This project aims to develop precisely-sized and biodegradable polyhydroxyalkanoate (PHA) microbeads through the application of advanced synthetic biology strategies, in order to address the environmental crisis caused by persistent plastic microbeads. We have modified E. coli using a modular system, enabling the controlled production of these microbeads. Our goal is to create PHA microbeads with high uniformity, monodispersity, and customizable sizes, to meet practical application requirements and contribute to green production.

Our modular system can be divided into the following parts (Table 1):

• ​​PHA synthesis module:​​ phaCAB operon for high-yield polymer production.
• ​​Size-regulation module:​​ employing hydrophobic proteins (PhaP and spider silk ADF3) to control granule uniformity.
• ​​Time control logic module:​​ using inducible promoters (T7, cspA, blue-light, arabinose) to separate synthesis and assembly phases, optimizing efficiency.

In this PHA synthesis module, we utilized PHA synthesis genes derived from Ralstonia eutropha, which exhibit high compatibility with heterologous hosts such as E. coli. Heterologous expression of phaC in E. coli enables the polymerization of short-chain-length (C3–C5) hydroxyacyl-CoA substrates into biodegradable polyesters [1]. However, although PhaC alone is capable of polymerizing PHA precursors, its activity is limited by the host’s innate ability to supply the essential monomer, (R)-3-hydroxyacyl-CoA, resulting in a substantial metabolic burden on the host cells. To address this limitation, we evaluated the ​​complete gene cluster—phaC, phaA, and phaB​​—which facilitates autonomous precursor generation and polymerization at the same time. This approach was employed to assess whether it mitigates host metabolic burden and enhances PHA yield (Table 1).

In the size-regulation module, we integrated a size-regulation system into the PHA synthesis framework using two distinct hydrophobic proteins: ​​PhaP​​ and ​​ADF3​​. PhaP, a native PHA granule-associated protein, modulates granule size via surface charge-mediated dispersion, while ADF3—a spider silk-derived protein—leverages its amphiphilic structure to promote controllable aggregation through hydrophobic adsorption and electrostatic repulsion. This strategy was designed to control the size of PHA particles, which is of importance for applications that require the fabrication of uniform microspheres.

In the time control logic module, T7 promoter​​, ​​cold-inducible cspA promoter​​, blue-light and araBAD promoter, to achieve precisely timed control over PHA synthesis and granule assembly. The ​​T7 promoter​​ was utilized to drive strong, IPTG-induced expression of the core PHA synthesis operon (phaCAB), ensuring high carbon flux toward polymer accumulation during the growth phase. Subsequently, the ​​cold-inducible cspA promoter​​, the ​​optogenetic blue-light-responsive system​​, and the ​​L-arabinose-inducible araBAD promoter​​ were employed to separately regulate the expression of size-modulating proteins (PhaP, ADF3). This temporal separation—synthesizing the PHA core first before initiating coating protein production—prevented metabolic burden and resource competition, significantly enhancing both the yield and uniformity of PHA granules.

Table 1 The related gene used in our modular system.

To establish a modular system for tunable PHA production, we employed distinct plasmid strategies for different genetic modules. For the core PHA synthesis module—comprising either the single gene phaC or the complete phaCAB operon—we selected the ​​pRSFDuet-1 vector​​ due to its compatibility with E. coli, medium copy number, and dual multiple cloning sites, facilitating stable co-expression of multiple genes. Separately, the gene encoding the hydrophobic protein ​​PhaP​​ was also cloned into the pRSFDuet-1 backbone to ensure coordinated regulation and expression balance with the synthesis genes.

For the incorporation of the ​​spider silk protein ADF3​, we adopted a ​​dual-plasmid co-transformation approach​​. The ADF3 gene was constructed into a compatible plasmid pBAD, allowing it to be co-transformed with the pRSFDuet-1 plasmid construct into the same E. coli host. This strategy enables the assembly of a multi-modular system where PHA synthesis.

Fig. 1 Plasmid schematic diagram of our project.

