A modular artificial photosynthetic microbial consortium for the CO₂-driven production of cis,cis-muconate

Abstract

With the increasing restrictions on petrochemical-derived polymers and growing demand for sustainable materials, cis,cis-muconate (ccMA) has emerged in recent years as a promising platform chemical for green manufacturing. However, ccMA biosynthesis in heterotrophic microorganisms still consumes sugar- or lignin-derived feedstocks, which constrains its sustainability profile and prevents it from being a truly green or carbon-negative process. Although photosynthetic microorganisms can be engineered as “negative-carbon factories” to convert photosynthetically fixed CO₂ into diverse value-added products, the direct synthesis of ccMA has so far been precluded by intrinsic metabolic limitations, including the feedback-regulated and largely unexplored shikimate pathway as well as an NADPH/ATP imbalance that conflicts with the high NADPH demand of ccMA biosynthesis. This study presents a two-species, three-module design for sustainable ccMA biomanufacturing. Specifically, a model photosynthetic chassis Synechococcus elongatus PCC 7942 was engineered with a sucrose and trans-cinnamic acid module to supply carbon and energy, while the heterotrophic chassis Corynebacterium glutamicum ATCC 13032 carried a reconstructed shikimate module for ccMA synthesis. This modular distribution not only alleviated metabolic burden in individual strains but also enhanced system robustness through rational medium optimization, fine-tuned inoculation ratios, and spatial stabilization of cyanobacteria using sodium alginate hydrogel immobilization. While the titers remain modest (7.04 and 8.61 mg L⁻¹ under non-immobilization and immobilization, respectively), this proof-of-concept system achieved a net carbon reduction of −15.10 kg CO₂e kg⁻¹ of produced ccMA, highlighting its environmental advantages and potential for carbon-negative biomanufacturing.

---

Green foundation

1. Our work advances green chemistry by introducing a two-species, three-module framework for sustainable ccMA biomanufacturing that enables CO₂-based synthesis of high-value chemicals and provides a conceptual foundation for future carbon-negative bioprocesses.

2. The artificial photosynthetic microbial consortia (APMC) increased ccMA yield by 1.41-fold compared with monoculture, achieving a net carbon reduction of −15.10 kg CO₂e kg⁻¹ of ccMA; cyanobacterial immobilization further boosted production by 22.3%, whereas heterotrophic fermentation of C. glutamicum from sucrose typically emits 81.78 kg CO₂e.

3. The rationally designed medium and spatially stabilized cultivation framework establish a paradigm for extending modular strategies to other microbial hosts and pathways, while future work could further partition biosynthetic routes within cyanobacteria across multiple chassis to enhance division of labor.

1. Introduction

cis,cis-Muconate (ccMA) is a versatile dicarboxylic acid that serves as a key intermediate in the production of bio-based plastics, such as adipic acid-derived nylon and polyethylene terephthalate (PET).¹ Due to its central role in aromatic compound metabolism, ccMA is a promising platform chemical in the transition towards more sustainable, bio-based manufacturing processes.² With the increasing restrictions on petrochemical-derived polymers and growing demand for sustainable materials, ccMA has attracted significant attention as a promising platform chemical for green manufacturing in recent years. However, conventional chemical production of ccMA involves oxidative cleavage of benzene or benzoic acid under harsh conditions, which is energy-intensive and environmentally harmful.

To address these challenges, microbial synthesis of ccMA from renewable resource feedstocks has emerged as a promising alternative.^3,4 Recent advances have established Corynebacterium glutamicum ATCC 13032 as a versatile and industrially relevant platform for ccMA biosynthesis, utilizing the shikimate pathway to convert simple carbon sources into aromatic compounds like chorismic acid,⁵ which are then processed into ccMA. Early work by Becker et al. provided a proof-of-concept conversion of lignin into ccMA, where engineered strains accumulated 1.8 g L⁻¹ ccMA from hydrolysates generated by hydrothermal decomposition of softwood lignin.⁶ This was followed by the first large-scale validation, where Lee et al. achieved 54.0 g L⁻¹ ccMA at a yield of 0.27 mol mol⁻¹ glucose in a 50 L fermenter through strain modification and metabolic pathway redesign.⁷ Subsequent systems metabolic engineering further upgraded C. glutamicum 13032 for lignin-based aromatics, enabling high-efficiency ccMA production from complex renewable feedstocks.⁸ Complementing these advances, Velasquez-Guzman et al. developed sensitive biosensors for chorismate and ccMA, providing essential tools for pathway optimization and real-time metabolic monitoring.⁹ Most recently, Li et al. reconstructed the shikimate pathway and optimized key biosynthetic nodes, enabling recombinant strains to achieve 88.2 g L⁻¹ ccMA with a yield of 0.30 mol mol⁻¹ glucose in a 5-L bioreactor—the highest reported to date for glucose-based production in C. glutamicum 13032.¹⁰ Together, these studies show steady advances, emphasizing the metabolic versatility, process robustness, and industrial relevance of C. glutamicum 13032 as a production chassis for sustainable ccMA. Nevertheless, ccMA biosynthesis in C. glutamicum 13032 still consumes sugar- or lignin-derived feedstocks, which constrains its sustainability profile and prevents it from being a truly green or carbon-negative process.

Cyanobacteria such as Synechococcus elongatus PCC 7942 represent promising chassis capable of being engineered as “negative-carbon factories” that convert photosynthetically fixed CO₂ into diverse value-added products.¹¹ Through metabolic engineering, S. elongatus 7942 has been adapted for the synthesis of phenylpropanoids and other aromatic compounds (e.g., p-coumaric acid, caffeic acid, resveratrol and p-hydroxystyrene),¹² biofuels and related alcohols (ethanol, isopropanol and 2,3-butanediol),¹³ as well as free fatty acids,¹⁴ biodegradable polymers such as poly(3-hydroxybutyrate),¹⁵ and secreted sugars like sucrose.¹⁶ While heterotrophic hosts such as C. glutamicum 13032 have successfully accomplished full ccMA synthesis through extensively engineered shikimate–β-ketoadipate pathways,¹⁰ the complexity of this process renders it challenging to directly implement in cyanobacteria. The native shikimate route in cyanobacteria is stringently regulated by strong feedback inhibition at key enzymatic nodes such as 3-deoxy-D-arabino-heptulosonate-7-phosphate synthase (DAHPS) and is yet to benefit from systematic metabolic engineering strategies to relieve these constraints.¹⁷ Moreover, ccMA biosynthesis requires substantial NADPH consumption, whereas photosynthetic energy metabolism in cyanobacteria often exhibits an imbalance between NADPH and ATP supply.¹⁸ While cyanobacteria face significant challenges in executing the direct ccMA biosynthesis from CO₂, they can nevertheless serve as efficient modular precursor factories. In particular, engineered S. elongatus 7942 has been shown to secrete high levels of sucrose—as exemplified by 3.8 g L⁻¹ under optimized light and CO₂ conditions¹⁹ and up to 5.8 g L⁻¹ through metabolic rewiring²⁰—and to direct fixed carbon into aromatic precursors such as cinnamic acid when expressing phenylpropanoid pathway enzymes in Synechocystis sp. PCC 6803 achieving ∼138 mg L⁻¹trans-cinnamic acid (tCA).²¹ These capabilities position cyanobacteria as promising upstream modules that channel fixed CO₂ into essential sugars and aromatic precursors.

