Regular articleAnalyses of extracellular protein production in Bacillus subtilis – I: Genome-scale metabolic model reconstruction based on updated gene-enzyme-reaction data
Introduction
Bacillus subtilis and its relatives are prodigious producers of industrial enzymes, such as proteases and amylases. Thus, B. subtilis has been the model of Firmiculates for decades, since the absence of an outer membrane combined with an efficient Sec-dependent secretion pathway [1] means that proteins can be secreted directly into the culture medium at high concentrations [2]. In recent years, considerable effort has been aimed at developing B. subtilis as host for the production of heterologous proteins. Meanwhile, the studies on: (i) promoters and suitable ribosome-binding sites [3], [4], [5], (ii) signal peptides [6], [7], (iii) secretion pathway [5], [8], [9], [10], (iv) proteolytic degradation of proteins [11], and (v) genome reductions for host generation [12], [13] underpin the crucial aim to increase recombinant protein (r-protein) production in B. subtilis and stimulate more research directed to metabolic and bioreactor engineering. Moreover, being sequenced and updated totally, B. subtilis [14], [15], [16] has also been shown to be an excellent model organism for systems biological analyses [2].
Intracellular reaction fluxes can be calculated by solving flux-balance based stoichiometric models based on intracellular reaction networks together with elaborate fermentation data [17]. Therefore, based on reliable models calculation of reaction fluxes is an important tool in metabolic engineering, allowing detailed quantification of in vivo fluxes which provide valuable information on cellular physiology that can be applied for engineering metabolic and regulatory pathways, dignity of which is dependent on the quantity and reliability of intracellular reactions.
Genome sequencing and subsequently annotation studies enable construction of genome scale reaction networks with high number of reactions interrelated with gene- and enzyme- data. B. subtilis genome was sequenced first by Kunst et al. [14] with 4100 protein-coding genes, and the annotated genome sequence stimulated the studies for B. subtilis network constructions [18], [19], [20], which were converted into in silico genome scale models (GEMs). Consequently, the first three GEMs reported by Oh et al. [18], Goelzer et al. [19], and Henry et al. [20] which can be regarded as the first-generation B. subtilis GEMs were based on the genome annotation of Kunst et al. [14]; thereafter, the second-generation GEMs of Tanaka et al. [21] and Hao et al. [22] were reconstructed after the genome annotation of Barbe et al. [15] (Table 1a). In order to evaluate the B. subtilis GEMs, type of data used in the construction [18] and reconstructions [19], [20], [21], [22] and in validation processes of the first- and second- generation B. subtilis GEMs are summarized in Table 1b. Furthermore, the strategies applied for optimum flux distributions in model constructions and their validation are summarized in Table 1c. Also, the accuracy of the GEMs are summarized in Table 1d.
Genome-scale reaction networks need to involve all accurate reactions based on correct gene-enzyme-reaction annotations. Therefore, resequencing of a genome stimulates reconstruction of genome scale model(s). In this context, the third re-sequencing of B. subtilis genome in 2013 by Belda et al. [16] and the subsequent re-annotation studies renewed the gene-enzyme-reaction data and created an updated data platform for the field of biochemical reaction engineering to reconstruct a renewed and expanded genome scale B. subtilis reaction network. The flow of updated gene-enzyme-reaction data enabled the authors to reconstruct the updated third-generation B. subtilis reaction network, BsRN-2016. Through thermodynamic analyses and elimination of unconnected-reactions, BsRN-2016 was converted into the first third-generation GEM, so-called iBsu1144. The dignity and robustness of iBsu1144 was tested in predicting intracellular reaction fluxes in a protein fermentation process, i.e., serine alkaline protease (SAP), using elaborate experimental data used in the first flux analysis work for a protein fermentation [29], [30]. The model presents a step towards more complete description of protein production in B. subtilis which is required to understand and optimize fermentation processes. Insights obtained from the intracellular flux distributions at well-defined low-, medium-, and high-oxygen transfer conditions throughout the bioprocess during batch-fermentations were used to determine metabolic engineering sites for the protein synthesis and for host cell constructions for protein productions. The reconstructed third generation GEM iBsu1144 was compared with a second- generation GEM [21], and in the light of information gathered from the GEM iBsu1144, future perspectives were discussed.
