Application of PHA surface binding proteins of alkali-tolerant Bacillus as surfactants

Abstract

Amphiphilic protein has lipophilic and hydrophilic domains, displaying the potential for development as a biosurfactant. The polyhydroxyalkanoate (PHA) surface binding protein derived from Bacillus is a type of protein that has not been studied for its emulsifying properties. In this study, PHA granule-associated protein (PhaP), PHA regulatory protein (PhaQ), and PHA synthase subunit (PhaR) derived from an alkali-tolerant PHA-producing Bacillus cereus HBL-AI were found and heterologously expressed in E. coli and purified to investigate their application as biosurfactants. It showed that the emulsification ability and stability of three amphiphilic proteins were higher than those of widely used chemical surfactants in diesel oil, vegetable oil, and lubricating oil. In particular, the PhaQ protein studied for the first time can form a stable emulsion layer in vegetable oil at a lower concentration (50 µg/mL), which greatly reduced the amount of protein used in emulsification. This clearly demonstrated that the PHA-binding protein of HBL-AI can be well applied as an environmentally friendly biosurfactants.

Introduction

Biosurfactants, amphipathic molecules of biological origin, have generated wide interest as nontoxic alternatives in most commercial and industrial applications [1]. Biosurfactant proteins, a type of biosurfactants, have hydrophilic and hydrophobic groups, which exhibits the surface-active properties commonly associated with chemical surfactants [2]. Many of these proteins are produced by microorganisms as fermentation products, which have potential advantages including biodegradability, compatibility with other co-surfactants, and low cost of production [3]. Within the range of bio-based products being developed, surfactant proteins are receiving increasing attention.

Polyhydroxyalkanoates (PHAs) surface binding proteins are a class of membrane proteins attached to the surface of PHA granules, which are produced by metabolism in microorganisms [4–6]. The PHA surface binding proteins mainly include PHA polymerase (PhaC), PHA depolymerase (PhaZ), PHA granule-associated protein (PhaP), and PHA regulatory protein (PhaQ and PhaR) [7, 8]. PhaR is also one of the subunits of class IV PHA synthase. Except for the covalent binding between PhaC and PHA particles, all the surface binding proteins of PHA have lipophilic domains and hydrophilic domains, which bind to PHA particles by hydrophobic interaction [9]. Due to the amphiphilic nature, it is possible for PHA surface binding protein to be directly used as a biosurfactant. Wei and his coworkers constructed two engineered Escherichia coli expressing PhaP [10] and PhaR [11] from Aeromonas hydrophila 4AK4. Both of them had thermal stability for application as a natural and environmentally friendly surfactant. Besides, PhaR showed a higher emulsification ability than PhaP and rhamnose. However, there is no research on the application of PHA surface binding protein from Bacillus, especially alkali-tolerant Bacillus. It is unknown whether the characteristics of proteins within resistant strains are unique. Thus, it is necessary to develop the function of PHA surface binding protein of such strains. PHA surface binding proteins of Bacillus mainly include PhaR, PhaP, PhaQ, and PhaC, which are different from Gram negative bacteria in amino acid composition and protein structure [12]. Among them, PhaQ is unique to Bacillus sp., and its characteristics as a surfactant are worth exploring.

In this study, an PHA-producing Bacillus cereus HBL-AI was obtained using Nile red and Sudan Black B in the condition of low-salt and high-alkali. Furthermore, combined with the complete genome data and functional annotation of protein-coding genes, the organization of the PHA synthesis gene cluster of this strain was found. Heterologous expression of PHA surface binding proteins (PhaR, PhaP, PhaQ, and PhaC) in E. coli was conducted. The emulsifying capacity and emulsion stability of these proteins were also investigated.

