Development of Process and Data Centric Inference System for Enhanced Production of L-Asparaginase From Halotolerant Bacillus Licheniformis PPD37

doi:10.21203/rs.3.rs-651645/v1

Download PDF

Research Article

Development of Process and Data Centric Inference System for Enhanced Production of L-Asparaginase From Halotolerant Bacillus Licheniformis PPD37

https://doi.org/10.21203/rs.3.rs-651645/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 30 Nov, 2021

Read the published version in Applied Biochemistry and Biotechnology →

You are reading this latest preprint version

The present study aims at bioengineering of medium components using – data and process centric approaches for enhanced production of L-asparaginase, an important biological molecule, by halotolerant Bacillus licheniformis PPD37 strain. To achieve this, first significant medium components were screened followed by optimization of a combination of media components and culture conditions such as L-asparagine, MgSO₄, NaCl, pH and temperature. Optimization study was carried out using statistical models such as response surface methodology (RSM) – process centric and artificial neural network (ANN) – data centric approaches. The production improved from 2.86 U/mL to 17.089 U/mL, an increase of approximately 6-times of the unoptimized L-asparaginase production. On comparing RSM and ANN models for optimised L-asparaginase production based on R² value, mean absolute percentage error (MAPE), root mean square error (RMSE), and mean absolute deviation (MAD) values, the ANN model emerged as the superior one. As this is the first report to the authors best knowledge on development of inference system using RSM and ANN models for enhanced L-asparaginase production using a halotolerant bacteria, this study could lead to more in-depth and large-scale L-asparaginase production.

Biotechnology and Bioengineering

artificial neural network

process centric approach

halotolerant

L-asparaginase

response surface methodology

Table 1: Plackett-Burman Design (PBD) for screening of medium components

Run No	Na₂HPO₄ (g/L)	KH₂PO₄ (g/L)	NaCl (g/L)	L-asparagine (g/L)	MgSO₄ (g/L)	Glucose (g/L)	CaCl₂(g/L)	Activity (U/mL)	Predicted Activity (U/ml)
1	3	5	30	2	1	0.5	0.05	4.985	4.711
2	8	5	10	8	1	0.5	0.25	10.654	10.446
3	8	1	30	8	0.1	4	0.05	8.727	8.861
4	8	5	30	2	1	4	0.05	4.505	4.773
5	5.5	3	20	5	0.55	2.25	0.15	7.026	7.02
6	3	1	10	2	0.1	0.5	0.05	4.204	4.181
7	3	1	30	8	1	0.5	0.25	8.952	8.563
8	3	5	10	2	0.1	4	0.25	4.001	4.428
9	3	5	30	8	0.1	4	0.25	9.484	9.046
10	8	1	30	2	0.1	0.5	0.25	3.622	3.293
11	8	5	10	8	0.1	0.5	0.05	9.484	9.683
12	3	1	10	8	1	4	0.05	10.841	10.513
13	8	1	10	2	1	4	0.25	5.1	5.006

Table 2: Experimental matrix for central composite design (CCD) using RSM and ANN for cultural parameters with actual and predicted L-asparaginase activity (U/ml)

