3.1 Single-factor experiments for optimizing γ-PGA production
The steamer bottom water of Baijiu brewing is a mixed liquid formed by the repeated condensation of water vapor in grains during grain gelatinization and fermented grain distillation. It is concentrated at the bottom of a steamer and original water. It contains large amounts of acids, esters, alcohols, starch, sugars, and other organic components, resulting in high COD, turbidity, and suspended solids [24]. It is a typical high-concentration organic wastewater. The synthesis of γ-PGA occurs via fermentation with organic wastewater, such as vermicelli processing wastewater [25], animal manure [26], or papermaking wastewater [27]. Our experimental results show that steamer bottom water can be used to replace water as a fermentation medium substrate to synthesize γ-PGA, with a yield of 24.37 g/L (Fig. 1A).
When 40 g/L dissolved solid in corn saccharifying solution and 6 g/L peptone were used as the supplementary carbon and nitrogen sources, respectively, the γ-PGA yield was the highest [28] (Fig. 1B,C). B. subtilis YB18 is a glutamate-dependent strain. As a substrate, glutamic acid should be supplemented in the medium for γ-PGA production. In the experiment, the addition of monosodium glutamate to the medium was not beneficial to the γ-PGA yield (Fig. 1D); this is likely because the steamer bottom water contains glutamic acid (30 mg/L), which activates the glutamic acid synthesis pathway of the strain that in turn uses its own metabolite glutamate acid as a substrate to produce γ-PGA [29].
The analysis of the effects of Mg2+, Ca2+, and Mn2+ on γ-PGA production via fermentation showed that the addition of K2HPO4 and MnSO4·H2O was beneficial to γ-PGA production. When the added K2HPO4 and MnSO4·H2O were 2.00 and 0.04 g/L, respectively, the γ-PGA yield was the highest (Fig. 1H–I). When MgSO4·7H2O, CaCl2, and NaCl were supplemented in the medium, they were not conducive to γ-PGA production, and the yield was lower than that without these supplements (Fig. 1E–G). This difference might be attributed to the presence of certain amounts of Mg2+, Ca2+, and Na+ in the steamer bottom water. When the fermentation time was 72 h, and the inoculation amount was 2%, the maximum γ-PGA yield was obtained (Fig. 1J–K).
3.2 PB design screening of significant factors affecting γ-PGA production
The significant factors affecting γ-PGA synthesis (P < 0.05) via microbial fermentation included dissolved solids in corn saccharification liquid, peptone, and K2HPO4 (Table 1 and Table 2). Dissolved solid in corn saccharifying liquid and peptone on the γ-PGA yield showed a positive effect, whereas K2HPO4 exhibited a negative effect.