Materials and methods

Bacterial strains and culture conditions

Bacillus subtilis HBL-WI, Bacillus methylotrophicus HBL-SI, Bacillus cereus HBL-AI, Bacillus amyloliquefaciens HBL-OI, and Bacillus velezensis HBL-FI were isolated previously in our laboratory [13]. E. coli DH5α and E. coli BL21(DE3)plysS were purchased from TransGen Biotech. Luria–Bertani (LB) broth consists of yeast extract 5.0 g/L, peptone10.0 g/L, and sodium chloride 10.0 g/L (pH 6.5), which is used for the growth of Bacillus sp. and E. coli. Saline-alkali LB, with the same composition as LB, except that the salt concentration is 50 g/L and pH is 8.0, is used for screening alkali-tolerant Bacillus sp.

Evaluation and morphological identification of PHA-producing strains

The PHA-producing Bacillus strains were screened through a viable staining method on saline-alkali LB solid medium using Nile red as the staining agent [14]. The colonies were examined by exposure to UV light to detect accumulation of PHA. Single colonies with fluorescence were isolated and selected by the Sudan Black B staining method [15]. This strain was inoculated into a 250 mL LB medium shake-flask at 37 °C for 72 h, and the cells were harvested by centrifugation. Then, TEM sample of strain was prepared through fixation, dehydration, penetration, embedding, and staining as described method by Sathiyanarayanan [16] for the observation of PHA granules.

Extraction and characterization of the polymer

The extraction of PHA was performed using the chloroform by stirring 1 g of freeze-dried cells in 10 mL of chloroform at 60 °C for 12 h and purified by re-precipitation with 5 volumes of ice-cold methanol [17]. The precipitates were separated by centrifugation and dried under vacuum to obtain the purified material. The purified material was dissolved in deuterated chloroform (CDCl₃) and subjected to ¹HNMR and ¹³CNMR (600 MHz) analysis in a Bruker AVANCE III NMR spectrophotometer. It was then prepared as KBr pellet and scanned in a Perkin Elmer RX-1 FTIR spectrophotometer in the range of 4000–400 cm⁻¹. The gel performance chromatography (GPC) measurement was taken to determine the molecular weight. Gas chromatography (GC) analysis was performed to determinate the PHA by methanolysis of the polymers in sulfuric acid and methanol as described by previous studies [18]. All of the analysis was conducted in triplicate.

Construction of plasmids including gene of PHA surface binding protein

Construction of the plasmid vector pET-28a, regulated by the T7 lac promoter, was employed to express the gene encoding phaP (ASZ65083.1), phaQ (ASZ65084.1), phaR (ASZ65085.1), and phaC (ASZ65087.1). These genes were amplified from the genome of B. cereus HBL-AI by polymerase chain reaction (PCR) techniques using the primers in Table 1. The PCR fragment was then digested with BamHI and XhoI and inserted into vector pET-28a, followed by digestion with BamHI and XhoI. The pET-28a vectors were introduced into E. coli BL21 (DE3) [19]. The strain was then sent to Sangon Biotech Co., Ltd. (Shanghai, China) for sequencing, and a strain with the correct target fragment was selected for subsequent experiments.

Table 1.

Primer sequences

Gene	Primer name	Sequences (5′ → 3′)
phaP	phaP-F	CGGGATCCATGGAAACTAAACCATACG
	phaP-R	CCCTCGAGTTACTTGATGGAAGTAAATAG
phaR	phaR-F	CGGGATCCATGGGCAGCGTTCTAGATTTG
	phaR-R	CCCTCGAG TCATTTTTTATTTTCTGGTTTATTCG
phaQ	phaQ-F	CGGGATCCATGCTACAAAATGAACCAGAACAAC
	phaQ-R	CCCTCGAGTTACTCTGCTGTACCACCAGGTGA
phaC	phaC-F	CGGGATCCATGACTACATTCGCAACAG
	phaC-R	CCCTCGAGTTAATTAGAACGCTCTTCAAG

The underlined mark represents the digestion site of BamHI(GGATCC) and XhoI(CTCGAG)