As shown in Fig. 2(A, C, D), the agarose gel electrophoresis confirmed successful amplification of all target DNA fragments: phaC gene (2249 bp), phaCAB operon (4338 bp), pRSFDuet-1 vector (3646 bp); phaP-Kan (3005 bp), Lac-phaC (3665 bp), Lac-phaCAB (5746 bp), Plight (3191 bp), ADF3-MBP (3061 bp) and pBAD vector (4053 bp). For pRSFDuet-1-phaC/phaCAB-P_cspA-phaP (Fig. 2B), the phaC or phaCAB gene and the P_cspA-phaP​​ module were individually digested with XbaI and HindIII, then ligated into the pRSFDuet-1 backbone. All samples showed sharp bands at their expected molecular weights.

Fig. 2 The agarose gel electrophoresis of DNA amplification.

(A) pRSFDuet-1-phaC/phaCAB. (B) pRSFDuet-1-phaC/phaCAB-P_cspA-phaP. (C) pRSFDuet-1-phaC/phaCAB-P_light-phaP. (D) pBAD-ADF3-MBP.

After obtaining the DNA fragments, they are assembled through homologous recombination or enzyme ligation. After transferring the recombinant product into E. coli DH5α, we conducted colony PCR identification. We expanded the cultures of transformants exhibiting bands of the correct size and extracted plasmids, which were then sent to the company for sequencing. The sequencing results confirmed that all extracted plasmids contained the correct sequences without mutations, indicating that we successfully obtained these recombinant plasmids for PHA synthesis and size regulation (Fig.3). Following the verification of plasmid sequences, the recombinant plasmids were successfully transformed into E. coli BL21(DE3). Positive clones, confirmed by colony PCR, were inoculated and cultured overnight for the following test.

Fig. 3 The colony PCR and sequencing results.

For the pBAD-ADF3-MBP plasmid, we conducted colony PCR and sequencing analysis to confirm the successful construction of it (Fig. 4). The sequence-verified pBAD-ADF3-MBP plasmid was subsequently transferred into E. coli BL21(DE3). Then, it was prepared into competent cells and separately introduced with pRSFDuet-1-phaC or pRSFDuet-1-phaCAB plasmids, thereby generating two strains. The strains possess a complete system, including the PHA synthesis module and the size regulation module (P_araBAD-ADF3).

Fig. 4 The colony PCR and sequencing results of (A) pBAD-ADF3-MBP, (B) Co-transformation of pBAD-ADF3-MBP and pRSFDuet-1-phaC, and (C) Co-transformation of pBAD-ADF3-MBP and pRSFDuet-1-phaCAB.

To assess the impact of these two genetic modules on bacterial growth, the OD₆₀₀ was monitored over 36 hours and detected by a microplate reader (Table 2).

 The strain harboring the complete phaCAB operon demonstrated significantly higher growth, reaching an OD₆₀₀ of 5.989 at 36 h compared to 4.805 for the phaC-only strain, confirming the metabolic advantage of autonomous precursor supply (Fig. 5A). Under cold-inducible (cspA) regulation of phaP, the phaCAB-P_cspA-phaP strain achieved an OD₆₀₀ of 2.668 versus 2.070 for its phaC counterpart (Fig. 5B). Notably, both strains showed severe growth suppression under low-temperature induction.​​ With blue-light-inducible (P_light) control, the phaCAB-P_light-phaP strain reached an OD₆₀₀ of 6.528, significantly outperforming the phaC-P_light-phaP strain (OD₆₀₀=5.611), demonstrating that optogenetic induction minimizes metabolic burden (Fig. 5C). In the dual-plasmid system, the phaCAB+ADF3-MBP strain maintained superior growth (OD₆₀₀=4.432) over phaC+ADF3-MBP(OD₆₀₀=3.436), highlighting the robustness of the complete operon even with complex genetic loads (Fig. 5D).​​

The complete phaCAB operon consistently outperformed the single-gene system across all conditions, validating its role in reducing metabolic burden. Blue-light induction proved more favorable than cold-induction for maintaining bacterial growth. The consistent growth advantage of phaCAB strains underscores its essential role in constructing efficient PHA production systems.