Table 1
Plackett-Burman experimental design and response value (n = 12).
Runs
|
Real levels (coded levels)
|
γ-PGA production (g/L)
|
Inoculation amount (% v/v)
|
K2HPO4 (g/L)
|
MnSO4 (g/L)
|
protein peptone (g/L)
|
corn saccharification liquid (g/L)
|
Fermentation time
(h)
|
Actual
|
Predicted
|
1
|
3.00(+1)
|
1.00(-1)
|
0.04(+1)
|
4.00(-1)
|
20.00(-1)
|
60(-1)
|
18.42 ± 0.445
|
14.86
|
2
|
3.00(+1)
|
2.00(+1)
|
0.02(-1)
|
6.00(+1)
|
20.00(-1)
|
60(-1)
|
19.74 ± 0.064
|
20.42
|
3
|
2.00(-1)
|
2.00(+1)
|
0.04(+1)
|
4.00(-1)
|
40.00(+1)
|
60(-1)
|
17.40 ± 0.056
|
21.19
|
4
|
3.00(+1)
|
1.00(-1)
|
0.04(+1)
|
6.00(+1)
|
20.00(-1)
|
72(+1)
|
20.74 ± 0.334
|
24.41
|
5
|
3.00(+1)
|
2.00(+1)
|
0.02(-1)
|
6.00(+1)
|
40.00(+1)
|
60(-1)
|
28.16 ± 0.056
|
27.48
|
6
|
3.00(+1)
|
2.00(+1)
|
0.04(+1)
|
4.00(-1)
|
40.00(+1)
|
72(+1)
|
20.30 ± 0.032
|
18.09
|
7
|
2.00(-1)
|
2.00(+1)
|
0.04(+1)
|
6.00(+1)
|
20.00(-1)
|
72(+1)
|
24.35 ± 0.445
|
23.67
|
8
|
2.00(-1)
|
1.00(-1)
|
0.04(+1)
|
6.00(+1)
|
40.00(+1)
|
60(-1)
|
35.59 ± 0.056
|
34.57
|
9
|
2.00(-1)
|
1.00(-1)
|
0.02(-1)
|
6.00(+1)
|
40.00(+1)
|
72(+1)
|
41.74 ± 0.032
|
39.75
|
10
|
3.00(+1)
|
1.00(-1)
|
0.02(-1)
|
4.00(-1)
|
40.00(+1)
|
72(+1)
|
25.00 ± 0.085
|
27.11
|
11
|
2.00(-1)
|
2.00(+1)
|
0.02(-1)
|
4.00(-1)
|
20.00(-1)
|
72(+1)
|
20.21 ± 0.032
|
19.31
|
12
|
2.00(-1)
|
1.00(-1)
|
0.02(-1)
|
4.00(-1)
|
20.00(-1)
|
60(-1)
|
22.35 ± 0.032
|
23.14
|
Table 2
Statistical analysis of Plackett-Burman design.
Variables
|
Effect
|
Coefficient
|
T Value
|
P Value
|
Constant
|
|
24.499
|
24.00
|
0.000
|
Inoculation amount
|
-4.878
|
-2.439
|
-2.49
|
0.055
|
K2HPO4
|
-5.612
|
-2.806
|
-2.86
|
0.035*
|
MnSO4
|
-3.402
|
-1.701
|
-1.74
|
0.143
|
protein peptone
|
7.769
|
3.885
|
3.96
|
0.011*
|
corn saccharification liquid
|
7.066
|
3.533
|
3.60
|
0.015*
|
Fermentation time
|
1.782
|
0.891
|
0.91
|
0.405
|
R2
|
97.22%
|
|
|
|
Adj R2
|
94.91%
|
|
|
|
* Significant at P < 0.05.
3.3 Path of steepest ascent
The PB test revealed that the concentrations of dissolved solids in corn saccharifying solution and peptone gradually increased, and the concentration of K2HPO4 gradually decreased. These results were then used to design the steepest ascent test. In the fifth experiment, the γ-PGA yield was the highest when the concentrations of dissolved solids in the corn saccharifying solution, peptone, and K2HPO4 were 60.00, 8.00, and 1.00 g/L, respectively (Table 3).
Table 3
Design and results of the steepest climbing test.
