Deionococcus proteotlycius Genomic Library Exploration Enhances Oxidative Stress Resistance and Poly-3-hydroxybutyrate Production in Recombinant Escherichia coli

Abstract

Cell growth is inhibited by abiotic stresses during industrial processes, which is a limitation of microbial cell factories. Microbes with robust phenotypes are critical for its maximizing the yield of the target products in industrial biotechnology. Currently, there are several reports on the enhanced production of industrial metabolite through the introduction of Deinococcal genes into host cells, which confers cellular robustness. Deinococcus is known for its unique genetic function thriving in extreme environments such as radiation, UV, and oxidants. In this study, we established that Deinococcus proteolyticus showed greater resistance to oxidation and UV-C than commonly used D. radiodurans. By screening the genomic library of D. proteolyticus, we isolated a gene (deipr_0871) encoding a response regulator, which not only enhanced oxidative stress, but also promoted the growth of the recombinant E. coli strain. The transcription analysis indicated that the heterologous expression of deipr_0871 upregulated oxidative-stress-related genes such as ahpC and sodA, and acetyl-CoA-accumulation-associated genes via soxS regulon. Deipr_0871 was applied to improve the production of the valuable metabolite, poly-3-hydroxybutyrate (PHB), in the synthetic E. coli strain, which lead to the remarkably higher PHB than the control strain. Therefore, the stress tolerance gene from D. proteolyticus should be used in the modification of E. coli for the production of PHB and other biomaterials

1. Introduction

Escherichia coli is one of the most common industrial hosts used as microbial cell factories for producing pharmaceuticals [1], biopharmaceuticals [2], fine chemicals [3], recombinant proteins, and biofuels [4,5]. To maximize the yield of bacterial metabolites, E. coli is subjected to high-cell-density fermentation [6] under aerobic conditions [7]. However, the high oxygen concentration and simultaneous accumulation of products such as fuels and chemicals can lead to diverse stresses, especially oxidative stress, which resulted in reduced industrial production [8,9]. Oxidative stress is induced by reactive oxygen species (ROS), such as hydroxyl radical (OH^·), hydrogen peroxide (H₂O₂), and superoxide anion (O₂⁻), through biological respiration including metabolic pathway destruction and DNA alteration [10]. Therefore, oxidative stress resistance is a critical factor in commercial strains, and various metabolic engineering techniques have been applied to enhance their robustness against oxidative stress [11].

Engineering regulatory factors, cell membranes, and biosynthetic pathways were employed to improve stress tolerance in E. coli [12], for example, the heterologous expression of the putative response regulator DRH632 from Antarctic bacteria into E. coli [13], the random mutation of cAMP receptor protein [11], and the overexpression of iron exporters, FetA and FetB, into E. coli [14]. As Deinococcus sp. is known for its ability to survive against multiple stresses such as γ- radiation, UV-C, oxidative stress, and DNA damage reagents, several studies have investigated its genes [15]. Heterologous expression of Deinococcus genes in E. coli successfully leads to strains with multiple stress resistance. For example, the cold-shock-domain-containing protein PprM [16], pyrroloquinoline-quinone (PQQ) synthase [17], manganese (Mn) transporter protein (MntH), and a small heat shock protein (Hsp20) confer oxidative stress tolerance to E. coli [18,19]. Moreover, dr_1558, a response regulator, confers on E. coli multiple stress resistances to oxidative, acidic, salt, and heat stresses [20]. Interestingly, some reports suggest that introducing Deinococcal genes into other hosts promotes not only oxidative stress resistance but also its metabolic activity. For instance, the Deinococcal gene irrE (pprI) improves the growth and ethanol production of Zymomonas mobilis [21], the proliferation of E. coli [18], and lactic acid production in Lactococcus lactis [19]. In addition, dr_1558 enhances the production of succinic acid [22], γ-aminobutyric acid (GABA) in E. coli [23], and cadaverine in Corynebacterium glutamicum [24]. Moreover, the biosynthesis of poly(3-hydroxybutyrate) (PHB), a biopolymer that can replace petroleum-based plastics [25], was highly produced in synthetic E. coli strains under dr_1558 regulation.