Run No L-asparagine MgSO₄ NaCl pH Temperature Activity

	Coded	Uncoded	Coded	Uncoded	Coded	Uncoded	Coded	Non-coded	Coded	Non-coded	Experimental	RSM Predicted	ANN Predicted
1	1	2	-1	0.125	1	2	1	7.5	-1	30	12.929	13.604	13.017
2	-1	1	1	0.2	1	2	1	7.5	-1	30	11.235	11.852	11.323
3	0	1.5	0	0.162	0	1.4	0	7	-2.378	25.175	9.652	9.984	10.228
4	0	1.5	-2.378	0.073	0	1.4	0	7	0	33.5	16.473	15.975	16.419
5	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.787	15.662	15.714
6	1	2	-1	0.125	-1	0.8	1	7.5	-1	30	15.422	14.679	15.57
7	1	2	1	0.2	1	2	-1	6.5	1	37	15.31	15.027	15.463
8	1	2	-1	0.125	1	2	1	7.5	1	37	14.862	14.764	14.626
9	-1	1	1	0.2	-1	0.8	1	7.5	-1	30	12.131	11.737	12.317
10	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.759	15.662	15.714
11	-1	1	1	0.2	-1	0.8	-1	6.5	1	37	13.644	13.171	13.336
12	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.857	15.662	15.714
13	1	2	1	0.2	1	2	1	7.5	1	37	15.338	15.072	15.135
14	-1	1	-1	0.125	-1	0.8	-1	6.5	-1	30	11.025	11.836	11.037
15	1	2	-1	0.125	-1	0.8	1	7.5	1	37	15.24	15.961	15.564
16	1	2	1	0.2	-1	0.8	1	7.5	1	37	17.089	16.875	16.946
17	-1	1	1	0.2	1	2	-1	6.5	1	37	12.383	13.024	12.36
18	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.717	15.662	15.714
19	2.378	2.689	0	0.162	0	1.4	0	7	0	33.5	17.073	17.115	17.096
20	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.016	15.662	15.714
21	-2.378	0.310	0	0.162	0	1.4	0	7	0	33.5	10.717	10.788	10.6
22	-1	1	-1	0.125	1	2	-1	6.5	1	37	13.266	13.897	13.236
23	-1	1	1	0.2	1	2	-1	6.5	-1	30	12.929	11.766	12.922
24	-1	1	-1	0.125	1	2	1	7.5	1	37	14.344	13.717	14.918
25	0	1.5	0	0.162	0	1.4	-2.4	5.8	0	33.5	13.308	13.058	13.475
26	0	1.5	0	0.162	2.378	2.827	0	7	0	33.5	13.014	13.409	13.032
27	-1	1	-1	0.125	1	2	1	7.5	-1	30	12.495	12.084	12.41

28	0	1.5	2.378	0.251	0	1.4	0	7	0	33.5	15.675	16.286	15.912
29	0	1.5	0	0.162	0	1.4	0	7	2.378	41.824	13.224	13.005	13.256
30	1	2	1	0.2	-1	0.8	1	7.5	-1	30	15.619	15.813	16.216
31	1	2	-1	0.125	-1	0.8	-1	6.5	1	37	16.403	16.478	16.425
32	1	2	1	0.2	1	2	-1	6.5	-1	30	14.19	14.242	14.276
33	-1	1	1	0.2	-1	0.8	-1	6.5	-1	30	11.515	11.79	11.482
34	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.703	15.662	15.714
35	-1	1	-1	0.125	1	2	-1	6.5	-1	30	12.649	12.417	12.764
36	1	2	1	0.2	-1	0.8	-1	6.5	-1	30	16.011	16.063	15.944
37	1	2	-1	0.125	-1	0.8	-1	6.5	-1	30	15.577	15.349	16.371
38	-1	1	-1	0.125	-1	0.8	-1	6.5	1	37	13.742	13.438	14.009
39	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.352	15.662	15.714
40	0	1.5	0	0.162	-2.378	0.270	0	7	0	33.5	15.731	15.281	14.5
41	1	2	-1	0.125	1	2	-1	6.5	1	37	15.338	15.14	15.229
42	-1	1	-1	0.125	-1	0.8	1	7.5	-1	30	11.375	11.362	11.524
43	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.969	15.662	15.714
44	-1	1	-1	0.125	-1	0.8	1	7.5	1	37	12.495	13.118	12.598
45	0	1.5	0	0.162	0	1.4	0	7	0	33.5	16.067	15.662	15.714
46	1	2	1	0.2	-1	0.8	-1	6.5	1	37	16.529	16.971	16.978
47	-1	1	1	0.2	1	2	1	7.5	1	37	13.35	13.265	13.373
48	0	1.5	0	0.162	0	1.4	0	7	0	33.5	15.156	15.662	15.714
49	1	2	-1	0.125	1	2	-1	6.5	-1	30	13.826	14.134	14.151
50	1	2	1	0.2	1	2	1	7.5	-1	30	14.918	14.133	14.824
51	-1	1	1	0.2	-1	0.8	1	7.5	1	37	13.532	13.272	13.787
52	0	1.5	0	0.162	0	1.4	2.378	8.189	0	33.5	12.229	12.596	12.461