Runs
|
protein peptone
(g/L)
|
corn saccharification liquid
(g/L)
|
K2HPO4
(g/L)
|
γ-PGA production (g/L)
|
1
|
4.00
|
20.00
|
2.00
|
11.18 ± 0.064
|
2
|
5.00
|
30.00
|
1.75
|
24.74 ± 0.064
|
3
|
6.00
|
40.00
|
1.50
|
32.02 ± 0.111
|
4
|
7.00
|
50.00
|
1.25
|
41.36 ± 0.222
|
5
|
8.00
|
60.00
|
1.00
|
49.69 ± 0.193
|
3.4 Optimal combination of significant factors obtained via CCD
A three-level response surface central composite experiment was designed using Minitab 19 software, with peptone, dissolved solid in corn saccharification liquid, and K2HPO4 as the three factors (Table 4). The results of CCD variance analysis showed that the regression model had P < 0.0001, which is credible. The regression model had a lack-of-fit P = 0.0578, which was > 0.05, indicating that the result was not significant and that the model was appropriate (Table 5). The adjusted R2 of the regression equation was 0.988, suggesting that this model could describe 98.8% of the change in the γ-PGA yield. The multiple quadratic regression equation of γ-PGA yield (Y) to peptone (A), dissolved solid in corn saccharifying solution (B), and K2HPO4 (C) is expressed as follows:
Y = − 630.810 + 120.003A + 5.282B + 77.438C + 0.096AB + 3.283AC − 0.625BC − 8.044A2 − 0.045B2 − 31.929C2 (1)
The regression Eq. (1) showed that the quadratic coefficient was negative, which represented the opening of the parabolic surface downward, indicating that the equation had a maximum value. Three-dimensional (3D) response surface plots for γ-PGA yield as a function of protein peptone, corn saccharification liquid, and K2HPO4 are shown in Fig. 2. The results reveal that the highest γ-PGA yield was 48.4605 g/L when peptone was 8 g/L, dissolved solid in corn saccharifying liquid was 60 g/L, and K2HPO4 was 1.00 g/L (Table 4). Three repeated experiments were performed under the optimal conditions obtained from the experiment. The average γ-PGA yield was 48.29 g/L, and the fitting degree with the prediction was 99.6%. Therefore, the model was reliable.
Table 4
Central composite experiment design with three significant variables and their responses.
Runs
|
Real levels (coded levels)
|
γ-PGA production (g/L)
|
protein peptone
(g/L)
|
corn saccharification liquid
(g/L)
|
K2HPO4
(g/L)
|
Actual
|
Predicted
|
1
|
7(-1)
|
50(-1)
|
1.0(0)
|
36.74 ± 0.066
|
36.55
|
2
|
9(+1)
|
50(-1)
|
1.0(0)
|
35.79 ± 0.114
|
35.37
|
3
|
7(-1)
|
70(+1)
|
1.0(0)
|
33.52 ± 0.014
|
34.11
|
4
|
9(+1)
|
70(+1)
|
1.0(0)
|
36.59 ± 0.016
|
36.78
|
5
|
7(-1)
|
60(0)
|
0.5(-1)
|
32.85 ± 0.008
|
32.35
|
6
|
9(+1)
|
60(0)
|
0.5(-1)
|
29.90 ± 0.090
|
29.82
|
7
|
7(-1)
|
60(0)
|
1.5(+1)
|
31.35 ± 0.008
|
31.44
|
8
|
9(+1)
|
60(0)
|
1.5(+1)
|
34.97 ± 0.163
|
35.47
|
9
|
8(0)
|
50(-1)
|
0.5(-1)
|
31.04 ± 0.012
|
31.72
|
10
|
8(0)
|
70(+1)
|
0.5(-1)
|
37.55 ± 0.233
|
37.45
|
11
|
8(0)
|
50(-1)
|
1.5(+1)
|
40.23 ± 0.154
|
40.33
|
12
|
8(0)
|
70(+1)
|
1.5(+1)
|
34.25 ± 0.242
|
33.57
|
13
|
8(0)
|
60(0)
|
1.0(0)
|
48.37 ± 0.147
|
48.29
|
14
|
8(0)
|
60(0)
|
1.0(0)
|
48.05 ± 0.140
|
48.29
|
15
|
8(0)
|
60(0)
|
1.0(0)
|
48.46 ± 0.136
|
48.29
|
Table 5
Analysis of variance (ANOVA) for CCD.
Terma
|
Sum of Squares
|
Degree of
freedom
|
Mean Square
|
F-value
|
P-value
|
Model
|
549.84
|
9
|
61.09
|
131.39
|
< 0.0001**
|
A
|
1.11
|
1
|
1.11
|
2.39
|
0.1825
|
B
|
0.536
|
1
|
0.536
|
1.15
|
0.332
|
C
|
11.21
|
1
|
11.21
|
24.1
|
0.0044**
|
AB
|
3.72
|
1
|
3.72
|
7.99
|
0.0368*
|
AC
|
10.78
|
1
|
10.78
|
23.19
|
0.0048**
|
BC
|
39.02
|
1
|
39.02
|
83.92
|
0.0003**
|
A²
|
238.89
|
1
|
238.89
|
513.77
|
< 0.0001**
|
B²
|
76.29
|
1
|
76.29
|
164.08
|
< 0.0001**
|
C²
|
235.26
|
1
|
235.26
|
505.97
|
< 0.0001**
|
Residual
|
2.32
|
5
|
0.465
|
|
|
Pure error
|
0.0905
|
2
|
0.0453
|
|
|
Lack-of-fit
|
2.23
|
3
|
0.7448
|
16.45
|
0.0578
|
Total
|
552.16
|
14
|
|
|
|
Std.Dev.