To date, 67 Deinococcus-type strains have been isolated and sequenced. However, most studies have focused on the D. radiodurans genes. Recently, Lim et al. reported that complicated stress resistance systems were widely located in the genome of Deinococcus as conserved and divergent forms [26], which are not fully understood yet.

In this study, to confer oxidative stress resistance and cellular robustness on E. coli, we explored the genomic library of Deinococcus proteolyticus. Through a series of screening steps, we selected a gene that has the potential to resist a high concentration of hydrogen peroxide. The selected gene was introduced subsequently to the PHB-generative E. coli strain to investigate its impact on PHB synthesis and metabolism.

2. Materials and Methods

2.1. Bacterial Strains and Plasmids and Culture Conditions

Deinococcus proteolyticus MRP (taxid:693977) was used as the genomic DNA donor strain and was obtained from the Korean Agricultural Culture Collection (KACC) of the Rural Development Administration. The bacterial strains Escherichia coli XL1-Blue (E. coli recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac F’ proAB lacI^qZΔM15 Tn10 [Tet^r]) was employed for the construction of a library, evaluation of survival rate under oxidative stress conditions, and fermentation for producing PHB. Plasmid pKMCAB (pKM212-MCS + pCnCAB) for PHB biosynthesis constructed previously [25] was introduced into XL1-Blue.

2.2. Screening of Oxidative-Resistant Clones Using the Genomic DNA Library

To construct the gDNA library of D. proteolyticus MRP, total DNA (1136 ng/µL) was extracted and smeared into 3–5 kb fragments. The fragments were then ligated into a modified pC31 vector provided by Macrogen (Seoul, Republic of Korea). The libraries for screening were cultured for 18 h at 37 °C with Bacto Luria-Bertani (LB) broth containing 50 µg/mL kanamycin and different concentrations of hydrogen peroxide (H₂O₂). The oxidative tolerance survival study was performed using exponential-phase cultures. Subsequently, another survival study to confirm deipr_0871 function isolated from the screening was executed as described previously using the E. coli strains harboring pRad-0871 and pRadGro. The surviving fraction was calculated by dividing the number of colonies in the samples by the number of colonies in the controls.

2.3. Construction of Deipr_0871 Expression Vector

The plasmid pRadGro was used as a vector for gene cloning and protein expression of the oxidative-stress-resistance gene, deipr_0871. The gene was amplified from the selected clones using primer Deipr_0871_F: 5-GAGGGCCCATGAACCGTTCCCAAGCTTCTCTTG-3′ and Deipr_0871_R: 5′-GATTAAGCTTCTAGTCCAGCAGACCGATGGTG-3′. A polymerase chain reaction was performed using nPfu-Forte DNA polymerase (Enzynomics, Daejeon, Republic of Korea) under the following conditions: initial denaturation at 95 °C for 5 min, followed by 30 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 45 s; followed by a final extension step at 72 °C for 1 min 30 s.

The amplified fragments were digested with Apa I and Hind III, then ligated to pRadGro vector treated with the same restriction enzymes. The resulting plasmid was transformed into E. coli, XL1-Blue, to create the pRad-0871 strain. The E. coli strain harboring the empty vector pRadGro was used as a control strain in this study.