The organization of natural microbial consortia provides a conceptual framework for engineering cooperative systems, as different species naturally execute sequential metabolic steps via division of labor.²² Inspired by natural symbiotic relationships, the artificial photosynthetic microbial consortia (APMC) of cyanobacteria and heterotrophic bacteria have in recent years emerged as a promising strategy to drive the conversion of photosynthetically fixed carbon into diverse products. Early efforts mainly focused on metabolic pathway optimization, such as redirecting carbon flux in S. elongatus 7942, which laid the groundwork for subsequent system construction.²³ The APMC system was then applied for functional validation, exemplified by cyanobacteria supplying photosynthates to heterotrophs for polyhydroxybutyrate (PHB) synthesis, thereby showing the feasibility of this approach.²⁴ More recently, research has extended to the production of platform chemicals, including 3-hydroxypropionic acid (3-HP)²⁵ and volatile hydrocarbons,²⁶ both efficiently synthesized using CO₂ as the sole carbon source. Inspired by these studies, although photosynthetic organisms face difficulties in the de novo autotrophic synthesis of ccMA, we can rationally assemble the “CO₂ to ccMA” pathway by leveraging the advantageous routes of both S. elongatus 7942 and C. glutamicum 13032, thereby establishing a new paradigm for sustainable biomanufacturing of ccMA.

By combining the photosynthetic carbon fixation of S. elongatus 7942 with the metabolic efficiency of C. glutamicum 13032, we established an APMC–ccMA system that directly converts CO₂ into ccMA. The conversion of “CO₂ to ccMA” is organized into three separate functional modules: the Suc module, the tCA module, and the ccMA module. In this framework, engineered S. elongatus 7942 provides sucrose and tCA from CO₂, which are utilized by C. glutamicum 13032 to synthesize ccMA in an APMC–ccMA system. Through rational pathway design, medium optimization, and inoculation ratio tuning, we achieved a ccMA titer of 7.04 mg L⁻¹. Using sodium alginate hydrogel immobilization, the advanced iAPMC–ccMA system boosted ccMA production by 72.55% over monoculture and 22.30% over the non-immobilized system (APMC–ccMA). Metabolomics analysis revealed that cyanobacterial immobilization enhanced glycolytic activity and energy metabolism in C. glutamicum 13032, thereby promoting a more efficient redistribution of carbon flux toward product synthesis.

2. Experimental

2.1 Strains, plasmid construction and culture conditions

S. elongatus 7942 and C. glutamicum 13032 were engineered and applied to the construction of the synthetic consortium system. The essential gene for sucrose transport, namely, the permease-coding gene cscB, was derived from E. coli W.²⁷ The PAL, which is responsible for removing the amino group from L-phenylalanine to form tCA in the pathway of phenylpropanoid biosynthesis, was cloned from A. thaliana.²⁸ The super-strong promoter P_cpc560 and the theophylline-inducible promoter P_tho were used to direct the gene expression in S. elongatus 7942. All plasmids were prepared in E. coli DH5α and then introduced into S. elongatus 7942 by conjugation. The genome of C. glutamicum 13032 was used as a template to amplify the upstream and downstream homology arms of the catB and phdR genes. The respective homology arms were sequentially fused and ligated into the expression vector pk18mobrpsl. The resulting recombinant plasmids were introduced into C. glutamicum 13032 via double-crossover homologous recombination.²⁹ Through this process, the engineered strain PM, capable of producing ccMA, was successfully constructed.

S. elongatus 7942 and the resulting engineered strains were cultivated in BG11 medium (pH = 7.5) under a light intensity of approximately 50 μmol photons per m² per s¹ in an illuminating shaking incubator (HNY-211B, Honour, Tianjin, China) at 160 rpm and 37 °C or on BG11 agar plates in an incubator (SPX-250B-G, Boxun, Shanghai, China). To maintain the stable phenotype of sucrose secretion, appropriate antibiotics were added when necessary. C. glutamicum 13032 and the resulting strains were grown in BHI medium or agar plates with appropriate antibiotics added to maintain plasmids at 30 °C in a shaking incubator at 160 rpm or in an incubator, respectively. All strains used in this study are listed in Table 1.

Table 1 Strains used in this study

Strains	Description	Reference
S. elongatus 7942	Wild type S. elongatus 7942	Our lab
Strain P	NSI:: Pcpc560-pal-Trbcl; Spe^R	Our study
Strain C	NSII:: Ptho-cscB-Trbcl; Cm^R	Our study
Strain PC	NSII:: Ptho-cscB-Trbcl; Cm^R in strain P	Our study
C. glutamicum 13032	Wild type C. glutamicum ATCC 13032	Our lab
Cgb-ΔcatB	catB knocked out in C. glutamicum 13032	Our study
Strain PM	phdR knocked out in Cgb-ΔcatB	Our study

---

2.2 Conjugation of S. elongatus 7942

For achieving engineered S. elongatus strains, the constructs were delivered into the cells through conjugation.³⁰E. coli HB101 harboring pRL443 and pRL623 (named “helper”) and E. coli DH5α harboring the integrated plasmid with the target gene were cultivated overnight and then transferred separately into a fresh liquid LB medium with the appropriate antibiotics at a 1 : 50 ratio. When both strains grew to the exponential phase (OD₆₀₀ = 0.3–0.5), 10 mL of the cells of each strain was collected by centrifugation and washed with the fresh LB medium three times to remove all the antibiotics. Then, 0.1 mL of fresh LB was used to resuspend each strain, and the cells were mixed together and incubated at 37 °C for 30 min. During this time, 10 mL of S. elongatus 7942 cells at the exponential phase (OD₇₅₀ ≈ 1) was collected by centrifugation and resuspended in 0.2 mL of fresh BG11 medium. S. elongatus 7942 cells were mixed with the E. coli mixture mentioned above and incubated at 37 °C under light for 30 min. Then, the mixture was spread on BG11 agar plates, which were then covered by sterile cellulose filters (0.45 μm pore size). The plates were incubated under light at an intensity of approximately 100 μmol photons per m² per s¹ for 24 h, and then, the cellulose filters were transferred onto new BG11 agar plates with appropriate antibiotics.

2.3 Construction of C. glutamicum mutants

The genome of C. glutamicum 13032 was used as a template to amplify the upstream and downstream homology arms of the catB and phdR genes. The respective homology arms were sequentially fused and ligated into the expression vector pk18mobrpsl. The catB gene was knocked out in the C. glutamicum 12032 strain that used the plasmid pk18mobrpsl-ΔcatB to obtain the strain Cgb-ΔcatB. The plasmid pk18mobrpsl-ΔcatB was transferred by electroporation into the competent cells and the transformation mixture was plated on BHI agar containing 10 μg mL⁻¹ kanamycin. After a two-step double homologous recombination event, a colony sensitive to kanamycin but resistant to streptomycin was isolated and confirmed by single-colony PCR to obtain a double-crossover strain. In the Cgb-ΔcatB strain, the phdR gene was knocked out using the plasmid pk18mobrpsl-ΔphdR, resulting in the strain PM. The process was the same as the above method. The plasmid pk18mobrpsl-ΔcatB contained two fragments; each fragment was approximately 500 bp in size, corresponding to the upstream and downstream regions of the catB gene. The two PCR fragments were amplified respectively and inserted into the empty vector pk18mobrpsl. The ligation products were transferred into E. coli DH5α using the blunt-end ligation method. The resulting plasmid was named pk18mobrpsl-ΔcatB. Similarly, the plasmid pk18mobrpsl-ΔphdR was constructed in the same way.

2.4 Culture conditions for batch monoculture

In batch monoculture, strain PC was cultivated in either BG-11 or NBG medium. Cultures were performed using 20 mL working volumes in 100 mL Erlenmeyer flasks, incubated at 30 °C or 37 °C with shaking at 160 rpm. The initial optical density was adjusted to OD₇₅₀ = 0.04. The BG-11 medium contained: 1.5 g L⁻¹ NaNO₃, 0.02 g L⁻¹ Na₂CO₃, 0.075 g L⁻¹ MgSO₄·7H₂O, 0.04 g L⁻¹ K₂HPO₄, 0.001 g L⁻¹ EDTA, 0.006 g L⁻¹ citric acid, 1 mL L⁻¹ trace element solution (containing 2.86 g L⁻¹ H₃BO₃, 1.81 g L⁻¹ MnCl₂·4H₂O, 0.222 g L⁻¹ ZnSO₄·7H₂O, 0.39 g L⁻¹ NaMoO₄·5H₂O, 0.079 g L⁻¹ CuSO₄·5H₂O and 0.0494 g L⁻¹Co(NO₃)₂·6H₂O), 0.006 g L⁻¹ ferric ammonium citrate and 0.036 g L⁻¹ CaCl₂·2H₂O, adjusted to pH = 7.5. NBG medium was prepared by supplementing 8.766 g L⁻¹ NaCl, 2.29 g L⁻¹ TES and 0.36 g L⁻¹ theophylline. BG11 medium was primarily used for growth, while NBG medium was used to enhance product formation by strain PC.

