 513.54 ± 115.17 510.47 
15 10.00 4.00 9.00 713.18 ± 66.06 655.00 
16 14.00 2.00 11.00 479.26 ± 47.37 515.50 
17 14.00 6.00 11.00 488.78 ± 45.95 494.47 
18 10.00 4.00 9.00 690.34 ± 95.86 655.00 
19 10.00 4.00 9.00 591.87 ± 89.86 655.00 
20 10.00 4.00 9.00 619.89 ± 45.82 655.00 
Casein (g/L)CMC (g/L)pHObserved (U/ml)Predicted (U/ml)
6.00 2.00 7.00 470.56 ± 111.51 498.05 
6.00 6.00 11.00 716.99 ± 40.18 754.66 
10.00 4.00 12.36 703.94 ± 51.26 737.88 
10.00 4.00 5.64 402.30 ± 151.90 406.89 
10.00 0.64 9.00 675.92 ± 55.54 636.50 
10.00 7.36 9.00 636.75 ± 47.59 629.25 
6.00 2.00 11.00 703.94 ± 51.26 652.42 
14.00 6.00 7.00 443.09 ± 32.85 442.31 
10.00 4.00 9.00 686.26 ± 66.75 655.00 
10 16.73 4.00 9.00 396.58 ± 65.75 390.77 
11 14.00 2.00 7.00 472.19 ± 55.23 467.70 
12 3.27 4.00 9.00 676.19 ± 66.92 635.08 
13 10.00 4.00 9.00 620.43 ± 36.62 655.00 
14 6.00 6.00 7.00 513.54 ± 115.17 510.47 
15 10.00 4.00 9.00 713.18 ± 66.06 655.00 
16 14.00 2.00 11.00 479.26 ± 47.37 515.50 
17 14.00 6.00 11.00 488.78 ± 45.95 494.47 
18 10.00 4.00 9.00 690.34 ± 95.86 655.00 
19 10.00 4.00 9.00 591.87 ± 89.86 655.00 
20 10.00 4.00 9.00 619.89 ± 45.82 655.00 

Determination of key properties of the screened protease

The effect of temperature on the screened protease was determined at different temperatures (25–60 °C). The protease activity at 50 °C was considered as 100%. To evaluate the thermal stability of the screened protease, it was incubated at 25–60 °C for 30 min and the residue activity was assayed. Similarly, the optimal pH of the screened protease was determined in different pH buffer (200 mM phosphate buffer (pH 6.0–8.0) and 50 mM borax-NaOH buffer (pH 9.0–11.0)). The protease activity at pH 9.3 was considered as 100%. With regard to pH stability, the screened protease was diluted with different pH buffer solutions and incubated at 40 °C at pH 9.3. Next, the solution was adjusted to pH 10.5 and the residual activity was assayed.

Enzymatic treatment of WAS by the screened protease

The obtained protease was partially purified by ammonium sulfate to reduce the interference of the high COD in the enzyme solution. According to our preliminary test, ammonium sulfate at 70% saturation could salt out almost all protease. After ammonium sulfate precipitation, the sample was centrifuged (10,000 × g, 4 °C, 10 min). The precipitate was dissolved in 50 mM borate buffer (pH 10.5) and dialyzed at 4 °C for 24 h against the same buffer. Then, protease was concentrated using PEG20000 and the concentrated protease was used for enzymatic treatment of WAS. For enzymatic treatment, 5 mL of partially purified protease was added to 45 mL of WAS in a 250 mL flask, with 5 mL of deionized water added as control group. Both groups were incubated in a reciprocal shaker (40 °C and 120 rpm) and sampled over time. Thereafter, the sludge was immediately centrifuged (5,000 × g, 4 °C, 10 min) and the dissolved organic matter (DOM) in the supernatant was determined.

Analytic methods

TS, VS and chemical oxygen demand (COD) were analyzed according to standard methods (APHA et al. 1998). SCOD is defined as the COD of the filtrate through a 0.45 μm membrane. The biomass was determined by measuring OD600 and dry cell weight (DCW). Optical density at 600 nm (OD600) was defined as the light absorbance of the fermentation broth at 600 nm, which was positively correlated with biomass (Francois et al. 2005). DCW was measured according to Bhunia et al. (2012).

The content of soluble polysaccharide was determined by phenol–sulfuric acid method using glucose as the standard, and the content of soluble protein was determined by the Lowry method using bovine serum albumin as the standard (Ibrahim et al. 2015). The COD conversion factors were 1.5 g-COD/g protein and 1.06 g-COD/g polysaccharide (Zhou et al. 2013). Protease activity was determined according to the modified method of Haddar et al. (2009). One unit (U) of protease activity represented the quantity of protease required to release 1 μg of tyrosine. Excitation-emission matrix (EEM) was determined using a Shimadzu RF-6000 spectrofluorometer (Kyoto, Japan). The emission spectra of 220–600 nm (2 nm increment) were scanned at the excitation wavelength of 220–450 nm (5 nm increments) (Pang et al. 2020b). The Parallel factor analysis (PARAFAC) algorithm was used to identify the component of the fluorophore according to Stedmon & Bro (2008).