2.4. Determination of Growth and Survival of the Recombinant E. coli Strains

Cell growth was analyzed with the optical density (OD₆₀₀) using a UV spectrophotometer (GENESYS 150 UV-visible spectrophotometer; Thermo Scientific, Waltham, MA, USA). All stress survival studies were performed using cells in the exponential phase after adjustment to approximately 10⁷ CFU/mL (OD₆₀₀ ≈ 0.1). For the γ-radiation tolerance assay, Deinococcus strains were irradiated to different doses (3~15 kGy) of γ-radiation using a ⁶⁰Co-gamma irradiator (Advanced Radiation Technology Institute, Jeongeup, Republic of Korea), then serially diluted in TGY media (0.5% Tryptone, 0.3% Yeast extract and 0.1% glucose), and spotted on TGY agar. To investigate the UV-C tolerance of D. radiodurans and D. proteolyticus, they were irradiated with 800 J/m² UV-C using a UV-C crosslinker (CX-2000 UV Crosslinker, UVP, Milwaukee, CA, USA) after successive dilution and dropping on a TGY plate. For oxidative stress survival, the Deinococcus strains were serially diluted in TGY broth and spotted onto TGY containing 0.4 mM of hydrogen peroxide (H₂O₂), then incubated at 30 °C for 36 h. The recombinant E. coli strains introducing pRadgro and pRad-0871 were also tested with the same process at 37 °C for 18 h as described above using LB and a 0.2 mM H₂O₂ LB plate. The surviving fraction was calculated by dividing the number of colonies in each sample by the number of cells in the control group.

2.5. Real-Time PCR (qPCR)

All sampling for RNA extraction was performed in the exponential phase of the E. coli strains with and without deipr_0871 under non-stressed and stressed conditions (0.45 mM H₂O₂), and without and with a high concentration of glucose (30 g/L). Total RNA was extracted using the RiboEx reagent (GeneAll, Seoul, Republic of Korea) supplied with DNase, and purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. RNA quality and concentration were estimated by measuring the optical densities of the solutions at 260 and 280 nm using a GeneQuant Pro instrument (Amersham Pharmacia Biotech, Amersham, UK). RNA (1 µg) from each sample was used for cDNA synthesis with random hexamers using a PrimeScript first-strand cDNA synthesis kit (TaKaRa Bio Inc., Kusatsu, Japan). RT-qPCR amplification was performed using SYBR Premix Ex Taq (TaKaRa) and an Eco Real-Time PCR system (Illumina Inc., San Diego, CA, USA) according to the manufacturer’s instructions. The housekeeping gene polA was used as an internal control. Primers used for RT-qPCR are listed in Table S1. All reactions were repeated in duplicates, and the relative expression was calculated using the relative quantification method.

2.6. Genome Analysis

The genome sequences of D. radiodurans R1 (GCF_020546685.1) and D. proteolyticus MRP (GCF_020546685.1) were obtained from the NCBI database, respectively. The two genome assemblies were examined to confirm if those qualities exceeded the proposed cutoff (over 95% completeness and under 5% contamination) using the CheckM software v1.1.6 [27]. The amino acid sequences and corresponding annotated gff files were obtained by executing the Prokka (v. 1.14.6) annotation pipeline, with the default parameters [28]. Then, the coded in-house pipeline was executed for obtaining the percentage of core and accessory COG category ratio, where DIAMOND sequence alignment (50% percent identity) and COG database were implemented [29,30]. From the annotated protein information, a donut plot depicting the positions of core and accessory genes of those two genomes was generated by using in-house code.

To predict the function of genes, the conserved domains (CDD) such as Pfam and the COG group were investigated using the InterPro database (https://www.ebi.ac.uk/interpro/, accessed on 3 April 2023).

2.7. Cultivation for PHB Production

To cultivate E. coli strains carrying the vector pKMCAB containing the PHB generative operon, seed cultures were cultivated at 30 °C in 5 mL LB medium for 18 h with shaking at 200 rpm. For the main shake flask cultures, 3% (v/v) seed cultures were added to 30 mL of LB medium (in 250 mL shake flask) supplemented with 30 g/L glucose. The cells were grown at 30 °C for 60 h at 200 rpm. Kanamycin (50 µg/mL) and ampicillin (100 µg/mL) were added to the culture medium to maintain plasmid stability. To measure glucose and PHB concentrations, growth sampling was performed at intervals of 12 h before 24 h cultures and at intervals of 6 h after 24 h.