Preparation of PHA surface binding protein

Four recombinant strains were separately grown in 1 L LB medium with 100 µg/mL kanamycin in 3 L shake flasks under 37 °C and 200 rpm. When OD600 of the LB cultures reached 0.6, isopropyl -d-1-thiogalactopyranoside (IPTG) was added to a final concentration of 1 mmol/mL to induce the expression of proteins, and the cells were further cultured at 28 °C and 200 rpm for 18 h to allow protein production. The cells were harvested by centrifugation (8000 × g, 10 min) at 4 °C, washed twice with 20 mmol/mL Tris–HCl buffer (pH 7.5), and lysed by sonication using an ultrasonic oscillator (Sonic Materials, Newton, CT, USA). The cell debris was removed by centrifugation (9000 × g, 15 min) at 4 °C, and the supernatant was filtered through 0.45-µm filters and applied to an affinity column (HisTrap HP; GE Healthcare, Piscataway, NJ, USA) equilibrated with 20 mmol/mL Tris–HCl (pH 7.5) and 0.5 mol/L NaCl. The sample was eluted in 20 mmol/mL Tris–HCl (pH 7.5), 0.5 mol/L NaCl, and 0.5 mol/L imidazole using a purifier system (ÄKTA pure; GE Healthcare). The purified enzyme solution was subsequently desalted and concentrated by ultrafiltration using an ultrafiltration membrane (Ultracel-10; Millipore, Billerica, MA, USA). The purified enzymes were evaluated by Coomassie Brilliant Blue staining of SDS–polyacrylamide gel electrophoresis (PAGE) gels [20].

Emulsion function of PHA surface binding protein

For emulsification activity studies, 0.5 mL aqueous protein sample (PhaR, PhaP, PhaQ and BSA) at a concentration of 500 µg/mL and 0.5 mL oil (diesel oil, vegetable oil, and lubricating oil) were mixed together in a cylindrical glass vial. Simultaneously, chemical surfactants SDS, Tween-20, and sodium oleate at a concentration of 500 µg/mL were selected as controls. The oil–water emulsions were obtained by vortex oscillation at 500 rpm for 120 s. All the samples were stored in a cool dark drawer at room temperature for 48 h or 30 d. All measurements were carried out in triplicate, and the heights of the water layer, oil layer, and emulsion layers were observed and recorded.

The emulsification value can be calculated as the following equation:

$$\mathrm{Emulsification}\mathrm{value}=\frac{\mathrm{height}\mathrm{of}\mathrm{the}\mathrm{emulsion}\mathrm{layer}\mathrm{height}}{\mathrm{height}\mathrm{of}\mathrm{all}\mathrm{layers}}\times 100\%$$

Effect of surface binding protein concentration on emulsion stability

The purified protein PhaR (1.23 mg/mL), PhaP (1.36 mg/mL), and PhaQ (1.27 mg/mL) were diluted with ultrapure water to prepare protein solutions with concentrations of 0 µg/mL, 50 µg/mL, 200 µg/mL, and 500 µg/mL, respectively. The protein solution was used for emulsification treatment with different oil (diesel oil, vegetable oil, and lubricating oil). Each group was subjected to three parallel tests and allowed to stand at a constant temperature of 25 °C. The emulsification results were recorded by taking photos.

Stability test of PhaR, PhaP, and PhaQ

The thermal stability of PhaR, PhaP, and PhaQ in diesel oil, vegetable oil, and lubricating oil was studied as follows: 500 µg/ml protein aqueous solution was heated at 60 °C and 90 °C for 30 min, then the solution was added to diesel, vegetable, and lubricating oil for emulsification study. All the samples were stored in a cool dark drawer at room temperature for 1 h. All measurements were carried out in triplicate, and the heights of the water layer, oil layer, and emulsion layers were observed and recorded as method above.

Results and discussion

Evaluation of PHA-producing strains and characterization of PHA

Out of the total Bacillus sp., one of them demonstrated to produce PHA using Nile red and Sudan Black B staining procedure on saline-alkali LB agar (pH 8.0). It can be observed that the PHA in the B. cereus HBL-AI specifically binds to Nile red and emits red fluorescence under the excitation of ultraviolet light (Fig. 1a). Under the microscope, black particles strained with Sudan Black inside the strain were observed (Fig. 1b). To further determine the accumulation of PHA, the strains fermented for 72 h were observed by TEM (Fig. 1c). It can be seen that PHA granules exist in the cells with a diameter of 200–1000 nm.