RESULTS AND DISCUSSION

Isolation and identification of the alkaline protease producer originated from WAS

Among various proteases, alkaline protease was the promising choice for sludge pretreatment due to its high hydrolytic activity and stability, and is widely used in the detergent industry (Yu et al. 2019). In this study, the bacteria were isolated from WAS based on whether they could grow and hydrolyze casein in alkaline environment (pH 10). Four morphologically different protease-producing bacteria were isolated. One strain, named as HJ1, exhibited the biggest size of clearance zone on agar plates and high protease activity (48.00 U/mL) in the broth. Therefore, it was chosen for further studies.

The morphological and biochemical characteristics of HJ1 are listed in Table 1, which showed a closely related characteristic to Bacillus amyloliquefaciens (Balows 1975). Especially, HJ1 grew well and secreted a large amount of protease at pH 8–11 (Fig. S1d), which was a typical characteristic of alkaliphile. Generally, the condition of alkaline treatment of the sludge was pH 10 (Pang et al. 2020b). Hence, it could be expected that the obtained protease would function well in alkaline treatment of WAS. Furthermore, the result of molecular evolutionary relationship revealed that HJ1 showed the highest similarity with Bacillus amyloliquefaciens (Figure 1). Likewise, it was reported that the genus Bacillus was predominant producer of alkaline protease (Deng et al. 2010; Guleria et al. 2016). The 16S rRNA gene sequence was deposited in the GenBank database (Accession number: MT678828) and this strain was named as Bacillus amyloliquefacien HJ1.

Figure 1

Phylogenetic tree showing the relationship between the isolate HJ1 and the related Bacillus strains based on 16S rRNA gene sequence. The number on the node is based on the percentage of the bootstrap value for 1000 repetitions.

Figure 1

Phylogenetic tree showing the relationship between the isolate HJ1 and the related Bacillus strains based on 16S rRNA gene sequence. The number on the node is based on the percentage of the bootstrap value for 1000 repetitions.

Optimization of the medium for protease production

Selecting the suitable nutritional and environmental factors in favor of protease production

For large-scale enzymatic treatment of WAS, a large amount of protease was a requisite. However, protease production was regulated by various conditions (e.g. the carbon and nitrogen sources, C/N ratio, pH and temperature) (Ibrahim et al. 2015). Hence, the composition and concentration of medium should be optimized to increase the yield of protease. As shown in Figure S1, the optimal medium components were starch (278.12 U/mL), casein (106.08 U/mL), CMC (356.21 U/mL) and initial pH 11.0 (552.16 U/mL), respectively. When simple substrates like glucose or peptone were used, protease production dramatically declined. This could be explained by the catabolic repression of simple carbon and nitrogen sources (Ibrahim et al. 2015). In contrast, the casein and starch could not be assimilated directly by bacteria, which stimulated alkaline protease production (Stammen et al. 2007; Geisseler & Horwath 2008). Surfactants were reported to change cellular permeability, so that the alkaline protease could be released out of the cell in a timely manner (Reese & Maguire 1969). In our study, different surfactants had different effects (i.e. promoting or inhibiting) on enzyme production, which was also found in the previous studies (Liu et al. 2004; Yao et al. 2009; Amore et al. 2015; Baweja et al. 2016). In addition, protease production was promoted in alkaline conditions (pH 8–11), which inferred that Bacillus amyloliquefacien HJ1 was an alkaliphilic microorganism. Likewise, alkaliphilic Bacillus sp. NPST-AK15 produced massive amounts of alkaline protease within the pH range from 8 to 12 (Ibrahim et al. 2015). Thus, four variables (starch, casein, CMC and initial pH 11.0) were selected for further study.

Establishing a model for selecting of the significant variables affecting protease production

Based on the above results, various nutrients (e.g. carbon and nitrogen sources) were selected for protease production, which affected protease production at different levels. Hence, PBD was employed to identify the variables significantly affecting protease production (Karan et al. 2011). Six variables (starch, casein, CMC, MgSO4·7H2O, KH2PO4 and initial pH) were studied at two levels (−1 and +1) (Table 2). A first-order equation model was constructed to analyze the significance of these variables based on the experimental data:
formula
(1)
where Y is the yield of protease, X1 is casein, X2 is starch, X3 is pH, X4 is CMC, X5 is KH2PO4 and X6 is MgSO4·7H2O. Based on the correlation coefficient in Equation (1), it clearly demonstrated that there were three variables (casein, starch and MgSO4·7H2O) negatively correlated with the yield of protease, while three others (pH, CMC and KH2PO4) correlated positively. As shown in Table 3, the P-value of casein, pH and CMC were less than 0.1, implying that they significantly affected protease production (Karan et al. 2011). Thus, they were selected for further optimization using CCD.