Strain PM was cultured in either BHI medium or CG12 medium, using the same culture volume and conditions (20 mL in 100 mL flasks, 30 °C or 37 °C, 160 rpm, initial OD₆₃₀ = 0.04). BHI medium, used for rapid growth, transformation, and screening, contained 38.5 g L⁻¹ brain–heart infusion broth and 91 g L⁻¹D-sorbitol. CG12 medium, optimized for growth and production in shake flasks, contained: 42 g L⁻¹ MOPS, 2 g L⁻¹ urea, 10 g L⁻¹ NH₄SO₄, 0.25 g L⁻¹ MgSO₄·7H₂O, 0.01 mg L⁻¹ CaCl₂, 1 ml L⁻¹ trace element solution (1 g of FeSO₄·7H₂O, 1 g of MnSO₄·H₂O, 0.1 g of ZnSO₄·7H₂O, 0.02 g of CuSO₄, and 0.002 g L⁻¹ NiCl₂·6H₂O dissolved in a small amount of ddH₂O adjusted to pH = 1.0 and then added with water to 90 mL), 0.2 mg L⁻¹ biotin, 33.33 mg L⁻¹ PCA, 1 g L⁻¹ KH₂PO₄, and 1 g L⁻¹ K₂HPO₄.

2.5 Consortium conditions and two-stage process

The strains PC and PM were grown together in MSC medium, which consisted of NBG medium supplemented with the following component: 1 g L⁻¹ urea, 0.5 g L⁻¹ NH₄SO₄, 0.125 g L⁻¹ MgSO₄·7H₂O, 0.005 mg L⁻¹ CaCl₂, 0.5 ml L⁻¹ trace element solution (1 g of FeSO₄·7H₂O, 1 g of MnSO₄·H₂O, 0.1 g of ZnSO₄·7H₂O, 0.02 g of CuSO₄, and 0.002 g L⁻¹ NiCl₂·6H₂O dissolved in a small amount of ddH₂O adjusted to pH = 1.0 and then added with water to 90 mL), 0.1 mg L⁻¹ biotin, 16.665 mg L⁻¹ PCA, 0.05 g L⁻¹ KH₂PO₄, and 0.05 g L⁻¹ K₂HPO₄. Consortia were prepared with 20 mL working volumes in 100 mL flasks at 30 °C, under constant illumination (65 μmol photons per m² per s¹), and shaken at 160 rpm. Due to the substantial difference in sucrose production and consumption rates between strain PC and strain PM, the consortium preparation process was divided into two stages. Stage 1: strain PC was pre-cultivated alone in NBG medium for 3 days to accumulate sufficient carbon sources and metabolic precursors. Stage 2: after the initial 3-day incubation, the components in MSC medium were added to the culture, and strain PM was introduced.

2.6 Quantification of S. elongatus and C. glutamicum

For monoculture experiments, the cell densities of S. elongatus 7942 and C. glutamicum 13032 were measured spectrophotometrically at OD₇₅₀ and OD₆₃₀, respectively, using a spectrophotometer (Infinite E Plex, TECAN, Shanghai, China). Under consortium conditions, S. elongatus 7942 cell numbers were using a hemocytometer under a light microscope (BX43, Olympus, Tokyo, Japan). C. glutamicum 13032 was serially diluted, plated on BHI agar, and incubated at 30 °C for 24 hours. Colony-forming units (CFUs) were counted to determine a viable cell number.

2.7 Determination of extracellular sucrose content

Sucrose secreted by strain PC into the culture medium was quantified using a Plant Sucrose Content Assay Kit from Boxbio (Beijing, China). Culture supernatants were collected by centrifuging 40 μL of sample at 13 000 rpm for 2 min. The resulting supernatants were used directly for the assay, following the manufacturer's instructions.

2.8 Quantification of tCA and ccMA

To quantify tCA and ccMA, 500 μL of culture was sampled daily and centrifuged at 13 000 rpm for 2 min. The supernatant was filtered through a 0.22 μm nylon membrane prior to analysis. The tCA and ccMA standard solutions were prepared using the same culture media to generate calibration curves. Analytical-grade tCA standard and ccMA standards were purchased from Aladdin (Shanghai, China) and Merck (Darmstadt, Germany), respectively. Quantification was performed using high-performance liquid chromatography (HPLC) with an Agilent 1260 series system equipped with a Waters Symmetry C18 column (5 μm, 4.6 × 250 mm) and a diode array detector (DAD) at 275 nm. The mobile phase consisted of acetonitrile and 0.1% phosphoric acid aqueous solution at a ratio of 61 : 39 (v v⁻¹), delivered at a flow rate of 0.4 mL min⁻¹. The column was maintained at 25 °C, and each sample run lasted 15 min.

2.9 Immobilization of cyanobacteria using sodium alginate

The immobilization of cyanobacterial cells with sodium alginate was performed following a previously reported method with minor modifications.³¹ Briefly, a 2.5% (m v⁻¹) sodium alginate solution was prepared by stirring until homogeneous, sterilized, and set aside. An aliquot of strain PC (volume = a mL, OD₇₅₀ = b) was centrifuged at 6000 rpm for 5 min, and then the supernatant was removed. The cell pellet was resuspended in 1 mL of sterile water to a final OD of ab. This suspension was mixed with the 2.5% sodium alginate solution at a ratio of 1 : 4 (v v⁻¹), transferred into a 5 ml syringe, and dispensed dropwise into a 2.5% (m v⁻¹) CaCl₂ solution. The droplets cross-linked and solidified to form strain PC beads (OD₇₅₀ = ab/5). After standing for 30 min, the beads were rinsed three times with sterile water and then transferred into a mL of culture medium (OD₇₅₀ = b).

2.10 Quantification of chlorophyll a (Chl_a)

To quantify chlorophyll a (Chl_a) of suspension cultures, 1 mL of culture was sampled and centrifuged at 12 000 rpm for 2 min and resuspended with an equal volume of methanol, followed by freezing overnight at 4 °C in the dark. Absorbance of the supernatant was measured at 665 nm and 720 nm using a UV-1750 spectrophotometer (Shimadzu, Kyoto, Japan). The quantification of Chl_a of beads was performed following a previously reported method with minor modifications.³² The beads from the culture (corresponding to 20 mL of strain PC) were placed in a 50 mL centrifuge tube, and an equal volume of methanol (20 mL) was added, followed by freezing for 48 h at 4 °C in the dark. Measurements were done the same way as for the suspension cultures.

2.11 Quantification of cell dry weight

To quantify cell dry weight, 20 mL of culture was sampled and centrifuged at 7830 rpm for 10 min at 4 °C, and the cell pellets were transferred to 1.5 mL centrifuge tubes, followed by freezing overnight at −80 °C. The samples were dried using an LGJ-10 vacuum freeze dryer (Songyuan Huaxing, Beijing, China) and reweighed.

