Engineering Success: Three-Stage DBTL for Revolutionary β-Carotene Production
Cycle 1: De Novo Construction of the β-Carotene Biosynthetic Pathway
Goal: Build substrate-free β-carotene biosynthesis in Saccharomyces cerevisiae
Design
We engineered S. cerevisiae BY4741 for substrate-independent β-carotene production by leveraging its native mevalonate pathway. The synthetic pathway featured two heterologous enzymes: carB (phytoene desaturase from Mucor lusitanicus) and carRP (bifunctional phytoene synthase/lycopene cyclase from M. circinelloides), integrated into the POX1 locus via CRISPR-Cas9. Expression was optimized using strong promoter PTEF1 (BBa_25SOU86A) with terminator TADH1(BBa_K4121040) for carB and constitutive promoter PTDH3 (BBa_K2637011) with TCYC1(BBa_K5370003) for carRP.
Build
Genetic parts were assembled via fusion PCR: codon-optimized carB/carRP were flanked by promoters/terminators amplified from S. cerevisiae BY4741 genomic DNA, and Gibson Assembly generated the integration cassette (BBa_25JCA3OI). The CRISPR system was established by Co-transformation of plasmid pRS426 alongside a CHOPCHOP-optimized POX1 targeting sgRNA. Electroporation delivered components into ΔURA3 hosts, with transformants selected on minimal medium.
Test
Genotype validation included colony PCR and Sanger sequencing of integration junctions. Engineered strain C. tropicalis1 showed progressive color change (yellow→orange) during fermentation (Figure 1A-C), confirming pathway activity. HPLC analysis detected β-carotene at 15 min and lycopene at 11 min (Figure 1D), quantifying production at 129.1 mg/L β-carotene with intermediate lycopene accumulation in shake flasks.
Figure 1. Fermentation and HPLC analysis
Learn
While de novo pathway functionality was confirmed, HPLC quantification identified substantial lycopene accumulation (45.8 mg/L), revealing inefficient substrate-to-product conversion by CarRP as the primary kinetic bottleneck. This limitation, compounded by cytosolic toxicity from hydrophobic intermediates, obstructed metabolic flux toward β-carotene; necessitating spatial compartmentalization to isolate intermediates and enhance enzymatic proximity.
We interviewed a general manager of Shanghai Huamao Chemical Company (Integrated HP). We also shared preliminary experimental results with her. He pointed out that our current β-carotene yield still falls significantly short of the requirements for industrial production, and further enhancements in yield are necessary.
Cycle 2: Signal Peptide Localization Validation and Linker Optimization
Goal: Determine HD2 signal peptide topology and optimize fusion constructs for lipid droplet targeting.
Design
Literature indicates that lipid droplet (LD)-targeting signal peptides typically localize at theC-terminus of proteins. To validate HD2’s targeting capability and structural constraints, we designed C-terminal fusion (GFP-HD2) and N-terminal fusion (HD2-GFP). Both constructs incorporated flexible GGGGS×2 linkers to minimize steric interference. Structural models generated by AlphaFold 3 quantitatively predicted linker-induced domain separation, guiding rational construct design.
Build
Codon-optimized yeGFP was fused to HD2 via N- or C-terminal flexible linkers (GGGGS×2), and the resulting expression cassettes—PTDH3-yeGFP-linker-HD2-TCYC1 (BBa_253ZON2D) and PTDH3-HD2-linker-yeGFP-TCYC1(BBa_25BE29BZ)—were cloned into E. coli vectors using Gibson Assembly for rapid functional validation.
Figure 2. Structures of the GFP fusion protein with LD-targeting signal peptide and linker
Test
Fluorescence microscopy revealed distinct localization patterns: HD2-C-terminal fusion (BBa_253ZON2D) exhibited strong colocalization with Nile Red-stained lipid droplets, while HD2-N-terminal fusion (BBa_25BE29BZ) showed diffuse cytosolic distribution with no significant LD targeting, confirming that HD2 fused to the C-terminus enables efficient lipid droplet localization whereas N-terminal fusion fails to target LDs.
Learn
HD2 exhibits strict C-terminal dependence for lipid droplet targeting, validating literature that localization signals function predominantly at the C-terminus. Structural modeling confirms flexible linkers are necessary. Consequently, for metabolic engineering, the key pathway enzymes (carRP/carB) should utilize C-terminal HD2 fusions, while N-terminal fusion fails to localize proteins to lipid droplets.
Cycle 3: Lipid Droplet Compartmentalization of the β-carotene Biosynthesis Pathway
Goal: Targeting the β-Carotene biosynthesis pathway to lipid droplets via HD2 significantly enhanced β-Carotene production
Design
To address cytotoxicity and enhance flux, we redesigned the pathway for lipid droplet (LD) targeting using the HD2 signal peptide. CarB and CarRP were C-terminally fused to HD2 with or without flexible GGGGS linkers based on the Kcat prediction (Model), exploiting LDs’ hydrophobic core for intermediate storage and product sequestration. This spatial strategy aimed to reduce endoplasmic reticulum stress while increasing pathway efficiency.
Build
The redesigned pathway cassettes were then integrated into the POX1 locus of SC-1 using the established CRISPR system, generating second-generation strain SC-2 (genotype: POX1::PTDH3-carRP-HD2-TCYC1-PTEF1-carB-linker-HD2-TADH1 BBa_25SXFI1M).
Test
Shake-flask fermentation yielded 314.8 mg/L β-carotene—a 2.4-fold increase over Cycle 1 (Figure 3). Scaling to a 5-L bioreactor (pH 5.5, DO >20%, fed-batch mode) further boosted production to 1.8 g/L at 288 h, representing a 5.7-fold improvement over flask conditions.
Figure 3. β-carotene production of SC-1 and SC-2 in shaking flask and 5-liter bioreactor
Learn
Compartmentalization increased intracellular β-carotene capacity 5.7-fold and achieved industrially relevant titers (>1 g/L threshold). In S. cerevisiae, LD saturation could be the new limiting factor, prompting design of DGAT1 overexpression constructs or hrd1 deletion for future LD size engineering.
Reference
-
Tang, S., Gao, W., Guo, Q., Wei, D., & Wang, F. Q. (2025). Orchestrating multiple subcellular organelles of Saccharomyces cerevisiae for efficient production of squalene. Bioresource Technology, 424, 132294.
-
Dasso, M. E., Centola, C. L., Galardo, M. N., Riera, M. F., & Meroni, S. B. (2025). FSH increases lipid droplet content by regulating the expression of genes related to lipid storage in Rat Sertoli cells. Molecular and Cellular Endocrinology, 595, 112403.