Establishing a model for accurately predicting protease production

It was reported that interactions between factors significantly affected protease production (Reddy et al. 2008). However, the constructed first-order equation model by PBD only contained linear terms and neglected the interactions between variables. Thus, CCD was further used to analyze interactions among variables (casein, pH and CMC) and to determine the optimal condition. The experimental results along with their corresponding predicted values were presented in Table 4. Based on the experimental data, a second-order polynomial equation was constructed to illustrate the relationship between protease activity and the independent variables:
formula
(2)
where Y is the predicted protease yield, X1 is casein, X2 is CMC and X3 is initial pH.

Furthermore, ANOVA was performed to evaluate the statistical significance of the independent variables and the fitted quadratic equation (Table 5). Generally, a larger F-value and smaller P-value (P < 0.1) suggested a more significant independent variable (Haaland 1989). Hence, it could be found that X1 -casein, X3-initial pH, X12, X32, X1X3 were significant. According to the evolution-economy principle (Allison et al. 2011), bacteria would elevate protease production when simple nutrients were scarce and complex nutrients were abundant. In this study, bacteria cannot directly utilize complex nitrogen source (i.e. casein), which stimulated the synthesis of protease. In addition, the isolated alkaliphile relied heavily on an alkaline environment for cell growth and enzyme production. Furthermore, it was reported that the structure of casein became more refractory at lower pH, but looser at higher pH (Liu & Guo 2008). It was clear that the pH-dependence of casein affected its availability. Consequently, the interactive effect between pH and casein was significant. Three-dimensional response surface and contour graph were further plotted to gain a visual interpretation of the interactive effect of variables on protease production (Figure 2). As expected, there were significant interactions between casein and pH and a clear peak inside the design boundary conditions, which indicated that the obtained optimal conditions were credible. The optimal level of three variables were as follow: initial pH 11.3, CMC 4 g/L, and casein 4 g/L, respectively. It also proved that RSM was an efficient method for protease optimization, which achieved a 15-fold increase in the optimized medium (713.00 U/mL) compared with the basal medium (48.00 U/mL). It was reported by Karan et al. (2011) that protease production from Geomicrobium sp. EMB2 was improved from 37 U/ml to 721 U/ml by means of RSM.

Table 5

Analysis of variance (ANOVA) for the variables and the regression model

SourceSSDFMSF-valueP-value
Model 222,931.01 9.00 24770.11 10.89 0.0004 
A-casein 72,049.47 1.00 72049.47 31.69 0.0002 
B-CMC 63.40 1.00 63.40 0.03 0.8707 
C-pH 72,771.40 1.00 72771.40 32.01 0.0002 
AB 714.72 1.00 714.72 0.31 0.5874 
AC 18,438.14 1.00 18438.14 8.11 0.0173 
BC 9.47 1.00 9.47 0.00 0.9498 
A2 36,363.01 1.00 36363.01 15.99 0.0025 
B2 881.81 1.00 881.81 0.39 0.5474 
C2 28,304.40 1.00 28304.40 12.45 0.0055 
Residual 22,735.51 10.00 2273.55   
Lack of Fit 10,722.37 5.00 2144.47 0.89 0.5481 
Pure Error 12,013.14 5.00 2402.63   
Cor Total 245,666.52 19.00    
SourceSSDFMSF-valueP-value
Model 222,931.01 9.00 24770.11 10.89 0.0004 
A-casein 72,049.47 1.00 72049.47 31.69 0.0002 
B-CMC 63.40 1.00 63.40 0.03 0.8707 
C-pH 72,771.40 1.00 72771.40 32.01 0.0002 
AB 714.72 1.00 714.72 0.31 0.5874 
AC 18,438.14 1.00 18438.14 8.11 0.0173 
BC 9.47 1.00 9.47 0.00 0.9498 
A2 36,363.01 1.00 36363.01 15.99 0.0025 
B2 881.81 1.00 881.81 0.39 0.5474 
C2 28,304.40 1.00 28304.40 12.45 0.0055 
Residual 22,735.51 10.00 2273.55   
Lack of Fit 10,722.37 5.00 2144.47 0.89 0.5481 
Pure Error 12,013.14 5.00 2402.63   
Cor Total 245,666.52 19.00    

R2 = 0.9075; Adj R2 = 0.8242; CV = 8.15%; Adeq precision = 10.793; SS, sum of squares; DF, degree of freedom; MS, mean square.