Section snippets
Mass flux balance-based analysis
Stoichiometric analysis of metabolic flux distributions [31] with flux balance (FB) methodology aims calculation of flow of metabolites through the mathematically represented intracellular reaction network without kinetic parameters. As it does not use kinetic parameters, it cannot predict metabolite concentrations, and it is only suitable for the calculation of fluxes at a mathematically well defined state. When fermentation data are used as constraints, an allowable solution space is
Reconstruction of B. subtilis reaction network and the genome-scale model
The third-generation B. subtilis reaction network, and then the GEM were constructed manually with a high level of accuracy and consistency through five interconnected steps (Fig. 2):
1. Based on the recent resequencing work [16] the updated annotations were collected from the databases; MicroCyc [36], SEED, GenoList, MetaCyc, UniProt, SubtiList, KEGG, ExPASy, BSORF, BRENDA, and the literature. The reactions were evaluated and B. subtilis reaction network BsRN-2016 containing 1144 genes, 1955
Conclusions
We presented B. subtilis reaction network BsRN-2016 and the first third-generation genome scale model (GEM) iBsu1144 with an updated gene-enzyme-reaction data for B. subtilis based on the third-sequencing of its genome [16] in 2013 and consequent re-annotation studies, following the first-generation GEMs [18], [19], [20] based on the sequencing of Kunst et al. [14] and second-generation GEMs [21], [22] based on the re-sequencing Barbe et al. [15], within a decade. iBsu1144 was used for an
Acknowledgements
This work was supported by the Scientific and Technological Research Council of Turkey (TÜBİTAK) through the project 109R025. PK was awarded a PhD scholarship (BIDEB: 2210) by the TÜBİTAK.
References (41)
- et al.
Bacillus protein secretion: an unfolding story
Trends Microbiol.
(2008) - et al.
Systematic screening of all signal peptides from Bacillus subtilis: a powerful strategy in optimizing heterologous protein secretion in gram-positive bacteria
J. Mol. Biol.
(2006) - et al.
Genome-scale reconstruction of metabolic network in Bacillus subtilis based on high-throughput phenotyping and gene essentiality data
J. Biol. Chem.
(2007) - et al.
Oxygen transfer effects in serine alkaline protease fermentation by Bacillus licheniformis: Use of citric acid as the carbon source
Enzyme Microb. Technol.
(1998) - et al.
Heading in the right direction: thermodynamics based network analysis and pathway engineering
Curr. Opin. Biotechnol.
(2015) - et al.
Mass flux balance-based model and metabolic pathway engineering analysis for serine alkaline protease synthesis by Bacillus licheniformis
Enzyme Microb. Technol.
(1999) - et al.
Bacillus subtilis: from soil bacterium to super secreting cell factory
Microb. Cell Fact.
(2013) - et al.
The influence of ribosome-binding-site elements on translational efficiency in Bacillus subtilis and Escherichia coli in vivo
Mol. Microbiol.
(1992) The sigma factors of Bacillus subtilis
Microbiol. Rev.
(1995)- et al.
Improving protein production on the level of regulation of both expression and secretion pathways in Bacillus subtilis
J. Microbiol. Biotechnol.
(2015)
Proteomics of protein secretion by Bacillus subtilis: separating the secrets of the secretome
Microbiol. Mol. Biol. Rev.
Protein traffic for secretion and related machinery of Bacillus subtilis
Biosci. Biotechnol. Biochem.
Protein secretion pathways in Bacillus subtilis: implication for optimization of heterologous protein secretion
Biotechnol. Adv.