2.8. Measurement of Amount of PHB

To determine the PHB concentration and glucose consumption, the cells were harvested via centrifugation at 10,000× g for 15 min. Cell pellets were dried in an 80 °C dry oven for 24 h, and the dry cell weight (DCW) was determined. The PHB concentration was analyzed using gas chromatography (SHIMADZU GC-2010 Plus, Shimadzu Inc., Kyoto, Japan) equipped with a flame ionization detector (FID) after solvent extraction using 3% (v/v) H₂SO₄ in methanol [31]. To determine the concentration of glucose consumption, cell supernatants were analyzed using high-performance liquid chromatography (HPLC; Agilent Technologies Inc., Santa Clara, CA, USA) equipped with a refractive index detector (RID).

2.9. Sequence Alignment

The amino acid sequences of the NarL/FixJ family (Deipr_0871, DR_1558: AAF11120, DRH577: OF298141, DRH632: LY802646, DRH1601, E. coli NarL: CAA48935, E. coli UhpA: YP002389147, E. coli NarP: CAD6002661, E. coli UvrY:CAD6011216, E. coli RcsB:CAD6002904, E. coli EvgA:CAD6007853, Bordetella pertussis BvgA:NP880570, Chromohalobacter salexigens EupR:ABE58223, Sinorhizobium meliloti FixJ:CAA79898) used in this study were obtained from NCBI (http://ncbi.nim.nih.gov/, accessed on 8 May 2023). Multiple sequence alignment was performed using Clustal Omega [32]. The conserved regions were visualized using GeneDoc 2.7 (Free Software Foundation, Inc., Boston, MA, USA). A phylogenetic tree was constructed using MEGA 11 based on the neighbor-joining method and evaluated using a bootstrap test with 1000 replicates [33].

3. Results

3.1. The Oxidative Stress Resistance Properties of Deinococcus Proteolyticus

D. proteolyticus, isolated from the feces of Lama glama in 1973, produces orange pigments and exhibits γ-radiation resistance similar to that of D. radiodurans [34]. To date, only the genome of D. proteolyticus has been sequenced and its stress resistance properties have not been studied [34]. To investigate D. proteolyticus’s resistance potential under various environmental conditions, the oxidative stress, UV-C stress, and γ-radiation tolerance of D. proteolyticus were examined and compared to those of D. radiodruans. It showed that D. proteolyticus exhibited as strong an endurance to radiation stress as D. radiodurans (Figure 1A). We also found that D. proteolyticus displays distinct resistance to UV-C and hydrogen peroxide. As shown in Figure 1, D. proteolyticus displayed approximately a 10-fold increase in UV-C resistance (800 J/m²) compared to D. radiodurans. In the presence of 0.4 mM hydrogen peroxide, D. proteolyticus showed minimal cell loss, whereas D. radiodurans exhibited a 4-log cycle decrease in viability (Figure 1B).

To explain the distinct stress resistance properties of D. proteolyticus, we compared the genomes of D. proteolyticus and D. radiodurans. The genome of D. proteolyticus was reported in 2012 and consists of one chromosome (CP002536.1) and four plasmids with a GC content of 65.6%. It contains a total of 2790 genes, of which 2688 are protein-coding. ANI analysis showed that D. proteolyticus had a 73.25% similarity to D. radiodurans. A comparative analysis showed that 1489 genes were homologous (core genome) between D. proteolyticus and D. radiodurans, while 1185 were D. protelyticus-specific (>50%) (Figure 2). Most of the core genome was involved in translation and energy-production-associated COG groups (Figure 2C), while high diversity was observed in the amino acid metabolism and inorganic ion metabolism COG groups. Most of the known stress-resistance-associated genes, including ddrO and pprI in D. radiodurans, were conserved in D. proteolyticus [26].

3.2. Screening Genomic Libraries of D. proteolyticus for Genes Improving Oxidative Stress Resistance to E. coli

To identify the oxidative-stress-associated genes in D. proteolyticus, we constructed a genomic library using 3~5 kb gene fragments. The isolation step was implemented under varying hydrogen peroxide concentrations (1.0–1.5 mM in 1st step and 1.5–2.0 mM in 2nd step). In the first step, 446 clones showed faster growth than the wild-type (WT) strain. In the second step, 50 clones exhibited stress resistance to hydrogen peroxides (Supplementary Figure S1). The candidate clones were treated with 20 mM hydrogen peroxide for 1 h, of which two exhibited much higher tolerance phenotypes (Supplementary Figure S2). The sequencing analysis showed that one clone contained the catalase gene (deipr_2034), while the other clone contained two genes, namely, histidine kinase (deipr_0870) and a response regulator (deipr_0871).