Table 3: Analysis of Variance (ANOVA) for Plackett-Burman Design for screening of medium components

Source	DF	Adj SS	Adj MS	F-Value	p-Value
Model	8	176.182	22.023	51.74	0
Linear	7	176.181	25.169	59.13	0
Na₂HPO₄	1	0.024	0.024	0.06	0.815
KH₂PO₄	1	0.465	0.465	1.09	0.311
NaCl	1	2.68	2.68	6.3	0.023
L-asparagine	1	167.693	167.693	393.98	0
MgSO₄	1	5.078	5.078	11.93	0.003
Glucose	1	0.094	0.094	0.22	0.645
CaCl2	1	0.147	0.147	0.35	0.564
Curvature	1	0.001	0.001	0	0.963
Error	17	7.236	0.426
Lack-of-Fit	4	2.463	0.616	1.68	0.215
Pure Error	13	4.773	0.367
Total	25	183.418

S		R-sq	R-sq(adj)	R-sq(pred)
0.65241		96.05%	94.20%	91.09%

Table 4 Analysis of Variance (ANOVA) of RSM-CCD for L-Asparaginase production

Source	DF	Adj SS	Adj MS	F-Value	P-Value
Model	20	158.443	7.9222	24.44	0.000
Linear	5	98.542	19.7084	60.81	0.000
Square	5	49.694	9.9388	30.67	0.000
2-Way Interaction	10	9.436	0.9436	2.91	0.011
Error	31	10.047	0.3241
Lack-of-Fit	22	8.957	0.4071	3.36	0.032
Pure Error	9	1.090	0.1211
Total	51	168.490
S	R-sq	R-sq(adj)		R-sq(pred)
0.569291	94.04%	90.19%		80.01%

Table 5: Estimated regression coefficient of L-asparaginase activity by RSM-CCD

Term	Coef	SE Coef	T-Value	p-Value
Constant	15.662	0.179	87.61	0
L-Asparagine - A	1.33	0.0865	15.38	0
MgSO₄ - B	0.0655	0.0865	0.76	0.455
NaCl – C	-0.3055	0.0897	-3.4	0.002
pH – D	-0.1074	0.0864	-1.24	0.223
Temperature - E	0.6352	0.0865	7.34	0
A²	-0.3023	0.0743	-4.07	0
B²	0.0829	0.0743	1.12	0.273
C²	-0.2697	0.0873	-3.09	0.004
D²	-0.4967	0.0737	-6.74	0
E²	-0.7367	0.0743	-9.92	0
AB	0.19	0.101	1.89	0.068
AC	-0.449	0.101	-4.46	0
AD	-0.049	0.101	-0.49	0.629
AE	-0.118	0.101	-1.18	0.249
BC	-0.151	0.101	-1.51	0.142
BD	0.105	0.101	1.04	0.304
BE	-0.055	0.101	-0.55	0.588
CD	0.035	0.101	0.35	0.73
CE	-0.031	0.101	-0.3	0.763
DE	0.039	0.101	0.38	0.704

Table 6: Comparison of predictive capability between RSM and ANN

Predictive Ability Measurement	RSM	ANN
R2 Value	0.9404	0.9846
Mean Absolute Percentage Error (MAPE)	2.632	1.519
Root Mean Square Error (RMSE)	0.439	0.322
Mean Absolute Deviation (MAD)	1.568	1.558