Fig. 5 The Growth curve of the strain.

Table 2 The OD600 values of strains with different growth times.

SDS-PAGE analysis confirmed the successful IPTG-induced expression of PHA synthesis enzymes in recombinant E. coli strains. The ​​phaC​​ strain showed a distinct band at ~66.1 kDa corresponding to PhaC synthase, while the ​​phaCAB​​ strain exhibited clear expression of all three enzymes: PhaC (66.1 kDa), PhaA (40.6 kDa), and PhaB (26.4 kDa). Notably, differential basal leakage was observed in uninduced controls, with PhaA showing higher background expression than PhaB (Fig. 6A). Under dual-induction conditions (IPTG + low temperature), both ​​phaC-PcspA-phaP​​ and ​​phaCAB-PcspA-phaP​​ strains expressed target proteins, though PhaA levels were significantly reduced under cold induction, suggesting impaired translation efficiency or metabolic burden (Fig. 6B). For the IPTG + blue light dual-induction system, PhaA did not show any reduction (Fig. 6C). The ​​phaCAB + ADF3-MBP​​ system demonstrated co-expression of all PHA synthases and the ~100 kDa ADF3-MBP fusion protein, confirming the feasibility of the IPTG + arabinose dual-induction system has been achieved (Fig. 6D).

Fig. 6 The SDS-PAGE results of protein expression in the whole bacteria.

Next, we quantitatively analyzed the PHA produced by E. coli using the Nile red staining method. The principle of this method is that Nile red specifically binds to intracellular PHA granules, and the fluorescence intensity is proportional to the PHA concentration. To obtain accurate quantitative results, we first prepared a PHA standard curve. The results showed a strong linear relationship between RFU and PHA concentration (y = 190.45x - 12.606) with an excellent coefficient of determination (R² = 0.9959) (Fig. 7). This indicated that the standard curve was highly reliable and suitable for accurate quantification of PHA in subsequent experimental samples.

Fig. 7 PHA standard curve.

As shown in Fig. 8, Nile red staining consistently demonstrated superior PHA accumulation in strains carrying the complete phaCAB operon compared to phaC-only constructs across all induction systems. Under IPTG induction alone, phaCAB showed intense pink coloration over time versus faint phaC signal. When coupled with cold-inducible phaP, phaCAB-PcspA-phaP maintained progressive pink development while phaC-based strain remained nearly colorless. Blue-light induction further enhanced this contrast, with phaCAB-Plight-phaP exhibiting intense magent. In the ADF3-MBP co-expression system, phaCAB+ADF3-MBP achieved deeper pigmentation than its phaC counterpart. These results confirmed that the autonomous precursor supply of phaCAB is essential for high-yield PHA production, while PhaP and ADF3 effectively modulate granule formation, with optogenetic control proving most efficient for simultaneous production and size regulation.

Fig. 8 Nile Red staining of the bacterial cells.

Quantitative analysis demonstrated that strains harboring the complete ​​phaCAB operon​​ consistently achieved significantly higher PHA yields compared to those with only the ​​phaC​​ gene across all experimental systems (Fig. 9, Table 3). The ​​phaCAB​​ strain itself reached a PHA concentration of 2.014 mg/mL, vastly outperforming the ​​phaC​​ strain (0.1-0.2 mg/mL). When coupled with various size-regulation modules, the ​​phaCAB​​ system maintained its superiority, yielding 0.987 mg/mL (with P_cspA-phaP), 1.568 mg/mL (with P_light-phaP), and 1.875 mg/mL (with ADF3-MBP), which was a 2.3-fold increase over its ​​phaC​​ counterpart. This enhanced production was visually confirmed by intense pink coloration in the cultures. However, the introduction of the cold-inducible P_cspA-phaP module resulted in a noticeable decrease in final PHA accumulation compared to the ​​phaCAB​​ system alone, suggesting that low-temperature induction imposes a metabolic burden that impacts overall yield.

Fig. 9 PHA quantitative results of the bacterial.