Catalase encoded by deipr_2034 shares 86% homology with catalase (DR_1998) in D. radiodurans which is known as the main H₂O₂-scavenging enzyme in D. radiodurans. The amino acid sequence of deipr_0871 showed high similarity (78%) to that of dr_0891 in D. radiodurans. However, its paired protein, DR_0892, showed only 58% similarity with Deipr_0870 (Figure 3A). Moreover, the neighboring genes of this cluster exhibited properties distinct from those of D. radiodurans (Supplementary Table S2).

To further characterize deipr_0871, we analyzed its domain composition using the InterPro database (https://www.ebi.ac.uk/interpro/, accessed on 8 May 2023). The response regulator Deipr_0871 included an N-terminal receiver domain and a LuxR C-like DNA-binding helix-turn-helix (HTH) domain (Figure 3C). CDD domain analysis indicated that Deipr_0871 is a response regulator of the NarL-like family. A phylogenetic tree with the well-known NarL/FixJ family of response regulators indicated that Deipr_0871 was independently located with C. salexigens EupR and closed with E. coli NarP and NarL response regulators (Figure 3B). Especially, the sequence of Deipr_0871 was so far from that of DR_1558 which was also classified into NarL-like family response regulator in D. radiodurans. The alignment analysis showed that Deipr_0871 has a relatively short linker region between the receiver domain and the HTH domain. Although the phosphorylation site (D91) was highly conserved, the putative DNA binding sites of Deipr_0871 differed slightly (Figure 3C).

3.3. Effect of Deipr_0871 on the Oxidative Stress Resistance of E. coli Strain

To investigate the effect of deipr_0871 on the oxidative stress resistance in E. coli, we constructed an auto-inducible expression vector, pRad-0871, and transformed it into the XL1-blue strain. The stress resistance potential of E. coli in the presence of hydrogen peroxide was examined using an LB medium containing 0.2 mM hydrogen peroxide. Under exposure to hydrogen peroxide, the cell count of E. coli in the control group was shown to be 10-fold lower than E. coli with deipr_0871 (Figure 4A).

We then assessed the expression of genes associated with oxidative stress in E. coli, including katG and katE (catalases), dps (DNA-binding protein from starved cells), ahpC (subunit C of alkyl hydroperoxide reductase), and sodA (superoxide dismutase) [35], using real-time PCR (qPCR) (Figure 4A). The expression of sodA was upregulated (2-fold increase) in the E. coli strain harboring deipr_0871 than that of the control strain, whereas the mRNA levels of katG, katE, and dps were not increased in the presence of deipr_0871. The main stress response sigma factor rpoS was unaffected by deipr_0871. Additionally, we examined the expression of two crucial E. coli genes, oxyR, and soxS, that encode regulators of the oxidative stress response [36]. The transcript levels of oxyR were comparable in both the control and deipr_0871 expression strains. However, soxS transcript levels increased 2.5-fold in the presence of deipr_0871 compared to that of the control strain (Figure 4B).

3.4. Effect of Deipr_0871 on Growth of E. coli Strain

We found that deipr_0871 affects not only oxidative stress resistance but also the growth rate in E. coli. Although the optical cell density of E. coli harboring deipr_0871 was not different from that of the control, the growth of the E. coli strain harboring deipr_0871 in the exponential phase was 1.4 times faster than that of the control strain (Figure 5A).

To examine changes in the metabolic ability of E. coli strain with deipr_0871, transcripts of 11 genes related to the central carbon pathway in E. coli were measured using qPCR (Figure 5B). Most genes were not affected by deipr_0871; icdA, pps, and ppc expression was downregulated 2-fold, 5-fold, and 2-fold, respectively. pck and mdh transcripts, encoding sequential enzymes that generate PEP from malate through oxaloacetate, increased by 4.5-fold and 2.3-fold, respectively. Moreover, aceE (pyruvate dehydrogenase E1) converting pyruvate to acetyl-CoA was upregulated 3-fold in the E. coli strain with deipr_0871 compared to that in the control, which showed acetyl-CoA accumulation.