2.12 Evaluation of carbon-negative capability

The evaluation of carbon-negative capability was performed following a previously reported method with minor modifications.^20,33 When strain PM was used as the chassis microorganism for producing the target products, sucrose was mainly used for producing the end products and accumulating biomass, resulting in a loss of one-third of the carbon atoms to CO₂. Assuming that all of the carbon atoms in sucrose were converted into biomass, end products and CO₂, by a back-of-the-envelope calculation, the minimum carbon emission during sucrose as the sole carbon source was calculated as follows: E_CO2 = [(12 × m_Suc/M_Suc) − (W_CB × C_C/12) − (m_product/M_product × C_product)] × M_CO2. Here, m_Suc/M_Suc represents the input weight and relative molecular mass of sucrose, respectively; W_CB is the increased cell biomass; C_C is the cell biomass carbon content (51.34%);^34,35m_product/M_product indicates the output weight and relative molecular mass of the end product; and C_product is the carbon number in the product. When culturing strain PM alone, based on the carbon conservation theory, the exogenously added tCA was converted into the corresponding mass of sucrose theoretically, calculated as follows: m_Suc = (9 × m_tCA/M_tCA) × (M_Suc/12). Here, 9 indicates the number of carbon atoms in one molecule of tCA, and 12 indicates the number of carbon atoms in one molecule of sucrose. m_tCA/M_tCA represents the input weight and relative molecular mass of tCA, while M_Suc indicates the relative molecular mass. In addition, sucrose conversion to tCA releases ∼1 CO₂ molecule per product molecule via the decarboxylation of perbenzoic acid. In our APMC–ccMA system, the CO₂ absorbed by strain PC was used to accumulate biomass and produce the end products. By a back-of-the-envelope calculation, the minimum carbon absorption was calculated as follows: A_CO2 = [(W_CB × C_C/12) + (m_product/M_product × C_product)] × M_CO2. Here, W_CB indicates the increased cell biomass; C_C indicates the cell biomass carbon content (51.34%); m_product/M_product indicates the output weight and relative molecular mass of the end product; and C_product indicates the carbon number in the product.

2.13 Metabolomics analysis

The liquid chromatography-mass spectrometry (LC-MS)-based targeted metabolomics was performed according to a previously reported method with minor modifications.³⁶ Cells were harvested (3 OD₇₅₀ units) after 48 hours and centrifuged at 7380 rpm for 5 minutes at 4 °C, quenched, and then rapidly extracted with 900 μL of 80 : 20 methanol/water (v v⁻¹; pretreated at −80 °C), followed by freezing in liquid nitrogen. Intracellular metabolites were extracted through three freeze–thaw cycles. The extraction process was repeated with another 500 μL of 80 : 20 methanol/water (v v⁻¹). The supernatants were combined and filtered through a 0.22 μm syringe filter. The solvent was removed using a vacuum freeze dryer, and then 100 μL of ddH₂O was added and mixed well. Liquid chromatography-quadrupole mass spectrometry (LC-QQQ-MS) was used. Analysis was performed in the multiple reaction monitoring (MRM) mode using an XBridge amide column (Waters, Milford, MA, USA; 150 × 2.1 mm, 3.5 μm). Data were processed using Qualitative Analysis B.0 software.

3. Results and discussion

3.1 Modular design of the “CO₂ to ccMA” pathway in the APMC–ccMA system

Cyanobacteria possess the capability to fix carbon dioxide via photosynthesis and convert it into organic carbon, which can be utilized to support the growth and metabolic activity of heterotrophic microbial chassis. To assess the feasibility of establishing an APMC system with S. elongatus 7942, we first examined whether C. glutamicum 13032 could utilize sucrose as its sole carbon source. As shown in Fig. S1, C. glutamicum 13032 exhibited robust growth under these conditions, indicating its capacity to metabolize sucrose as the primary carbon substrate. In addition, efficient ccMA biosynthesis in C. glutamicum 13032 requires not only a carbon source but also a suitable metabolic precursor supplied by the cyanobacterial partner. To address this, we adopted a modular design strategy based on a tCA-centered biosynthetic route. The pathway was rationally partitioned between the two microbial chassis, with S. elongatus 7942 engineered to convert CO₂ into both sucrose and tCA, and C. glutamicum 13032 subsequently utilizing these inputs to complete the downstream conversion to ccMA (Fig. 1). In S. elongatus 7942, the pal gene encoding phenylalanine ammonia-lyase (PAL) from Arabidopsis thaliana was introduced to catalyze the conversion of phenylalanine to tCA, while the cscB gene encoding sucrose permease from E. coli W was integrated to facilitate sucrose secretion, generating strain PC (Fig. 2A). In C. glutamicum 13032, tCA enters the benzoate degradation pathway, where it is converted to catechol and subsequently to ccMA.³⁷ The latter could be further transformed into muconolactone by the activity of catechol 1,2-dioxygenase (CatB, encoded by cg2635). To redirect the metabolic flux toward ccMA accumulation, we deleted the catB gene, thereby preventing the downstream conversion of ccMA. Additionally, to alleviate transcriptional repression of the benzoate degradation pathway, the phdR gene (cg0343), encoding a MarR-type repressor implicated in the negative regulation of the phdR gene cluster, was also deleted. These modifications yielded the engineered strain designated as strain PM (Fig. 2B). These modifications collectively optimized the C. glutamicum 13032 chassis for efficient ccMA production within the synthetic consortium.

Fig. 1 Schematic diagram of the APMC–ccMA system. The tCA-centered biosynthetic route for ccMA was modularly distributed between S. elongatus 7942 and C. glutamicum 13032, whereby engineered S. elongatus 7942 utilizes CO₂ to synthesize sucrose and tCA, and engineered C. glutamicum 13032 subsequently utilizes these intermediates to synthesize ccMA.

Fig. 2 Rational design of chassis suitable for establishing the APMC–ccMA system. Schematic representation of the construction of the engineered S. elongatus 7942 strain PC (A) and C. glutamicum 13032 strain PM (B). The growth and production performance of strain PC were evaluated in NBG medium by measuring OD, sucrose and tCA titer at 30 °C (C) and 37 °C (D), respectively. The effect of temperature on ccMA production by the engineered strain PM was evaluated by measuring ccMA titers in cultures supplemented with 15 mg L⁻¹tCA, 2 g L⁻¹ sucrose, and an additional 1 g L⁻¹ sucrose at 12 h, at 30 °C (E) and 37 °C (F), respectively.

Despite the significant differences between S. elongatus 7942 and C. glutamicum 13032 in terms of carbon and nitrogen source utilization, energy acquisition strategies, and light dependence, both require basic inorganic ions (such as Mg²⁺, Ca²⁺, K⁺, and Fe²⁺/³⁺) and trace elements (such as Mn²⁺, Zn²⁺, and Mo⁶⁺) in their media to support cellular metabolism and enzymatic activity. Using our previously established sucrose-producing medium, NBG,²⁵ we evaluated growth and production at both 30 °C and 37 °C, the optimal temperatures for C. glutamicum 13032 and S. elongatus 7942, respectively. The results showed that sucrose production in strain PC was more sensitive to temperature (day 5: 0.49 g L⁻¹ at 30 °C vs. 0.82 g L⁻¹ at 37 °C), while tCA production was relatively unaffected by temperature (Fig. 2C and D). When culturing strain PM, we supplemented either NBG or CGXII medium with sucrose and tCA and assessed its growth and production. Our results indicate that C. glutamicum exhibited comparable growth at both 30 °C and 37 °C in NBG medium (SI, Fig. S2); however, no detectable production of ccMA was observed under either condition (Fig. 2E and F). In contrast, cultivation in CGXII medium resulted in a higher ccMA synthesis rate at 30 °C compared to 37 °C. These findings suggest that further optimization of the NBG medium composition was necessary to support the compatible growth and production of both microbial strains.