Figure 2

3D-response surface and contour graph of protease production showing the interactive effect between (a and b) casein and pH, hold value: CMC 4 g/L; (c and d) CMC and casein, hold value: pH 9.0; (e and f) CMC and pH, hold value: casein 10 g/L. K2H2PO4 was fixed at a high level (4 g/L) while MgSO4·7H2O and starch were maintained at a low level (0.1 g/L and 20 g/L). Bacillus amyloliquefacien HJ1 was incubated at 30 °C and 180 rpm for 48 h.

Figure 2

3D-response surface and contour graph of protease production showing the interactive effect between (a and b) casein and pH, hold value: CMC 4 g/L; (c and d) CMC and casein, hold value: pH 9.0; (e and f) CMC and pH, hold value: casein 10 g/L. K2H2PO4 was fixed at a high level (4 g/L) while MgSO4·7H2O and starch were maintained at a low level (0.1 g/L and 20 g/L). Bacillus amyloliquefacien HJ1 was incubated at 30 °C and 180 rpm for 48 h.

Growth kinetics and protease production of Bacillus amyloliquefacien HJ1

The dynamics of cell growth and alkaline protease production varied greatly with culture conditions (Geisseler & Horwath 2008). Therefore kinetic analysis was performed in the optimized production medium (starch 20 g/L, KH2PO4 4 g/L, MgSO4·7H2O 1 g/L, CMC 4 g/L, casein 4 g/L and initial pH 11.3). As shown in Figure 3, Bacillus amyloliquefacien HJ1 grew rapidly but scarcely produced protease in phase I. In phase II (12–36 h), the yield of protease increased substantially to 713.46 U/mL. According to above results, it could be concluded that protease production was partially associated with microbial cell growth. Similar dynamics of protease production have been also reported in previous studies (Rao et al. 2009; Jain et al. 2012).

Figure 3

Dynamics of protease production and growth of Bacillus amyloliquefacien HJ1 in the optimal medium (starch 20 g/L, KH2PO4 4 g/L, MgSO4•7H2O 1 g/L, CMC 4 g/L, casein 4 g/L and initial pH 11.3). I: exponential growth phase; II: early stage of stationary phase; III: late-stationary phase.

Figure 3

Dynamics of protease production and growth of Bacillus amyloliquefacien HJ1 in the optimal medium (starch 20 g/L, KH2PO4 4 g/L, MgSO4•7H2O 1 g/L, CMC 4 g/L, casein 4 g/L and initial pH 11.3). I: exponential growth phase; II: early stage of stationary phase; III: late-stationary phase.

The determination of key characterization of the protease for sludge pretreatment

Previous studies have reported that alkali-protease coupling treatment under mesophilic conditions effectively promoted the deflocculation of WAS (Haddar et al. 2009; Jie et al. 2014). Hence, it is a prerequisite for the obtained protease to be stable under alkaline and mesophilic conditions before being applied in sludge treatment. As shown in Figure 4(a), the protease showed marked activity at pH 7.0–11.0 and reached the highest at pH 9.0, which demonstrated that it was a typical alkaline protease (Yu et al. 2019). In addition, the protease retained almost all protease activity at the broad range of pH 6–11 (Figure 4(b)) after being incubated at 40 °C for 30 min. As shown in Figure 4(c) and 4(d), the protease was highly active and stable at broad ranges of temperature from 30 to 60 °C with an optimal protease activity at temperature 50 °C, which was in line with a previous report (Yu et al. 2019). The above results confirmed that the isolated protease was highly stable in alkaline and mesophilic conditions, which ensured high efficiency of sludge treatment.

Figure 4

Effect of pH on protease (a) activity and (b) stability. The pH stability was determined by incubating the protease in different pH buffer solutions at 40 °C for 30 min and the residual activity was assayed. Effect of temperature on protease (c) activity and (d) stability. Thermal stability was determined by pre-incubating the crude protease at 25–60 °C for 30 min and the residual activity was assayed. Relative activity was calculated as the ratio of protease activity under different conditions (pH or temperature) to the optimal condition (i.e. pH 9.3 or 50 °C). Residual activities were calculated as the ratio of protease activity after incubation to the protease activity without incubation.

Figure 4

Effect of pH on protease (a) activity and (b) stability. The pH stability was determined by incubating the protease in different pH buffer solutions at 40 °C for 30 min and the residual activity was assayed. Effect of temperature on protease (c) activity and (d) stability. Thermal stability was determined by pre-incubating the crude protease at 25–60 °C for 30 min and the residual activity was assayed. Relative activity was calculated as the ratio of protease activity under different conditions (pH or temperature) to the optimal condition (i.e. pH 9.3 or 50 °C). Residual activities were calculated as the ratio of protease activity after incubation to the protease activity without incubation.