Interestingly, the addition of glucose to the medium significantly promoted growth (Figure 5C). The optical density at 600 nm (OD_600nm) was 16.12 ± 0.34 in deipr_0871-harboring strains, which was 2-fold higher than that of the control strain (7.96 ± 0.58). The transcripts of zwf, icdA, and sucA, which contribute to the energy cycle in the pentose phosphate pathway and tricarboxylic acid (TCA) cycle, were upregulated 2.3-fold, 2.4-fold, and 3-fold, respectively. Especially, the transcript of aceE was highly enhanced by 20-fold in the presence of glucose (Figure 5D).

3.5. Effect of Deipr_0871 on Valuable Metabolite, PHB, Production in the E. coli Strain

To use these properties to produce the valuable metabolite PHB, we introduced PHB synthesis genes phaA, phaB, and phaC, into the E. coli strain using the pKMCAB vector. Cell growth and glucose consumption rates were significantly higher in deipr_0871-overexpressing strains than in controls. After 60 h of incubation with E. coli, up to 80% of the glucose was consumed in both strains.

Table 1. Effects of deipr_0871 on cell concentration and PHB production (60 h).

Strains	DCW (g/L)	Yield (gPHB/gGlucose)	PHB Conc. (g/L)	PHB Content (wt%)
E. coli XL1-Blue (pKMCAB+pRad-0871)	3.88 ± 0.09	0.12 ± 0.004	3.139 ± 0.07	80 ± 3
E. coli XL1-Blue (pKMCAB+pRadGro)	1.96 ± 0.05	0.03 ± 0.007	0.731 ± 0.03	38 ± 1

reveals that the DCW of E. coli carrying pKMCAB and pRad-0871 reached 3.88 ± 0.09 g/L by consuming 26.06 ± 0.81 g/L glucose, whereas that of the control strain only reached 1.96 ± 0.05 g/L with a glucose consumption of 27.62 ± 1.79 g/L. In addition, PHB production was increased from 0.731 ± 0.03 g/L in control E. coli to 3.139 ± 0.07 g/L in E. coli with pKMCAB and pRad-0871 (Table 1).

4. Discussion

In this study, we found that D. proteolyticus exhibited high resistance potential toward oxidative stress. In the core-genome analysis between D. proteolyticus and D. radiodurans, important genes for stress resistance were observed in both strains with high similarity, which was also observed in the comparative analysis of 11 genomes in the Deinococcus species [26].

TCSs facilitate the detection of environmental signals and regulate diverse stress-resistance-associated genes [37]. Among the stress response genes in D. proteolyticus, we revealed that D. proteolyticus has 11 two-component systems, 6 orphan histidine kinases, and 6 orphan response regulators. A comparative analysis with the genome of D. radiodurans showed that most of the response regulators (73%) have a high degree of similarity (>50%) with response regulators from D. radiodurans (Supplementary Figure S3). It was proposed that highly preserved response regulators may play a pivotal role in the stress response.

From a genomic library constructed using the gDNA of D. proteolyticus, we isolated a gene (Deipr_0871) that enhanced the oxidative stress resistance of E. coli strains. The selected clone contained a response regulator (RR) domain involved in a two-component signal transduction system (TCSs) with histidine kinase (HK) [37]. Deipr_0871 was composed of two domains, receiver domain and NarL-like LuxR type-HTH domain. The NarL/FixJ family has divergent roles in bacterial systems such as nitrogen fixation and sugar phosphate transport [38].