Table 3 PHA concentration of the bacterial.

1. Comparative Analysis of PHA Synthesis Efficiency

The significantly higher PHA yield observed in strains harboring the complete phaCAB operon, compared to those expressing only the phaC gene, underscores the critical advantage of an autonomous precursor supply system. The phaCAB operon encodes a self-sufficient pathway where PhaA (β-ketothiolase) condenses acetyl-CoA to acetoacetyl-CoA, and PhaB (acetoacetyl-CoA reductase) reduces it to (R)-3-hydroxybutyryl-CoA, the direct monomer for polymerization by PhaC [3]. This integrated system relieves the host E. coli's endogenous metabolism from the burden of monomer supply, which is a major limitation for the phaC-only strain. Consequently, the phaCAB strains not only accumulated more polymer but also demonstrated superior growth kinetics, indicating reduced metabolic stress.

2. The Role of Hydrophobins in Controlling PHA Granule Size and Uniformity​

The inherent tendency of PHA to form polydisperse granules in recombinant systems poses a significant challenge for applications requiring precise material properties. Our results confirmed that co-expression of hydrophobins like PhaP or ADF3 is an effective strategy to achieve granular size control, albeit through distinct mechanisms. PhaP (phasin), a weakly hydrophobic protein, binds to the hydrophobic PHA core and acts as an emulsifier, reducing interfacial tension and preventing granule coalescence, leading to the formation of numerous, small, and uniform particles [4].

In contrast, the spider silk protein ADF3, with its strong hydrophobic domains, promotes granule aggregation and fusion, resulting in larger PHA particles. This divergent effect highlights the programmable nature of PHA morphology through surface engineering. By selecting and tuning the expression of specific hydrophobins, it is possible to tailor PHA granules for different downstream applications, such as uniform microbeads for cosmetics, thereby enhancing the versatility of bio-derived PHA

3. Impact of Induction Systems on Metabolic Burden and PHA Production​​

The choice of induction system profoundly influences the metabolic state of the production host and, consequently, the efficiency of PHA synthesis. Our findings demonstrated that cold-shock induction (cspA promoter) for PhaP expression, while functionally inducible, imposed a substantial metabolic burden on E. coli, leading to suppressed cell growth and reduced PHA accumulation. Low temperatures can impair translation efficiency, ribosomal function, and overall metabolic activity.

In contrast, optogenetic induction using a blue-light-responsive system provided a more favorable alternative. This system enabled precise temporal control without the stress associated with a temperature downshift, resulting in healthier cell growth and higher final PHA titers. The lower metabolic burden associated with blue-light induction underscores the importance of selecting induction strategies that minimize physiological stress on the host organism.

1. Precise Control of PHA Granule Size via Hydrophobin Regulation​

To meet various application requirements, our future research will focus on achieving precise control of the particle size of PHA by more effectively regulating the expression and function of hydrophobic proteins. Key strategies include the use of tunable expression systems (e.g., different promoters of other inducible system) to dynamically regulate the timing and level of hydrophobic protein production during fermentation, adjust the ratio with PHA particles, and thereby systematically control their final size and distribution.

In addition to adjusting the expression level, we will also study the structural diversity of hydrophobic proteins to establish the correlation between their hydrophobicity strength and the size regulation of particles. For instance, proteins with high hydrophobicity like ADF3 can promote the fusion of particles into larger ones (923.2 ± 147.3 nm), while PhaP usually produces smaller and more uniform particles (289.9 ± 64.44 nm). Through the construction of predictive models linking protein characteristics—such as domain hydrophobicity and charge distribution—to PHA morphology, our goal is to enable the customization of particles according to specific application requirements.

2. Functionalized PHA-Based Hybrid Systems for Advanced Cosmetics

​​ Building on the inherent biodegradability and biocompatibility of PHA, future research will focus on developing ​​multifunctional hybrid systems​​ by integrating PHA particles with natural polymers, such as chitosan, collagen, and hyaluronic acid. These composites are expected to exhibit synergistic effects: PHA will provide structural integrity and controlled degradation, while supplementary biopolymers can impart moisturizing, anti-inflammatory, or regenerative properties—critical for skincare applications such as post-inflammatory repair and anti-aging formulations [5].