Fig. 1.

The image of different characteristics. a Fluorescence microscope image. b Sudan Black staining. c TEM image of B. cereus HBL-AI. d Infrared spectrum of polymer. d Gas chromatography–mass spectrometry of PHB sample. f ¹HNMR analysis results of extracted samples. g ¹³CNMR analysis results of extracted samples

The PHA produced by this strain was extracted. The infrared spectrum of the product is shown in Fig. 1d. 1724 cm⁻¹ is the C = O stretching vibration absorption peak, which is the characteristic absorption peak of PHA. It can be confirmed that the extracted polymer is PHA. The monomer types of the PHA polymers were analyzed by GC–MS. The GC–MS spectrum (Fig. 1e) shows that the main peak with a retention time of 4.20 min is methyl 3-hydroxybutyrate, indicating that the intracellular polymer is poly-3-hydroxybutyrate (PHB). The structure of the product is similar to the previous results by mostly isolated Bacillus strains [21]. The final product was extracted and further identified by NMR. It also proved that the intracellular polymer obtained from HBL-AI was PHB, and its monomer was 3-hydroxybutyric acid (3HB) (Fig. 1f, g).

Construction of the expression system and purification of PHA surface binding protein

The PHB synthesis gene cluster of B. cereus HBL-AI was found according to the genome-wide functional annotation of protein-coding genes [13]. HBL-AI has a pha locus that consists of phaR-phaB-phaC operon and phaJ-phaP-phaQ operon in the opposite direction (Fig. 2a). The phaR (366 bp) and phaC (1086 bp) genes encode PHA synthase subunits, whereas phaB (744 bp) encodes an NADPH dependent acetoacetyl-CoA reductase (PhaB), which plays a role in the supply of (R)-3HB-CoA monomer for PHA polymerization [22]. The phaP gene (525 bp) encodes PHA granule-associated protein (PhaP). PhaP is a non-enzymatic protein localized on the surface of PHA granules in the cells and functions to block the binding of unnecessary proteins. The phaQ gene (453 bp) encodes a P(3HB)-responsive transcriptional regulator (PhaQ) that negatively controls the expression of both phaQ and phaP [23]. The phaJ gene (540 bp) encodes R-specific enoyl-CoA hydratase (PhaJ), which is a member of the MaoC-like protein family. PhaJ is a monomer supplying enzyme from fatty acid β-oxidation [24].

Fig. 2.

PHB synthesis gene cluster of B. cereus HBL-AI (a), fragments of phaC, phaR, phaP, and phaQ by PCR amplification (b), and SDS-PAGE analysis of purified PhaP, PhaQ, and PhaR (c)

The whole genome of HBL-AI was used as a template, and the PHA surface binding protein genes of phaR (149 ng/µL), phaP (162 ng/µL), phaQ (135 ng/µL), and phaC (143 ng/µL) were successfully cloned by PCR amplification (Fig. 2b). These gene fragments were inserted into plasmids pET-28a at a ratio of 7:1, and recombinant plasmids were transformed into E. coli BL21(DE3)plysS separately. Protein expression and purification of these recombinant strains were conducted, and the purified PhaR, PhaP, and PhaQ were obtained except for PhaC that is an inclusion body. As shown in SDS-PAGE (Fig. 2c), those protein bands are single; the size of the protein is the same as the predicted molecular weight, which are 17.3 kDa, 23.9 kDa, and 20.7 kDa, respectively, for PhaR, PhaP, and PhaQ.