3.2 Medium compatibility assessment for rational design of the APMC–ccMA system

To identify a medium compatible with both strain PC and strain PM, the components of CGXII medium were categorized into four functional groups: (I) nitrogen sources (urea and ammonium sulfate), (II) metal ions, trace elements and vitamins (MgSO₄·7H₂O, CaCl₂, FeSO₄·7H₂O, MnSO₄·H₂O, ZnSO₄·7H₂O, CuSO₄, NiCl₂·6H₂O and biotin), (III) growth-promoting and iron-facilitating factors (protocatechuic acid), and (IV) phosphates (KH₂PO₄ and K₂HPO₄). These component groups were sequentially added into the NBG medium, resulting in the formulation of four distinct media, designated M1 through M4. Using media M1 to M4, we simultaneously evaluated the performance of three modules—a Suc module, a tCA module, and a ccMA module—at 30 °C and 37 °C to assess their temperature responsiveness and expression efficiency under varying cultivation conditions. At both 30 °C and 37 °C, the stepwise addition of medium components led to a reduction of sucrose and tCA production in strain PC (left panels in Fig. 3A and C). However, strain PM was only able to produce ccMA when cultured at 30 °C in the M4 medium (left panels in Fig. 3B and D). The stepwise addition of medium components had divergent effects on the supply module (Suc and tCA) and production module (ccMA). To address this, we further tested a reduced formulation (1/2 M1–M4) (right panels in Fig. 3A–D), in which the concentration of the added components was decreased to 50% of their original levels, to explore whether lower concentrations could better harmonize their activity. When the added components were reduced by half, both sucrose and tCA production increased, with the effect being more pronounced at 37 °C. However, ccMA synthesis was still only detectable at 30 °C, and the yield in 1/2 M4 was nearly identical to that observed in M4 (Fig. 3B). These results allowed us to preliminarily identify 1/2 M4 as a culture system well-suited for the APMC–ccMA system.

Fig. 3 Medium formulation optimization for the APMC–ccMA system. The components of CGXII medium were divided into four groups—I: nitrogen sources (urea and ammonium sulfate); II: metal ions, trace elements and vitamins (MgSO₄·7H₂O, CaCl₂, FeSO₄·7H₂O, MnSO₄·H₂O, ZnSO₄·7H₂O, CuSO₄, NiCl₂·6H₂O and biotin); III: growth-promoting and iron-facilitating factors (protocatechuic acid); and IV: phosphates (KH₂PO₄ and K₂HPO₄)—and sequentially added to evaluate their effects on the production performance of strains PC and PM. During the cultivation of strain PM, sucrose and tCA were supplemented exogenously. Production of sucrose and tCA production by strain PC (A), and ccMA production by strain PM (B) at 30 °C. Production of sucrose and tCA production by strain PC (C), and ccMA production by strain PM (D) at 37 °C. Various concentrations of phosphate (i.e., 0.01, 0.05, 0.1, 0.2 and 0.5 g L⁻¹) were supplemented into the M3 medium for analyzing the effect of phosphate on production in strain PC and strain PM (E and F).

Meanwhile, we found that phosphate supplementation appeared to be a critical factor, as revealed by comparisons between the M3 and M4 medium formulations, as well as between 1/2 M3 and 1/2 M4. However, when only phosphate was added to M3 or 1/2 M3 to generate M5 and 1/2 M5, respectively, strain PM failed to produce ccMA, despite strain PC maintaining normal sucrose and tCA production (SI, Fig. S3A and B). We also evaluated different combinations of the four medium components to formulate a series of media (M6, M7 and M8) and their half-strength counterparts (1/2 M6–1/2 M8) and found that the previously defined 1/2 M4 medium remained the most suitable for supporting APMC performance. All these results confirm that although phosphate is not the sole factor required for ccMA production, it is indeed a critical component of the medium. Consequently, we further investigated whether optimizing phosphate levels could enhance the overall synthetic capacity of the APMC system. Phosphate at varying concentrations (0.01, 0.05, 0.1, 0.2 and 0.5 g L⁻¹) was added to the M3 medium. Increasing phosphate levels had opposing effects on the two chassis strains in the APMC–ccMA system: it inhibited the growth and productivity of the strain PC (Fig. 3E and SI – Fig. S3C), while promoting the growth and production of strain PM (Fig. 3F and SI – Fig. S3D). Notably, ccMA synthesis was only detectable when the phosphate concentration reached 0.05. Based on these results, we identified 1/2 M3 supplemented with 0.05 phosphate as the optimal medium for this consortium, hereafter designated as MSC. Moreover, a control experiment without PCA supplementation (medium M7) showed no significant difference in ccMA titre (17.74 mg L⁻¹) compared with the PCA-containing medium M4 (18.67 mg L⁻¹), suggesting that PCA does not contribute to ccMA biosynthesis. This is consistent with previous studies indicating that PCA cannot be metabolized into ccMA without the aroY gene.^8,10

In synthetic microbial consortium systems, where distinct microbial species with differing metabolic requirements must coexist and cooperate, the medium not only serves as a source of essential nutrients but also plays a central role in regulating physiological compatibility, metabolic exchange, and overall system stability.³⁸ Standard media optimized for monocultures often fail to meet the nutritional and physicochemical needs of all partners in a mixed system, leading to unbalanced growth, metabolic competition, or reduced productivity. Several studies have highlighted the importance of rational medium design in synthetic microbial consortium systems. For example, Zhao et al. reformulated BG-11 to include buffering agents, antioxidant components, and appropriate micronutrient levels to support heterotrophic viability under phototrophic stress, resulting in significantly improved metabolic output.³⁹ Similarly, Chen et al. established a marine autotroph–heterotroph consortium system and used Synechococcus sp. PCC 7002 and Rhodopseudomonas palustris for the production of β-caryophyllene.⁴⁰ The successful implementation of this system relied heavily on the design of a tailored medium (BG11-RP) that supported both organisms’ physiological needs. By replacing nitrate with ammonium and glutamate, adjusting salt concentrations to accommodate marine cyanobacteria, and incorporating buffering agents, the authors created a balanced environment that enabled robust consortium performance. In this study, we focused on the development of a compatible culture medium, MSC, tailored to support the establishment of an APMC–ccMA system comprising C. glutamicum and cyanobacteria. Although several synthetic microbial consortium media have been developed for other heterotrophs, such as E. coli or P. putida, no dedicated formulations have been designed for C. glutamicum, a chassis organism widely used in industrial biotechnology. Our work fills this gap and provides a valuable reference for future engineering of synthetic microbial consortia involving photosynthetic microorganisms and C. glutamicum.

3.3 Optimization of inoculation ratios for APMC–ccMA system construction

The initial inoculation ratio plays a pivotal role in synthetic consortium systems, as it not only significantly influences the final community composition but also regulates carbon source utilization and interspecies interactions.⁴¹ Studies using E. coli and Pseudomonas putida as model organisms have shown that certain ratios (such as 1 : 1 or 1000 : 1) are more conducive to establishing metabolic cooperation, thereby markedly enhancing carbon utilization efficiency. In contrast, extreme ratios (e.g., 1 : 1000) can impair system functionality. These findings highlight that optimizing the initial inoculation ratio between S. elongatus 7942 and C. glutamicum 13032 is a critical prerequisite for constructing a stable and efficient synthetic microbial consortium. We implemented a range of initial autotroph-to-heterotroph inoculation ratios (16 : 1, 8 : 1, 4 : 1, 2 : 1, 1 : 1, 1 : 2, 1 : 4 and 1 : 8) and cultivated the consortium in a precisely formulated MSC medium (Fig. 4A). Samples were collected at 12-hour intervals to quantify the concentrations of sucrose, tCA, and ccMA, enabling assessment of metabolic dynamics under different inoculation conditions. Regardless of the initial inoculation ratio, sucrose was rapidly depleted in all synthetic consortium conditions (Fig. 4B). In contrast, the consumption rate of tCA decreased as the proportion of strain PM in the inoculum was reduced (Fig. 4C). Notably, the production of ccMA did not exhibit a linear correlation with the abundance of the heterotrophic partner. Instead, the highest titer of ccMA (7.04 mg L⁻¹) was observed at an initial autotroph-to-heterotroph ratio of 1 : 1, suggesting that optimal metabolic coordination, rather than maximum heterotroph density, was key to maximizing product yield (Fig. 4D). In the 1 : 1 inoculation group, the production of ccMA began earlier and reached its peak at 48 h, whereas other groups showed delayed onset or lower maximal titers. At higher heterotroph inoculation ratios (e.g., 1 : 8), strain PM rapidly outcompeted the cyanobacterial population, leading to an imbalance in carbon exchange and reduced productivity (SI, Fig. S4A and B).