The dynamics of protease activity and solubilization of WAS during enzymatic treatment

Until recently, many attempts have been made to improve WAS solubilization using protease (Yang et al. 2010; Pang et al. 2020b). However, it was reported that the activity lifetime of commercialized protease in sludge was short (<12 h), which limited the application of enzymatic treatment (Odnell et al. 2016). Indeed, the stability of enzyme in sludge environment was crucial to the practical application. Hence, enzymatic treatment of WAS by the screened protease was conducted to verify its stability and efficiency of sludge solubilization. COD was the sign of WAS solubilization. As shown in Figure 5(a), there were two distinct phases. First, both CODs in enzymatic treatment and control group increased rapidly within the first 12 h. For enzymatic treatment, the net increase in COD was 2,431.06 mg/L, while for control group this was only 1,378.44 mg/L. The reason for the enhanced efficiency of sludge hydrolysis might be that hydrolase breaks down the cell wall of the microorganisms and hydrolyzes biomacromolecules into simpler ones (Song et al. 2010; Yang et al. 2015). Similarly, the endogenous cellulase improved the hydrolysis rate of cellulose and thus elevated the biogas production (Speda et al. 2017). Second, COD trended to decline up to the end of treatment (12–72 h). This phenomenon could be explained by the active cryptic growth of spore-producing microorganisms like Bacillus, whose spores could withstand high temperatures without being killed (Jonghe et al. 2010). Spore-producing microorganisms started to utilize organic nutrients for growth once the spore was activated after 6 h incubation. When the rate of organics utilization was greater than the rate of release, COD started to decline. Meanwhile, the reduction of COD in the enzyme treatment group was significantly higher than in the control group, indicating that more organics were consumed by the microorganisms. Thus, it could be concluded that treatment by the screened protease resulted in higher bioavailability of WAS.

Figure 5

Dynamics of (a) COD, (b) polysaccharide, (c) protein and (d) protease activity in liquid phase of WAS during sludge treatment.

Figure 5

Dynamics of (a) COD, (b) polysaccharide, (c) protein and (d) protease activity in liquid phase of WAS during sludge treatment.

In addition, the variation in DOM was further characterized by the variation in protein and polysaccharide. The protein showed a similar pattern with COD (Figure 5(c)). In the control group, the protein decreased from 893.30 mg/L at 6 h to 590.05 mg/L at 72 h. However, the protein in the enzymatic treatment group decreased rapidly from 3,278.02 mg/L at 6 h to 1,180.10 mg/L at 72 h, which indicated that the addition of protease significantly promoted the utilization of protein. It is worth noting that the total reduction of protein from 6 to 72 h (303.25 mg/L for the control group, 2,097.92 mg/L for enzymatic treatment) was almost equal to that of COD (751.88 mg/L for the control group, 2,681.69 mg/L for enzymatic treatment), which indicated that protein was the main nutriment utilized by the microorganisms in WAS (Yu et al. 2008). The dynamic pattern of polysaccharide was completely different to those of protein and COD (Figure 5(b)). Polysaccharide increased rapidly from 0–12 h, beyond which it remained almost constant. In general, the amount of released polysaccharide was less than that of protein. This was in accordance with the fact that protein is the main organic in WAS (Yu et al. 2008). Similarly, Kavitha et al. (2017) also observed a higher level of release of protein than polysaccharide. Thus, it could be speculated that the elevated bioavailability was mainly caused by the release of protein.

Furthermore, the protease activity in sludge supernatant was also measured to determine the stability of the screened protease in the sludge (Figure 5(d)). In agreement with the trends of COD and protein, the dynamics of protease activity could be distinctly divided into two stages (stage I and stage II). In stage I (0–36 h), the protease activity decreased exponentially with time, while in stage II (36–72 h) the protease activity remained relatively constant. The screened protease had a 47.92% activity at 12 h, 30.33% at 24 h and 10.94% at 72 h. In a previous report, Odnell et al. (2016) treated the sludge with commercial protease (subtilisin and trypsin). They found that the enzymes had almost lost all activity at about 6 h and were completely inactivated after 12 h. The improvement of the stability of the screened protease might be explained by the endogeneity (Speda et al. 2017). The endogenous protease was defined as protease naturally existing in the sludge, which was evolutionarily adapted to the sludge environment. Based on the result above, it could be concluded that the screened protease was more stable than commercial enzymes in the sludge environment. This enabled a lower dosage to achieve the similar effect as commercial protease, resulting in high cost–effectiveness for sludge treatment. In addition, DOM solubilization of enzymatic treatment and bio-utilization by the microorganisms were reported to be two simultaneous and competitive processes (Xin et al. 2016). Thus, protease activity could also be used to explain the change of DOM in sludge. For example, the protease activity in stage I was relatively high, so DOM solubilization exceeded the bio-utilization capacity by the microorganisms, which resulted in the rapidly increased COD. Therefore, it could be concluded that the protease screened from the sludge was highly tolerant to the sludge environment, which will be helpful for enzymatic treatment of WAS.