Molecular engineering of secretory machinery components for high-level secretion of proteins in Bacillus species
J. Ind. Microbiol. Biotechnol.
Extracytoplasmic proteases determining the cleavage and release of secreted proteins lipoproteins, and membrane proteins in Bacillus subtilis
J. Proteom Res.
Enhanced recombinant protein productivity by genome reduction in Bacillus subtilis
DNA Res.
Combined effect of improved cell yield and increased specific productivity enhances recombinant enzyme production in genome-reduced Bacillus subtilis strain MGB874
Appl. Environ. Microbiol.
The complete genome sequence of the gram positive bacterium Bacillus subtilis
Nature
From a consortium sequence to a unified sequence: the Bacillus subtilis 168 reference genome a decade later
Microbiology
An updated metabolic view of the Bacillus subtilis 168 genome
Microbiology
Cited by (23)
Improving microbial bioproduction under low-oxygen conditions
2023, Current Opinion in BiotechnologyMachine learning for metabolic pathway optimization: A review
2023, Computational and Structural Biotechnology JournalChemical pretreatment and saccharification of corncob for poly-γ-glutamic acid production by Bacillus subtilis SCP010-1
2022, Process Safety and Environmental ProtectionCitation Excerpt :The simultaneous hydrolysis of cellulose and its fermentation is called simultaneous saccharification and fermentation (SSF), while separate hydrolysis and fermentation (SHF) involves separate processes for enzymatic saccharification and fermentation (Ask et al., 2012). Although B. subtilis SCP010–1 contained gene sequences for different types of extracellular enzymes, including cellulase, amylase, xylanase, mannase and esterase amongst others (Seo et al., 2010; Pınar et al., 2017), its ability to produceγ-PGA by SSF fermentation was rather low. Therefore, the SHF method whereby enzymatic saccharification is performed separately from fermentation to produce γ-PGA was preferred.
Analysis and modeling tools of metabolic flux
2022, Current Developments in Biotechnology and Bioengineering: Designer Microbial Cell Factories: Metabolic Engineering and ApplicationsNaturally occurring novel promoters around pyruvate branch-point for recombinant protein production in Pichia pastoris (Komagataella phaffii): Pyruvate decarboxylase- and pyruvate kinase- promoters
2018, Biochemical Engineering JournalCitation Excerpt :The stoichiometric reactions and abbreviations of the metabolites are presented in Table S2. The fluxes were calculated using GAMS 2.25 (GAMS Development Corp., Washington DC, USA), by minimizing the objective function defined as the difference between synthesis- and secretion- rates of rhGH [32,37,38]. The variables of the model were the fluxes of metabolic reactions expressed in mmol gDCW−1 h-1; flux towards the cell was represented with the specific growth rate, μ (h−1).
Analyses of extracellular protein production in Bacillus subtilis – II: Responses of reaction network to oxygen transfer at transcriptional level
2017, Biochemical Engineering JournalCitation Excerpt :Period I (0 < t < 8 h) is the exponential cell growth phase; Period II (8 < t < 14 h) is the growth interruption phase starts by the addition of protease inhibitor where rhGH secretion starts; Period III (14 < t < 20 h) is the rhGH secretion phase where the cell growth proceeds in the early stationary phase, and Period IV (20 < t < 32 h) is the late stationary phase where rhGH secretion was the highest. The reaction fluxes at LOT, MOT, and HOT conditions in each period at t = 0, 12, 16, and 24 h were calculated as a single optimal flux distribution on the edge of the solution space formed [19] by the rhGH fermentation data as constraints; however, only the results in Period III at t = 16 h are presented in Supplementary Tables 3.1–3.15. Microarray analyses were conducted with samples isolated from bioreactors operated at LOT, MOT, and HOT conditions at t = 0, 12, and 16 h, which correspond to respectively, Period I, Period II, and Period III.