Fig. 4 Medium formulation optimization for the APMC–ccMA system. Optimization of inoculation ratios for the APMC–ccMA system. Optimization of the initial autotroph-to-heterotroph inoculation ratios (16 : 1, 8 : 1, 4 : 1, 2 : 1, 1 : 1, 1 : 2, 1 : 4 and 1 : 8) was carried out as illustrated in the schematic (A). To assess the metabolic dynamics under different inoculation conditions, samples were collected to quantify the concentrations of sucrose and tCA from strain PC (B and C) and ccMA from strain PM (D).

To further investigate the contribution of cyanobacteria to product synthesis, we conducted a control experiment in which strain PM was cultured alone with exogenously supplied sucrose (0.27 g L⁻¹) and tCA (7.79 mg L⁻¹) at the same initial concentrations used in the synthetic consortium system. Under these conditions, the final titer of ccMA reached only 4.99 mg L⁻¹, representing a nearly 30% lower compared to the consortium system (SI, Fig. S5). This result suggests that the continuous supply of the carbon source and precursor by cyanobacteria played a crucial role in supporting sustained biosynthesis and enhancing the product yield. Similar findings have been reported in previous studies, where photosynthetic consortia outperformed monocultures due to stable metabolic interactions and dynamic precursor exchange. For instance, Hays et al. showed that the continuous production of sucrose by engineered cyanobacteria enhanced the performance of heterotrophic partners.⁴² Zhou et al. showed that distributing metabolic pathways across microbial consortia led to higher productivities than single-strain systems.⁴³ Collectively, these results support the notion that the APMC system offers distinct advantages for bioproduction through sustained metabolic cooperation and resource supply.

By distributing different pathway modules across specialized microbial hosts, researchers can optimize each chassis for distinct biochemical functions, thereby improving flux balance, minimizing toxic intermediate accumulation, alleviating metabolic burden, and ultimately enhancing system modularity, tunability, and production efficiency.⁴³ Leveraging this capability, we previously constructed a synthetic consortium of cyanobacteria and E. coli for the biosynthesis of 3-hydroxypropionic acid (3-HP).²⁵ In this study, to enable the direct synthesis of ccMA from CO₂, the biosynthetic pathway was distributed between S. elongatus 7942 and C. glutamicum 13032. The cyanobacterial chassis was engineered to produce both sucrose and tCA, which prompted us to consider whether separating these two biosynthetic routes into different strains could further optimize the supply of carbon sources and precursors. To explore this idea, we conducted experiments in which the Suc module and tCA module were allocated to separate microbial hosts. Interestingly, this separation led to an increase in tCA production, suggesting that pathway segregation can alleviate metabolic burden and enhance precursor yield (SI, Fig. S6). These findings point to the potential of employing a three-strain consortium system to further improve ccMA production. However, such a system would require systematic investigation into key parameters such as strain ratios and inoculation timing to achieve optimal coordination and productivity.

3.4 Engineering and metabolic analysis of an advanced spatially fixed APMC–ccMA system

In synthetic photosynthetic consortium—particularly the APMC–ccMA systems composed of photosynthetic microorganisms such as cyanobacteria paired with various heterotrophic partners—the spatial transfer efficiency of metabolites is often a major bottleneck limiting overall productivity.⁴⁴ Improving the local availability of key metabolites is therefore critical for enhancing the performance of APMC–ccMA systems. One promising strategy is to immobilize autotrophic cells within defined spatial structures, such as hydrogel beads or microcarrier surfaces, which can concentrate secreted metabolites in close proximity to the heterotrophic partner, thereby facilitating rapid uptake and utilization.³² Building on this rationale, we implemented a spatially directed arrangement strategy by immobilizing cyanobacterial cells (strain PC) in a sodium alginate hydrogel. Microscopic observation and culture analysis confirmed that cells were effectively entrapped, with no detectable free cells in the surrounding medium (SI, Fig. S7A). This immobilization maintained the functionality of both sucrose and tCA biosynthetic pathways (SI, Fig. S7B and C). Although the overall sucrose titer was slightly lower compared with free-cell cultures, the sucrose production on a per-cell basis was substantially higher (SI, Fig. S7D). These findings suggest that cell immobilization can confer physiological and metabolic advantages: the physical confinement likely reduces energy investment in motility or stress responses, while the hydrogel microenvironment may stabilize cell physiology and sustain metabolic activity.^45,46 Taken together, the immobilized cyanobacteria exhibited notable advantages in metabolic performance, as evidenced by the significantly increased per-cell production of both sucrose and tCA.

This observation prompted us to ask whether such enhancement could further improve the overall efficiency of the APMC–ccMA system. To address this, we established a photosynthetic consortium using the immobilized cyanobacterial cells as the autotrophic partner, designated as the iAPMC–ccMA system (Fig. 5A). As one of the microbial partners was immobilized, it was necessary to re-optimize the initial inoculation level of the heterotrophic strain. We tested several starting concentrations (0.1, 0.2, 0.3, 0.4 and 0.5) of strain PM and found that all tested conditions supported robust cell growth (Fig. 5B), accompanied by rapid sucrose consumption (Fig. 5C). In contrast, tCA consumption correlated positively with the growth of the heterotrophic strain (Fig. 5D). Notably, the highest product titer was achieved when the initial inoculation density was 0.2, reaching 8.61 mg L⁻¹, suggesting that this level provided an optimal balance between substrate availability and metabolic demand. By comparing the product titers across monocultures, conventional synthetic consortium (APMC–ccMA) systems, and the immobilized synthetic consortium (iAPMC–ccMA) system, we found that the system constructed with immobilized cyanobacteria yielded the highest production (Fig. 5F).

Fig. 5 Engineering and metabolic analysis of an advanced iAPMC–ccMA system. Schematic of iAPMC–ccMA system construction with the immobilized cyanobacteria strain PC (A). Growth of strain PM in the iAPMC–ccMA system at different initial inoculation concentrations (B), and the corresponding content of sucrose and tCA (C and D) and ccMA (E). Comparison of ccMA yields at 48 h in three cultivation modes—monoculture, APMC–ccMA and iAPMC–ccMA (F). Metabolomics analysis of strain PM in three cultivation modes (G). In monoculture, strain PM was supplemented with sucrose and tCA at levels equivalent to those synthesized by strain PC.

To further understand the underlying metabolic basis, we analyzed the levels of central carbon metabolic intermediates in strain PM under these three conditions. One challenge in the APMC system is the difficulty of extracting metabolites, since the two microbial partners are intermingled within the same culture environment. To address this, and drawing on previous experience from our group,⁴⁷ we adopted a compartmentalization strategy in which strain PC was enclosed in a dialysis bag to facilitate sample collection (SI, Fig. S8A). Growth and product formation of strain PM remained comparable to the unseparated conditions, indicating that the dialysis-based compartmentalization did not significantly disrupt metabolite exchange or metabolic flux within the consortium (SI, Fig. S8B and C). Metabolomic profiling revealed that strain PM exhibited markedly enhanced metabolic activity in the APMC system compared with the monoculture (Fig. 5G). Specifically, intracellular levels of GLU, F6P, and FBP were significantly elevated, indicating that sucrose supply from cyanobacteria effectively boosted carbon availability and activated glycolysis.⁴⁸ In summary, gel immobilization proves to be an effective strategy for optimizing the APMC system. It not only enhances metabolic flux in the heterotrophic strain by promoting more stable and efficient substrate exchange but also facilitates physiological analysis and product recovery through improved spatial control and simplified sampling. Concomitantly, the accumulation of AMP suggested a more active energy turnover, consistent with accelerated ATP consumption to support nucleotide synthesis and cell proliferation.⁴⁹ Moreover, the general increase in amino acid pools further highlighted the reinforcement of nitrogen metabolism, reflecting elevated demands for protein biosynthesis and biomass expansion.⁵⁰ Meanwhile, levels of NAD⁺ and NADP⁺ were decreased, suggesting a more reduced intracellular redox state.⁵¹ In parallel, the downregulation of key TCA intermediates, including malate, succinate, and fumarate, indicated that carbon flux through the cycle was diminished, consistent with redistribution of metabolic resources toward biosynthetic pathways.⁵² Collectively, these results show that the APMC system stimulated both carbon and nitrogen metabolism in strain PM, thereby establishing a highly active metabolic state favorable for downstream product synthesis.