It was reported that the biodegradability of protein depended on its type and chemical property (Pang et al. 2020a). Thus, EEM was applied to characterize the dynamics of fluorophores (e.g. protein and humid acid). PARAFAC confirmed two main components (Figure 6). Component I (aromatic protein-like substances) located at the Ex/Em of 225/300–350 nm and 275/300–350 nm, which was the preferred substrate for the microorganisms (Luo et al. 2013). Component II (lignin-like substance) located at the Ex/Em of 240/400 nm and 300/400 nm, which could be biodegraded into acid-precipitable polymerized lignin (APPL) (Zeng et al. 2017). With the release of DOM, fluorescence intensities (FI) of components I and II gradually increased within 6 h beyond which FI decreased significantly. In addition, FI of aromatic protein-like substances in enzymatic treatment was significantly higher than control group, which indicated that the screened protease effectively improved the hydrolysis of WAS and promoted the utilization of aromatic protein. A similar phenomenon was also observed by Luo et al. (2013) who found that the treatment of WAS by commercial protease and α-amylase released more protein-like substances, which were the fermentation substrates of volatile fatty acids (VFA). Thereby, enzymatic treatment elevated the bioavailability of WAS. Taken together, the screened protease offered a robust source for alkaline protease-based pretreatment of WAS considering its stability and high efficiency of sludge hydrolysis.

Figure 6

EEM spectra of liquid phase of WAS during enzymatic treatment. All samples were diluted by 200 times.

Figure 6

EEM spectra of liquid phase of WAS during enzymatic treatment. All samples were diluted by 200 times.

CONCLUSION

This study provided a new insight into the isolation of new enzyme and enzymatic treatment of WAS. The protease isolated from native community in the sludge was active and stable in the sludge environment. Furthermore, the yield of protease was elevated using RSM. Treatment of WAS by the screened protease was shown to significantly promote the hydrolysis of WAS and elevate its bioavailability. The screened protease was endowed with superior properties: (1) good catalytic performance and stability over broad pH and temperature ranges, (2) high stability in the sludge environment and (3) an alkaline nature with high production level, laying the foundations for its application as a sludge hydrolysis enhancer.

ACKNOWLEDGEMENTS

This work was supported by the Sichuan Lutianhua Co., Ltd, China [2016-CDLZ-G11-SCU]. JC was supported by the Fundamental Research Funds for the Central Universities [YJ202014], Chengdu Science and Technology Bureau [2019-YF05-01962-SN] and Science & Technology Department of Sichuan Province [2021YFS0292]. We also would like to thank the reviewers for their useful comments to improve this manuscript.

DATA AVAILABILITY STATEMENT

All relevant data are included in the paper or its Supplementary Information.