Emulsion function of PHA surface binding protein

Diesel oil, vegetable oil, and lubricating oil were selected as the oil phase to test the emulsifying properties of solutions of PhaR, PhaP, PhaQ, and BSA compared with the traditional chemical emulsifiers SDS, Tween-20, and sodium oleate. As shown in Fig. 3a, PhaR, PhaP, and PhaQ are of good emulsifying effects on diesel oil, with an obvious emulsifying layer. Tween-20 and sodium oleate can emulsify diesel oil to form an emulsifying layer, but the emulsifying layer is small, indicating that their emulsifying ability of diesel oil is weak. SDS produces slight emulsification at the water/oil interface. BSA fails to emulsify diesel oil. It can be seen from Fig. 3b that the emulsification values of PhaR, PhaP, and PhaQ do not change significantly after 30 d of sample standing, which indicates that the emulsion system formed is stable and the three proteins have good emulsion layer stability.

Fig. 3.

Emulsion function of PHA surface binding protein. a The emulsification effect of different surfactants on diesel. b Emulsification value of diesel emulsified with different surfactants. c Emulsifying effect of different surfactants on lubricating oil. d Emulsification value of lubricating oil emulsified with different surfactants. e Emulsification effect of different surfactants on vegetable oil. f Emulsification value of vegetable oil emulsified with different surfactants

PhaR, PhaP, and PhaQ also have a good emulsifying effect on lubricating oil, and an obvious emulsion layer can be observed. Other surfactants such as BSA, SDS, Tween 20, and sodium oleate have failed to form emulsification systems. PhaP is slightly better than PhaR and PhaQ in emulsifying effect in the lubricating oil (Fig. 3c). The emulsification values of PhaR, PhaP, and PhaQ did not change significantly after the samples were kept for 30 d (Fig. 3d), indicating that the emulsification system formed by these three proteins had good stability.

As shown in Fig. 3e, PhaR, PhaP, PhaQ, BSA, and SDS had good emulsifying effects on vegetable oil, and an obvious emulsifying layer can be observed; Tween 20 can emulsify vegetable oil, but its emulsifying effect is not obvious; sodium oleate cannot emulsify vegetable oil. It can be seen that PhaP is slightly better than PhaR and PhaQ in emulsifying vegetable oil, and the emulsifying properties of the three proteins to vegetable oil are better than other surfactants. After the sample was left standing for 30 d, the emulsification values of PhaR, PhaP, and PhaQ did not change significantly, but the emulsification values of SDS, Tween-20, and BSA all decreased significantly (Fig. 3f). It is indicated that the stability of the emulsification system formed by emulsifying vegetable oil with three protein solutions is better than that formed by other surfactants, and all three proteins have good emulsion layer stability.

Effect of surface binding protein concentration on emulsion stability

Diesel oil, vegetable oil, and lubricating oil emulsions with different concentrations of surface binding protein (PhaR, PhaP, and PhaQ) were prepared under the same vortex intensity. After 48 h of still-standing, various emulsion stabilities were observed. It can be concluded that the higher concentration of the PhaR, PhaP, and PhaQ, the more stable the emulsion (Fig. 4). Whereas the emulsifying effect of protein solutions with different concentrations on the oil phase is not the same. In the case of the diesel oil, emulsion layers started to form stably when the concentration of PhaR, PhaP, and PhaQ exceeded 500 µg/mL. Stable emulsions could be observed in the presence of 200 µg/mL of PhaR and PhaP in the case of vegetable and lubricating oil. However, the concentration of PhaQ in vegetable oil to form a stable emulsion layer was as low as 50 µg/mL, while in lubricating oil, the emulsion layer was formed at a high concentration of 500 µg/mL (Table 2). PhaR and PhaP from B. cereus have the same emulsion layers and emulsion indexes as that of PhaR and PhaP from A. hydrophila 4AK4 in the three oil phases at the same protein concentration [10, 11]. However, the minimum concentration of PhaQ (50 µg/mL) from Bacillus that achieves the same emulsification efficiency in vegetable oil is significantly lower than that of PhaR and PhaP, which also shows that PhaQ can reduce the amount of protein used in emulsification and indirectly save costs.

Fig. 4.