|
0.6819
|
R2
|
0.996
|
|
|
Mean
|
37.32
|
Adj R2
|
0.988
|
|
|
C.V.%
|
1.83
|
Pred R2
|
0.935
|
|
|
a A, protein peptone; B, corn saccharification liquid; C, K2HPO4
*Significant at(P < 0.05); **Highly significant at P < 0.01.
3.5 Composition analysis of fermentation broth
The N, P, K, and total nutrient (sum of N, P, and K) contents were 4,100, 3,405, 885, and 8,390 mg/L, respectively; the number of viable B. subtilis was 4.5×107 CFU/mL. We detected 12 volatile chemical components in the fermentation broth (Table 6). At the late fermentation stage, when the carbon source and energy in the medium were depleted, 2,3-butanediol was used [30]. The forms of heavy metals in soil are closely related to soil pH. Acetic acid can affect soil pH, which is the most important factor affecting the solubility and retention of heavy metals in soil; it is also an important factor influencing a series of reactions such as dissolution, precipitation, adsorption, and desorption in soil [31]. Furthermore, acetic acid inhibits spore germination and mycelial growth caused by plant pathogens; it elicits an induction effect on the protective function of plant pathogenic factor infection. It is also an effective solvent of chitosan, a new nontoxic, pollution-free, environmental-friendly, and safe biological pesticide [32]. 2,3-Butanedione, acetic acid, and 2-methyl-butanoic acid can kill nematodes [33]. 2-Methyl-butanoic acid can significantly inhibit the hatching of nematode eggs and effectively promote the growth of Limonium sinense Kuntze [34, 35]. Acetoin, also known as methyl acetyl methanol, acetoin, and 3-hydroxybutanone, and its reducing product 2,3-butanediol can be used as plant growth promoters to trigger and induce systemic resistance, thereby protecting plants against pathogens. As an insect pheromone, acetoin, together with other volatile substances, can be used as an insect attractant. For example, acetoin combined with black rice vinegar can be used as an active component of fly attractant. Thus, a new strategy for environmental protection and pest control using acetoin as a raw material offers the potential to reduce the use of pesticides and labor input [36].
Table 6
Determination of fermentation broth composition by GC-MS.
Number
|
Molecular Formula
|
Compound Name
|
Area %
|
1
|
C25H44N2O5S
|
2-Myristynoyl pantetheine
|
0.76
|
2
|
C4H8O2
|
Acetoin
|
36.06
|
3
|
C2H4O2
|
Acetic acid
|
3.18
|
4
|
C4H10O2
|
[R-(R*,R*)]-2,3-Butanediol
|
39.62
|
5
|
C4H8O2
|
2-methyl-Propanoic acid
|
0.92
|
6
|
C4H10O2
|
[R-(R*,R*)]-2,3-Butanediol
|
4.25
|
7
|
C5H12O2
|
1-methoxy-2-Butanol
|
0.30
|
8
|
C4H8O2
|
Butanoic acid
|
1.08
|
9
|
C5H10O2
|
2-methyl-Butanoic acid
|
0.61
|
10
|
C8H14O4
|
3,4-dihydroxy-3,4-dimethyl-2,5-Hexanedione
|
0.96
|
11
|
C3H8O2
|
1,3-Propanediol
|
0.35
|
12
|
C4H7NO
|
2-Pyrrolidinone
|
0.42
|