REFERENCES

Allison
S. D.
,
Weintraub
M. N.
,
Gartner
T. B.
,
Waldrop
M. P.
2011
Evolutionary-economic principles as regulators of soil enzyme production and ecosystem function
. In:
Soil Enzymology
(
Shukla
G.
&
Varma
A.
, eds).
Springer Berlin Heidelberg
,
Berlin, Heidelberg
, pp.
229
243
.
Amore
A.
,
Parameswaran
B.
,
Kumar
R.
,
Birolo
L.
,
Vinciguerra
R.
,
Marcolongo
L.
,
Ionata
E.
,
La Cara
F.
,
Pandey
A.
&
Faraco
V.
2015
Application of a new xylanase activity from Bacillus amyloliquefaciens XR44A in brewer's spent grain saccharification
.
Journal of Chemical Technology and Biotechnology
90
(
3
),
573
581
.
APHA, AWWA and WEF
.
1998
Standard Methods for the Examination of Water and Wastewater
.
American Public Health Association
,
Washington, DC
,
USA
.
Balows
A.
1975
Bergey's Manual of Determinative Bacteriology, 8th edition
.
American Journal of Public Health
65
(
3
),
729
759
.
Baweja
M.
,
Tiwari
R.
,
Singh
P. K.
,
Nain
L.
&
Shukla
P.
2016
An alkaline protease from Bacillus pumilus MP 27: functional analysis of Its binding model toward Its applications As detergent additive
.
Frontiers in Microbiology
7
,
1195
.
Bhunia
B.
,
Basak
B.
,
Bhattacharya
P.
&
Dey
A.
2012
Kinetic studies of alkaline protease from Bacillus licheniformis NCIM-2042
.
Journal of Microbiology and Biotechnology
22
(
12
),
1758
1766
.
Deng
A.
,
Wu
J.
,
Zhang
Y.
,
Zhang
G.
&
Wen
T.
2010
Purification and characterization of a surfactant-stable high-alkaline protease from Bacillus sp. b001
.
Bioresource Technology
101
(
18
),
7100
7106
.
Devlin
D. C.
,
Esteves
S. R. R.
,
Dinsdale
R. M.
&
Guwy
A. J.
2011
The effect of acid pretreatment on the anaerobic digestion and dewatering of waste activated sludge
.
Bioresource Technology
102
(
5
),
4076
4082
.
Erden
G.
&
Filibeli
A.
2010
Ultrasonic pre-treatment of biological sludge: consequences for disintegration, anaerobic biodegradability and filterability
.
Journal of Chemical Technology and Biotechnology
85
(
1
),
145
150
.
Francois
K.
,
Devlieghere
F.
,
Standaert
A. R.
,
Geeraerd
A. H.
,
Cools
I.
,
Van Impe
J. F.
&
Debevere
J.
2005
Environmental factors influencing the relationship between optical density and cell count for Listeria monocytogenes
.
Journal of Applied Microbiology
99
(
6
),
1503
1515
.
Gessesse
A.
,
Dueholm
T.
,
Petersen
S. B.
&
Nielsen
P. H.
2003
Lipase and protease extraction from activated sludge
.
Water Research
37
(
15
),
3652
3657
.
Guleria
S.
,
Walia
A.
,
Chauhan
A.
&
Shirkot
C. K.
2016
Molecular characterization of alkaline protease of Bacillus amyloliquefaciens sp1 involved in biocontrol of fusarium oxysporum
.
International Journal of Food Microbiology
232
,
134
143
.
Guo
W.
,
Zheng
H.
,
Li
S.
,
Hoa
S.
,
Yang
S.
,
Fenga
X.
,
Changa
J.
,
Wang
X.
&
Ren
N.
2016
Promotion effects of ultrasound on sludge biodegradation by thermophilic bacteria Geobacillus stearothermophilus TP-12
.
Biochemical Engineering Journal
105
,
281
287
.
Haaland
P.
1989
Separating signals from the noise
. In:
Experimental Design in Biotechnology
(P. Haaland, ed.)
Marcel Dekker
,
New York
, pp.
61
83
.
Haddar
A.
,
Agrebi
R.
,
Bougatef
A.
,
Hmidet
N.
,
Kamoun
A. S.
&
Nasri
M.
2009
Two detergent stable alkaline serine-proteases from Bacillus mojavensis A21: purification, characterization and potential application as a laundry detergent additive
.
Bioresource Technology
100
(
13
),
3366
3373
.
Hasegawa
S.
,
Shiota
N.
,
Katsura
K.
&
Akashi
A.
2000
Solubilization of organic sludge by thermophilic aerobic bacteria as a pretreatment for anaerobic digestion
.
Water Science and Technology
41
(
3
),
163
169
.
Holt
J. G.
,
Krieg
N. R.
,
Pha
S.
,
Staley
J. T.
&
Williams
S. T.
1994
Bergey's Manual of Determinative Bacteriology
.
Williams & Wilkins
,
Philadelphia
.
Hu
W.
,
Xie
Y.
,
Zeng
Y.
,
Li
P.
&
Wang
Y.
2018
A new approach for excess sludge reduction by manganese dioxide oxidation: performance, kinetics and mechanism studies
.
Environmental Science and Pollution Research
25
,
29356
29365
.
Ibrahim
A. S. S.
,
Salamah
A. A. A.
,
Badawil
Y. B. E.
&
Tayeb
M. A. E.
2015
Production of extracellular alkaline protease by new halotolerant alkaliphilic Bacillus sp. NPST-AK15 isolated from hyper saline soda lakes
.
Electronic Journal of Biotechnology
18
,
236
243
.
Jie
W.
,
Peng
Y.
,
Ren
N.
&
Li
B.
2014
Utilization of alkali-tolerant stains in fermentation of excess sludge
.
Bioresource Technology
157
,
52
59
.
Jonghe
V. D.
,
Coorevits
A.
,
Block
J. D.
,
Coillie
E. V.
,
Grijspeerdt
K.
,
Herman
L.
,
Vos
P. D.
&
Heyndrickx
M.
2010