The emulsification effect of different concentrations of protein solution of PhaR, PhaP, and PhaQ in diesel oil, vegetable oil, and lubricating oil

Table 2.

The minimum protein concentration in the system to achieve stable emulsification effect

	Diesel oil	Vegetable oil	Lubricating oil
PhaR	500 µg/mL	200 µg/mL	200 µg/mL
PhaP	500 µg/mL	200 µg/mL	200 µg/mL
PhaQ	500 µg/mL	50 µg/mL	500 µg/mL

Stability test of PhaR, PhaP, and PhaQ

The capability of inversion and thermal stability of PhaR, PhaP, and PhaQ in emulsion was studied. Firstly, the emulsion systems of PhaR, PhaP, and PhaQ solutions with diesel oil, vegetable oil, and lubricating oil were turned upside down, and whether flow changes in the emulsion layer were observed. It can be seen from Fig. 5a that all water layers stably stayed on the upper part of the vial without leaking no matter which emulsion system it is. These also suggested that PhaR, PhaP, and PhaQ could form a fairly stable emulsion structure with the oil phases at a protein concentration of 1000 µg/mL.

Fig. 5.

The capability of inversion (a) and thermal stability of 60 °C (b) and 90 °C (c) of PhaR, PhaP, and PhaQ in diesel oil, vegetable oil, and lubricating oil

Secondly, the thermal stabilities of PhaR, PhaP, and PhaQ were investigated at 60 °C and 90 °C for 30 min. Similar to the emulsification of unheated proteins in oil, it was observed that when the three proteins were heated to 60 °C, emulsified layer was formed in the three oil phases. The emulsification values of the three proteins were 46.7%, 66.7%, and 27.6% in diesel fuel; 83.3%, 100%, and 76.7% in vegetable oil; and 100%, 100%, and 96.7% in lubricating oil, respectively. When the temperature rose to 90 °C, the three proteins could only keep good emulsification in vegetable oil, and the emulsification values of PhaR, PhaP, and PhaQ were 80%, 86.7%, and 66.7%, respectively. But the emulsifying ability of the three proteins in lubricating oil was obviously reduced and that was completely lost in diesel oil (Fig. 5b). The poor performance of the three proteins in diesel oil is in sharp contrast to the good thermal stability of PhaR from A. hydrophila 4AK4 [11], which might relate to the source of diesel oil. The structural stability of PhaR, PhaP, and PhaQ in vegetable oil after heating is also conducive to their application in food or cosmetics, because vegetable oils such as soybean oil and related derivatives are usually used as additives [25]. Amphiphilic proteins as surfactants have better biocompatibility and safety than chemical surfactants.

Conclusions

In summary, an alkali-tolerant B. cereus HBL-AI was evaluated, which had the ability to synthesize PHB. The genes encoding PHA surface binding proteins PhaR, PhaP, and PhaQ in this strain were found and heterologously expressed in E. coli. Purified soluble proteins have been proven to have excellent emulsification ability for various oils in water, which are superior to commonly used chemical surfactants. Meanwhile, PhaR, PhaP, and PhaQ also have good storage stability and thermal stability, showing great application potential in the field of food and cosmetics.

Author contribution

Conceptualization: HZ and WL; methodology: XF and SF; software: XF, SF, and JJ; validation: HZ and WL; formal analysis: XL and DL; investigation: XF, SF, XL, and DL; resources: HZ and WL; data curation: XF, SF, and JJ; writing—original draft preparation: XF and SF; writing—review and editing: HZ; visualization: HZ and WL; supervision: HZ and WL; project administration: HZ, JJ, and WL; funding acquisition: HZ and WL. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Natural Science Foundation of China (grant number 3210120332), the Central Government Guided Local Science and Technology Development Fund of Hebei Province (CN) (grant number 216Z2603G), and the Bureau of Science and Technology of Hebei Province (CN) (grant number 20322901D).

Data availability

All relevant data are within the paper.

Declarations

Conflict of interest

The authors declare no competing interests.