3.5 Carbon reduction and environmental implications

Constructing an artificial consortium system with photosynthetic microorganisms reduces the requirement for added organic carbon and provides a sustainable model for biosynthesis. As the current system is still at the laboratory stage, we employed a stoichiometric carbon balance approach to estimate the theoretical CO₂ sequestration potential of the APMC system. This approach follows the standard theoretical carbon balance method widely applied in cyanobacterial metabolic engineering to benchmark photosynthetic production routes.^53,54 This stoichiometric carbon-balance framework was also consistent with recent studies on phototrophic–heterotrophic consortia that reported theoretical carbon-negative biosynthesis based on CO₂-derived carbon flow.³³ Compared to conventional heterotrophic fermentation platforms that rely on sugar-based substrates and result in net CO₂ emissions, the APMC system offers a promising carbon-reduction strategy for the biosynthesis of value-added chemicals such as ccMA. Specifically, producing 1 kg of ccMA with the APMC–ccMA system resulted in a net reduction of 15.10 kg CO₂e, whereas heterotrophic fermentation of C. glutamicum from sucrose typically emits 81.78 kg CO₂e. Recent studies have highlighted the potential of autotrophic or hybrid systems to achieve net-negative CO₂ fluxes in biomanufacturing, particularly when integrated with closed photobioreactors or CO₂-rich waste streams.⁵⁵ Nevertheless, future work at a larger scale should include techno-economic analysis and life cycle assessment to support progress toward economically viable and environmentally sustainable processes.^56,57 In addition to reducing dependence on agricultural feedstocks, the APMC system represents a sustainable bioproduction model with potential for carbon offset or climate mitigation, especially if scaled using renewable light and industrial CO₂ emissions. This study shows the effectiveness of a rationally engineered APMC system, in which metabolic pathways are strategically divided between two microbial species. By assigning distinct modules to each partner, the system alleviates metabolic burden, enhances robustness, and enables fine-tuned interspecies coordination to improve production efficiency.^58,59 Moreover, the modular architecture developed here is not limited to the direct biosynthesis of ccMA from CO₂, instead, it provides a transferable framework that can be adapted to other microbial hosts. This portability expands the utility of our system, paving the way for broader applications in carbon emission reduction in biomanufacturing and sustainable production of diverse high-value compounds.

4. Conclusions

This work shows that a rationally engineered APMC system can serve as an environmentally sustainable route to produce ccMA directly from CO₂. By distributing the cinnamic acid-derived biosynthetic pathway between S. elongatus 7942 and C. glutamicum 13032 and optimizing medium compatibility and inoculation ratios, the APMC–ccMA system achieved a yield of 7.04 mg L⁻¹ with a net carbon reduction of −15.10 kg CO₂e kg⁻¹ of the product. Further immobilization of cyanobacterial cells in a sodium alginate hydrogel enhanced ccMA production by 22.3%, reaching 8.61 mg L⁻¹. In comparison, under monoculture conditions where strain PM was supplemented with sucrose and tCA equivalent to the amounts synthesized by strain PC, strain PM released 81.78 kg CO₂e kg⁻¹ of ccMA produced, highlighting the substantially higher carbon footprint of the heterotrophic route. These results underscore the environmental advantage of the APMC system, in which direct CO₂ fixation by cyanobacteria reduces the dependency on externally supplied substrates and enables carbon emission reduction in biomanufacturing.

Author contributions

K. M. drafted the manuscript; K. M., J. W. and Y. J. constructed the engineered strains and performed functional analyses; L. C. contributed to the manuscript revision; and X. S. and W. Z. designed the experiment, drafted the manuscript, and contributed to the funding for the research. All authors read and approved the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Abbreviations

G6P	Glucose-6-phosphate
S6P	Sucrose-6-phosphate
CMA	Chorismic acid
BZA	Benzoic acid
CAT	Catechol
MUC	Muconolactone
PYR	Pyruvic acid
Phe	Phenylalanine
Lys	Lysine
Asp	Aspartic acid
Leu	Leucine
Val	Valine
Ser	Serine
Ala	Alanine
NAD^+	Nicotinamide adenine dinucleotide
3-PG	3-Phosphoglycerate
PEP	Phosphoenolpyruvic acid
GLU	Glucose
AKG	α-Ketoglutaric acid
MAL	Malate
SUC	Succinate
FUM	Fumarate
NADP^+	Nicotinamide adenine dinucleotide phosphate
AMP	Adenosine monophosphate
FBP	Fructose-1,6-bisphosphate

Data availability

All the data have been included in the file and there are no other available data and materials.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5gc04738b.

Acknowledgements

This research was supported by the National Key R&D Program of China (grant number: 2024YFA0919700), the National Natural Science Foundation of China (no. 32270091), and the Haihe Laboratory of Sustainable Chemical Transformations.