Toxinogenic and spoilage potential of aerobic spore-formers isolated from raw milk
.
International Journal of Food Microbiology
136
(
3
),
318
325
.
Kavitha
S.
,
Subbulakshmi
P.
,
Rajesh Banu
J.
,
Gobi
M.
&
Tae Yeom
I.
2017
Enhancement of biogas production from microalgal biomass through cellulolytic bacterial pretreatment
.
Bioresource Technology
233
,
34
43
.
Lee
J. H.
,
Lee
J. M.
,
Lim
J. S.
,
Park
T. J.
&
Byun
I. G.
2015
Enhancement of microwave effect with addition of chemical agents in solubilization of waste activated sludge
.
Journal of Industrial Engineering and Chemistry
24
,
359
364
.
Li
H.
,
Jin
Y.
,
Mahar
R.
,
Wang
Z.
&
Nie
Y.
2008
Effects and model of alkaline waste activated sludge treatment
.
Bioresource Technology
99
(
11
),
5140
5144
.
Li
H.
,
Zou
S.
,
Li
C.
&
Jin
Y.
2013
Alkaline post-treatment for improved sludge anaerobic digestion
.
Bioresource Technology
140
(
3
),
187
191
.
Liu
Y.
&
Guo
R.
2008
pH-dependent structures and properties of casein micelles
.
Biophysical Chemistry
136
(
2
),
37
73
.
Liu
J.
,
Chen
X.
,
Dai
X.
,
Tang
B.
&
Peng
Z.
2004
Effect of surfactant on the production of thermophilic protease from Bacillus stearothermophilus
.
Journal of Microbiology
119
(
52
),
58
59
.
61
.
Liu
S.
,
Fang
Y.
,
Lv
M.
,
Wang
S.
&
Chen
L.
2010
Optimization of the production of organic solvent-stable protease by Bacillus sphaericus DS11 with response surface methodology
.
Bioresource Technology
101
(
20
),
7924
7929
.
Odnell
A.
,
Recktenwald
M.
,
Stensén
K.
,
Jonsson
B.-H.
&
Karlsson
M.
2016
Activity, life time and effect of hydrolytic enzymes for enhanced biogas production from sludge anaerobic digestion
.
Water Research
103
,
462
471
.
Rai
C. L.
,
Struenkmann
G.
,
Mueller
J.
&
Rao
P. G.
2004
Influence of ultrasonic disintegration on sludge growth reduction and its estimation by respirometry
.
Environmental Science and Technology
38
(
21
),
5779
5785
.
Rao
C. S.
,
Sathish
T.
,
Pendyala
B.
,
Kumar
T. P.
&
Prakasham
R. S.
2009
Development of a mathematical model for Bacillus circulans growth and alkaline protease production kinetics
.
Journal of Chemical Technology and Biotechnology
84
(
2
),
302
307
.
Reese
E. T.
&
Maguire
A.
1969
Surfactants as stimulants of enzyme production by microorganisms
.
Applied Microbiology
17
(
2
),
242
245
.
Sonali
P.
,
Sukanta
K. P.
&
Pratima
R.
2020
Production, characterization and application of thermostable, alkaline α-amylase (AA11) from Bacillus cereus strain SP-CH11 isolated from Chilika Lake
.
International Journal of Biological Macromolecules
145
,
804
812
.
Speda
J.
,
Johansson
M. A.
,
Odnell
A.
&
Karlsson
M.
2017
Enhanced biomethane production rate and yield from lignocellulosic ensiled forage ley by in situ anaerobic digestion treatment with endogenous cellulolytic enzymes
.
Biotechnology for Biofuels
10
(
1
),
1754
6834
.
Stammen
S.
,
Müller
B. K.
,
Korneli
C.
,
Biedendieck
R.
,
Gamer
M.
,
Francolara
E.
&
Jahn
D.
2007
Enhanced productivity of serine alkaline protease by Bacillus sp. using soybean as substrate
.
Malaysian Journal of Microbiology
3
(
1
),
4037
4046
.
Stedmon
C. A.
&
Bro
R.
2008
Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial
.
Limnology and Oceanography: Methods
6
(
11
),
572
579
.
Yang
Q.
,
Luo
K.
,
Li
X. M.
,
Wang
D. B.
,
Zheng
W.
,
Zeng
G. M.
&
Liu
J. J.
2010
Enhanced efficiency of biological excess sludge hydrolysis under anaerobic digestion by additional enzymes
.
Bioresource Technology
101
(
9
),
2924
2930
.
Yao
G.
,
Cheng
J.
,
Sun
P.
&
Li
N.
2009
Preparation of alkaline protease by fermentation of Bacillus subtilis
.
Food Science
30
(
23
),
347
351
.
Yu
G.
,
He
P.
,
Shao
L.
&
Lee
D.
2008
Extracellular enzymes in sludge flocs collected at 14 full-scale wastewater treatment plants
.
Journal of Chemical Technology & Biotechnology
83
(
12
),
1717
1725
.
Zeng
Z.
,
Zheng
P.
,
Ding
A.
,
Zhang
M.
,
Abbas
G.
&
Li
W.
2017
Source analysis of organic matter in swine wastewater after anaerobic digestion with EEM-PARAFAC
.
Environmental Science and Pollution Research
24
(
7
),
6770
6778
.
Zhou
A.
,
Guo
Z.
,
Yang
C.
,
Kong
F.
,
Liu
W.
&
Wang
A.
2013
Volatile fatty acids productivity by anaerobic co-digesting waste activated sludge and corn straw: effect of feedstock proportion
.
Journal of Biotechnology
168
,
234
239
.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY-NC-ND 4.0), which permits copying and redistribution for non-commercial purposes with no derivatives, provided the original work is properly cited (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Supplementary data