A phylogenetic tree with other known NarL family regulators revealed that Deipr_0871 was found in close proximity to C. salexigens EupR, E. coli NarP, and NarL response regulators (Figure 3). C. salexigens EupR was previously recognized as playing a function in compatible solute absorption, whereas NarP and NarL can regulate nitrogen metabolism. The heterologous expression of DR_1558, a NarL-like family response regulator in D. radiodurans, resulted in similar phenotypic properties with those of deipr_0871, such as high oxidative stress resistance, despite their low homology (38%). A previous study revealed that the oxidative stress resistance of the E. coli strain harboring DR_1558 was enhanced via an increase in rpoS transcripts. Additionally, other DR_1558 homologs in Antarctic bacteria can upregulate rpoS transcripts [13]. RpoS is an alternative sigma factor that plays a central role in adaptation to many suboptimal growth conditions by controlling the expression of many genes, including those affecting phenotypic traits such as metabolic pathways and the expression of genes required to survive nutrient deprivation [20].

Multiple alignments of amino acid composition located in the putative DNA binding site suggested that Deipr_0871 may control the stress resistance of E. coli strains in a manner different from that of other regulators. It was also confirmed by a transcriptional change in the recombinant E. coli strain in the presence of deipr_0871. The heterologous expression of deipr_0871 did not regulate rpoS or its downstream genes (Figure 4B). Under normal growth conditions, SodA and SoxS transcripts were highly expressed in the presence of deipr_0871, which increased the stress resistance of the E. coli strain. In E. coli, two major regulatory defense systems respond to oxidative stress: oxyR and soxRS. [36]. OxyR responds to hydrogen peroxide and induces the expression of catalases such as katG, dps, and ahpC, whereas SoxRS responds to redox-active compounds and regulates superoxide dismutase (SodA). [36]. SodA plays a role in decreasing the levels of cytotoxic ROS [39], which has the properties of cellular responses to oxidative stresses for detoxification [36]. Moreover, SoxS can improve NADPH pools and promote antioxidant defense by mediating the reduction of thioredoxins or glutaredoxins [40,41].

According to Shery et al. and Henard et al., SoxS activates the expression of genes involved in carbon metabolic pathways, such as glycolysis and the TCA cycle [40,41]. The SoxS deletion mutant shows reduced glucose uptake and growth rates [42]. This can explain the effect of deipr_0871 on growth, as the growth rate in the presence of deipr_0871 was higher than that in the control strain. Interestingly, glucose supplementation altered the metabolic ability of the E. coli strain in the presence of deipr_0871, which boosts the growth rate and cell density of the E. coli (Figure 6). In addition, it stimulated PHB production in the recombinant E. coli strain. PHB is a bioplastic produced intracellularly by microorganisms to save energy as a carbon source [43,44]. The sufficient provision of acetyl-CoA and the availability of the cofactor NADPH significantly impact the synthetic efficiency of PHB, and metabolic engineering to expand the supply of NADPH has been attempted to boost PHB production [45].

As shown in Figure 7, this suggests that Deipr_0871 differentially regulates the central carbon metabolism compared to DR_1558. NADPH-generating genes (zwf, icdA, sucA, and mdh) in the pentose phosphate pathway and TCA cycle were upregulated in the presence of deipr_0871, indicating an improvement in cell growth by generating energy (ATP, NADPH, etc.) through the pentose phosphate and TCA pathways. The higher expression levels of genes related to the pentose phosphate pathway, glycolysis, and TCA cycle produce more energy in the form of ATP, NADH, and NADPH for cell growth, resulting in the increased proliferation period of pRad-0871 strains. High concentrations of NADPH also activate acetoacetyl-CoA reductase (phaB) involved in the synthesis of PHB. In addition, the overflow of glycolysis and the TCA pathway generate and accumulate numerous pyruvate and acetyl-CoA precursors of the product (PHB), which are elevated.

Using a genomic library of D. proteolyticus, we isolated the NarL-like response regulator Deipr_0871. The introduction of deipr_0871 in E. coli showed improved oxidative stress resistance and growth rate through the soxS regulation system. This is different from the other Deinococcal NarL-like response regulator, DR_1558. Moreover, its mechanism in metabolic pathways and stress resistance was employed to improve PHB production by boosting the acetyl-CoA and NADPH generation pathways, respectively. It concluded that deipr_0871 can also be employed to improve the capabilities of industrial strains.