References

1. F. Molinari, A. Salini, A. Vittore, O. Santoro, L. Izzo, S. Fusco, L. Pollegioni and E. Rosini, Bioresour. Technol., 2024, 408, 131190.
2. N. Z. Xie, H. Liang, R. B. Huang and P. Xu, Biotechnol. Adv., 2014, 32, 615–622.
3. S. C. Mokwatlo, B. C. Klein, P. T. Benavides, E. C. Tan, C. M. Kneucker, C. Ling, C. A. Singer, R. Lyons, V. S. i. Nogué and K. V. Hestmark, Green Chem., 2024, 26, 10152–10167.
4. G. Wang, A. Tavares, S. Schmitz, L. Franca, H. Almeida, J. Cavalheiro, A. Carolas, S. Ozmerih, L. M. Blank, B. S. Ferreira and I. Borodina, Biotechnol. Bioeng., 2022, 119, 376–387.
5. J. Kalinowski, B. Bathe, D. Bartels, N. Bischoff, M. Bott, A. Burkovski, N. Dusch, L. Eggeling, B. J. Eikmanns, L. Gaigalat, A. Goesmann, M. Hartmann, K. Huthmacher, R. Kramer, B. Linke, A. C. McHardy, F. Meyer, B. Mockel, W. Pfefferle, A. Puhler, D. A. Rey, C. Ruckert, O. Rupp, H. Sahm, V. F. Wendisch, I. Wiegrabe and A. Tauch, J. Biotechnol., 2003, 104, 5–25.
6. J. Becker, M. Kuhl, M. Kohlstedt, S. Starck and C. Wittmann, Microb. Cell Fact., 2018, 17, 115.
7. H. N. Lee, W. S. Shin, S. Y. Seo, S. S. Choi, J. S. Song, J. Y. Kim, J. H. Park, D. Lee, S. Y. Kim, S. J. Lee, G. T. Chun and E. S. Kim, Sci. Rep., 2018, 8, 18041.
8. F. Weiland, N. Barton, M. Kohlstedt, J. Becker and C. Wittmann, Metab. Eng., 2023, 75, 153–169.
9. J. C. Velasquez-Guzman, H. M. Huttanus, D. P. Morales, T. S. Werner, A. L. Carroll, A. M. Guss, C. M. Yeager, T. Dale and R. K. Jha, J. Ind. Microbiol. Biotechnol., 2024, 51, e2415213121.
10. M. Li, J. Chen, K. He, C. Su, Y. Wu and T. Tan, Metab. Eng., 2024, 82, 225–237.
11. C. Chenebault, V. Blanc-Garin, M. Vincent, E. Diaz-Santos, A. Goudet, C. Cassier-Chauvat and F. Chauvat, Biomolecules, 2023, 13, 504.
12. J. Ni, F. Tao, Y. Wang, F. Yao and P. Xu, Green Chem., 2016, 18, 3537–3548.
13. N. J. Oliver, C. A. Rabinovitch-Deere, A. L. Carroll, N. E. Nozzi, A. E. Case and S. Atsumi, Curr. Opin. Chem. Biol., 2016, 35, 43–50.
14. W. Meng, M. Xia, J. Hu, C. Wang, C. Qian, M. Zhang and W. Fu, Front. Mar. Sci., 2025, 12, 1542670.
15. L. Gonzalez-Resendiz, L. Sanchez-Garcia, I. Hernandez-Martinez, G. Vigueras-Ramirez, L. F. Jimenez-Garcia, R. Lara-Martinez and M. Morales-Ibarria, Bioresour. Technol., 2021, 337, 125508.
16. M. Santos-Merino, L. Yun and D. C. Ducat, Front. Microbiol., 2023, 14, 1126032.
17. F. Gu, C. Li, H. Zheng and J. Ni, Blue Biotechnol., 2024, 1, 2.
18. M. C. Lai and E. I. Lan, Metabolites, 2015, 5, 636–658.
19. L. Yun, R. Zegarac and D. C. Ducat, Front. Plant Sci., 2024, 15, 1378573.
20. X. Bai, G. Zhang, L. Xu, K. Li, M. Zhang and Y. Huang, Bioresour. Technol., 2025, 434, 132850.
21. K. Kukil and P. Lindberg, Microb. Cell Fact., 2024, 23, 57.
22. L. Lin, Biotechnol. Biofuels Bioprod., 2022, 15, 14.
23. D. C. Ducat, J. A. Avelar-Rivas, J. C. Way and P. A. Silver, Appl. Environ. Microbiol., 2012, 78, 2660–2668.
24. T. L. Weiss, E. J. Young and D. C. Ducat, Metab. Eng., 2017, 44, 236–245.
25. L. Zhang, L. Chen, J. Diao, X. Song, M. Shi and W. Zhang, Biotechnol. Biofuels, 2020, 13, 82.
26. Y. Cui, F. Rasul, Y. Jiang, Y. Zhong, S. Zhang, T. Boruta, S. Riaz and M. Daroch, Front. Microbiol., 2022, 13, 965968.
27. M. Sahin-Toth, S. Frillingos, J. W. Lengeler and H. R. Kaback, Biochem. Biophys. Res. Commun., 1995, 208, 1116–1123.
28. X. Zhang and C.-J. Liu, Mol. Plant, 2015, 8, 17–27.
29. T. Wang, Y. Li, J. Li, D. Zhang, N. Cai, G. Zhao, H. Ma, C. Shang, Q. Ma, Q. Xu and N. Chen, Microb. Biotechnol., 2019, 12, 907–919.
30. S. Li, T. Sun, C. Xu, L. Chen and W. Zhang, Metab. Eng., 2018, 48, 163–174.
31. T. Sun, H. Huo, Y. Zhang, Y. Xie, Y. Li, K. Pan, F. Zhang, J. Liu, Y. Tong, W. Zhang and L. Chen, ACS Nano, 2024, 18, 17694–17706.
32. G. S. Toth, V. Siitonen, L. Nikkanen, L. Sovic, P. Kallio, R. Kourist, S. Kosourov and Y. Allahverdiyeva, Biotechnol. Biofuels Bioprod., 2022, 15, 146.
33. C. Li, R. Wang, J. Wang, L. Liu, H. Li, H. Zheng and J. Ni, Angew. Chem., Int. Ed., 2023, 62, e202215013.
34. J. Lange, E. Münch, J. Müller, T. Busche, J. Kalinowski, R. Takors and B. Blombach, Genes., 2018, 9, 297.
35. J. Buchholz, M. Graf, B. Blombach and R. Takors, Biochem. Eng. J., 2014, 90, 162–169.
36. J. Ma, T. Guo, M. Ren, L. Chen, X. Song and W. Zhang, Biotechnol. Biofuels Bioprod., 2022, 15, 69.
37. N. Kallscheuer, M. Vogt, J. Kappelmann, K. Krumbach, S. Noack, M. Bott and J. Marienhagen, Appl. Microbiol. Biotechnol., 2016, 100, 1871–1881.
38. R. V. Kapoore, G. Padmaperuma, S. Maneein and S. Vaidyanathan, Crit. Rev. Biotechnol., 2022, 42, 46–72.
39. R. Zhao, A. Sengupta, A. X. Tan, R. Whelan, T. Pinkerton, J. Menasalvas, T. Eng, A. Mukhopadhyay, Y. S. Jun, H. B. Pakrasi and Y. J. Tang, Sci. Rep., 2022, 12, 22163.
40. W. Chen, Y. K. Park, L. Studena, D. Bell, P. Hapeta, J. Fu, P. J. Nixon and R. Ledesma-Amaro, Bioresour. Technol., 2024, 410, 131232.
41. C. H. Gao, H. Cao, P. Cai and S. J. Sorensen, ISME J., 2021, 15, 29–40.
42. S. G. Hays, L. L. Yan, P. A. Silver and D. C. Ducat, J. Biol. Eng., 2017, 11, 1–14.
43. K. Zhou, K. Qiao, S. Edgar and G. Stephanopoulos, Nat. Biotechnol., 2015, 33, 377–383.
44. N. Martin, T. Bernat, J. Dinasquet, A. Stofko, A. Damon, D. D. Deheyn, F. Azam, J. E. Smith, M. P. Davey, A. G. Smith, S. Vignolini and D. Wangpraseurt, J. Appl. Phycol., 2021, 33, 2805–2815.
45. T. G. Johnston, S. F. Yuan, J. M. Wagner, X. Yi, A. Saha, P. Smith, A. Nelson and H. S. Alper, Nat. Commun., 2020, 11, 563.
46. X. Liu, M. E. Inda, Y. Lai, T. K. Lu and X. Zhao, Adv. Mater., 2022, 34, e2201326.
47. J. Ma, T. Guo, M. Ren, L. Chen, X. Song and W. Zhang, Biotechnol. Biofuels Bioprod., 2022, 15, 69.
48. C. Wittmann, P. Kiefer and O. Zelder, Appl. Environ. Microbiol., 2004, 70, 7277–7287.
49. Y. Deng, D. R. Beahm, S. Ionov and R. Sarpeshkar, BMC Biol., 2021, 19, 101.
50. N. Rehm and A. Burkovski, Appl. Microbiol. Biotechnol., 2011, 89, 239–248.
51. S. Takeno, R. Murata, R. Kobayashi, S. Mitsuhashi and M. Ikeda, Appl. Environ. Microbiol., 2010, 76, 7154–7160.
52. Y. Mao, G. Li, Z. Chang, R. Tao, Z. Cui, Z. Wang, Y. J. Tang, T. Chen and X. Zhao, Biotechnol. Biofuels, 2018, 11, 95.
53. M. Santos-Merino, A. K. Singh and D. C. Ducat, Front. Bioeng. Biotechnol., 2019, 7, 33.
54. S. A. Angermayr, A. G. Rovira and K. J. Hellingwerf, Trends Biotechnol., 2015, 33, 352–361.
55. M. Pavan, K. Reinmets, S. Garg, A. P. Mueller, E. Marcellin, M. Kopke and K. Valgepea, Metab. Eng., 2022, 71, 117–141.
56. S. C. Mokwatlo, B. C. Klein, P. T. Benavides, E. C. D. Tan, C. M. Kneucker, C. Ling, C. A. Singer, R. Lyons, V. Sànchez i Nogué, K. V. Hestmark, M. A. Ingraham, K. J. Ramirez, C. W. Johnson, G. T. Beckham and D. Salvachúa, Green Chem., 2024, 26, 10152–10167.
57. A. Corona, M. J. Biddy, D. R. Vardon, M. Birkved, M. Z. Hauschild and G. T. Beckham, Green Chem., 2018, 20, 3857–3866.
58. X. Li, Z. Zhou, W. Li, Y. Yan, X. Shen, J. Wang, X. Sun and Q. Yuan, Nat. Commun., 2022, 13, 1554.
59. X. Chen, C. He, Q. Zhang, S. Bayakmetov and X. Wang, ACS Synth. Biol., 2024, 13, 183–194.

---