PHA beads can be functionally enhanced by engineering their surface with bioactive compounds such as ​​antioxidants​​, ​​anti-inflammatory agents​​, or ​​tissue-repair molecules​​. This is primarily achieved through ​​protein engineering​​ of granule-associated proteins (GAPs), particularly ​​PHA synthase (PhaC)​​ and ​​phasins (PhaP)​​, which naturally anchor to the PHA surface. Functional domains (e.g., antioxidant enzymes like superoxide dismutase or anti-inflammatory cytokines) can be fused to these GAPs, enabling their stable display on the bead surface during in vivo assembly.

For instance, ​​PhaC fusion proteins​​ allow covalent attachment of functional moieties, ensuring high stability and density, while ​​PhaP-based display​​ exploits non-covalent hydrophobic interactions for simpler integration. Additionally, hybrid systems incorporating ​​natural polymers​​ (e.g., chitosan or hyaluronic acid) can be co-assembled to impart moisturizing or regenerative properties. This one-step biofabrication approach eliminates costly purification and chemical conjugation, yielding customizable, multifunctional beads ideal for biomedical applications like skincare or wound healing [5].

3. Development of Green and Sustainable PHA Extraction Methods​​

In order to reduce the environmental impact of the production process of PHA, future work must prioritize the adoption of solvent-free and energy-efficient extraction methods. The current methods often rely on chlorinated solvents (such as chloroform) or powerful chemicals, which can damage the integrity of PHA and pose processing challenges. To reduce environmental pollution and ensure the stability of PHA particles, in the future, we will use the following promising alternatives [6]:

• ​​Biological lysis​​: Use modified enzymes (such as lysozyme or specific proteases) to gently break down the cell wall and release PHA particles without damaging their structure.
• Water-based two-phase system (ATPS): Utilizing non-toxic polymers (such as polyethylene glycol/gelatin) to separate polyhydroxy fatty acid ester particles from cell debris in a water-based system, thereby minimizing the use of organic solvents.
• Supercritical fluid extraction: Utilizing supercritical carbon dioxide to dissolve and recover polyhydroxy fatty acid esters. The operation temperature is mild, which can maintain the quality of the polymer and enable the recycling of the solvent.

[1] CATALDI P, STEINER P, RAINE T, et al. Multifunctional Biocomposites Based on Polyhydroxyalkanoate and Graphene/Carbon Nanofiber Hybrids for Electrical and Thermal Applications [J]. ACS Applied Polymer Materials, 2020, 2(8): 3525-3534.

[2] Guerette PA, Ginzinger DG, Weber BH, Gosline JM. Silk properties determined by gland-specific expression of a spider fibroin gene family. Science. 1996 Apr 5;272(5258):112-5.

[3] Madison LL, Huisman GW. Metabolic engineering of poly(3-hydroxyalkanoates): from DNA to plastic. Microbiol Mol Biol Rev. 1999 Mar;63(1):21-53.

[4] RAN Gan-Qiao, TAN Dan, LU Xiao-Yun. Polyhydroxyalkanoate Nanoparticles: Structure, Biosynthesis, and Applications in Biotechnology and Biomedicine[J]. Chinese Journal of Biochemistry and Molecular Biology, 2016, 32(7): 745-754.

[5] Parlane NA, Gupta SK, Rubio-Reyes P, Chen S, Gonzalez-Miro M, Wedlock DN, Rehm BHA. Self-Assembled Protein-Coated Polyhydroxyalkanoate Beads: Properties and Biomedical Applications. ACS Biomater Sci Eng. 2017 Dec 11;3(12):3043-3057.

[6] Mondal, S., Syed, U. T., Gil, C., Hilliou, L., Duque, A. F., Reis, M. A. M., & Brazinha, C. (2023). A novel sustainable PHA downstream method. Green Chemistry, 25(3), 1137-1149.