Design
Engineering Principles: To circumvent the drawbacks of glucose metabolism (such as the Crabtree effect leading to by-product ethanol production), utilize non-conventional carbon source red algae hydrolysate to improve the theoretical yield and process sustainability of Rh1 synthesis.
Model-Assisted Design: Adopt the latest consensus metabolic model GEM-Yeast9 as the computational foundation, which comprehensively covers the metabolic network of Saccharomyces cerevisiae. Then introduce heterologous synthesis pathways: Add the complete synthesis pathway reactions and corresponding metabolites from 2,3-oxidosqualene to ginsenoside Rh1 into the model. Then introduce the AHG metabolic pathway: Add the AHG absorption and degradation pathways, enabling it to finally enter central metabolism. Objective function definition: To simulate the resource trade-off between cell "growth" and "production", construct a weighted dual objective function Z = w · vbiomass + (1-w) · vRh1, where w is the weight factor.
Build
In silico Construction: In the expanded metabolic model, use the COBRApy toolbox to set reaction boundary conditions, simulating the implementation effect of genetic engineering. For example, close the glucose exchange reaction, open the input of galactose and AHG, and allow the output of Rh1.
Test
In silico Testing and Functional Analysis:
Step 1: Determine the optimal weight (w): Vary w from 0 to 1, maximizing the objective function Z. By analyzing the curves of biomass synthesis flux vbiomass and Rh1 synthesis flux vRh1 against w, find the balanced point w=0.55. This point avoids excessive dominance by a single objective, better aligning with the metabolic phenotype of actual engineered strains.
Step 2: Compare different carbon source strategies: Under the optimal weight w=0.55, fix the total carbon source input concentration, use glucose and galactose+AHG mixture as carbon sources respectively, and calculate the Rh1 synthesis rate vRh1 when Z is maximized.
Step 3: Analyze metabolic flux distributions: At a specific carbon source concentration (e.g., 1 mmol/gDW/h), analyze the flux distribution differences in the central metabolic pathways.
Learn
Computational test results indicate that under the same carbon molar concentration, the predicted Rh1 synthesis rate using the galactose+AHG mixed carbon source is significantly higher than using the glucose single carbon source. This theoretically validates the initial design principle, indicating the advantage of red algae hydrolysate as a carbon source.
Flux distribution analysis revealed the differences in the central metabolic network under the two carbon source strategies, explaining the potential reasons for the different Rh1 synthesis efficiencies (e.g., carbon flow direction, cofactor regeneration efficiency).
Model improvement and iterative design inspiration: The success of this cycle proves the reliability of the expanded model. The key parameters and metabolic bottleneck information learned provide precise guidance for the next DBTL cycle. For example, the rate-limiting enzymes in the red algae decomposition process can be further adjusted to seek higher yield.
CYCLE 1: Expressing Agarase and Neoagarobiose Hydrolase
Design
Through literature research, it was found that the agar in red algae needs to be hydrolyzed into monosaccharides in two steps. First, agarase hydrolyzes it into neoagarobiose, and then neoagarobiose hydrolase hydrolyzes it into galactose and 3,6-anhydro-L-galactose[1][2]. We decided to introduce heterologous hydrolases into Saccharomyces cerevisiae to achieve extracellular secretion of hydrolases.
Build
We obtained the signal peptide gene from the α-factor mutant[3](BBa_258RYIFY) of Saccharomyces cerevisiae, which contains the Kozak sequence and helps to enhance the efficiency of translation initiation. Three types of agarases[4][5][6][7][31] and two types of neoagarobiose hydrolases[5][8][32] from different sources were screened, and 6 types of plasmids were constructed. We introduced the 6 types of plasmids into the Saccharomyces cerevisiae CEN.PK2-1D by lithium acetate transformation method, and obtained 6 engineered strains (namely Sq-Ag1, Sq-Ag2, Sq-Ag3, Sq-Ag4, Sq-Ag5, Sq-Ag6).
Table 1 Gene Sources of 5 Hydrolases
Table 2 Gene Sources of 5 Hydrolases
A
B
C
D
E
F
Figure Plasmids of 6 Hydrolases
Test
We cultured the 6 strains in defective medium and conducted qualitative detection using the Lugol's iodine plate method. As shown in the figure, large transparent hydrolysis circles were formed around the colonies, indicating that all 6 strains had the ability to secrete agarase extracellularly.
Lugol's iodine staining result diagram
Learn
This cycle initially verified the ability of the reconstructed yeast to decompose agar, but we need to further conduct quantitative research on the activity of hydrolases to obtain the combination with the best activity.
CYCLE 2: Expressing Agarase and Neoagarobiose Hydrolase
Design
In Cycle 2, we confirmed the activity of two enzymes. To improve the dual-enzyme catalytic system, we plan to use a computational model to predict how well enzymes bind to substrates, find the best enzyme combination, and test it with wet experiments. The main goal is to build an efficient process for predicting and verifying enzyme function, to support future metabolic pathway construction.
Build
We made a prediction model (see Model Part 3 for details) based on protein structure modeling and molecular docking. It uses enzyme sequences to predict enzyme-substrate binding，which shows enzyme activity level. The model found that the AqAga + agaNash combination scored highest—meaning it should work best.
To check the model, we designed 7 different enzyme combinations for wet experiments, to prepare for later comparison.
Test
First, we made a standard curve for substrate concentration (to measure enzyme activity accurately).
Then we tested the 7 enzyme combinations.
Results showed AqAga + agaNash had the highest activity—matching the model. This proves the model works.
Learn
We analyzed the model’s docking results and experimental data (see Model Part 3).
AqAga and agaNash act on different agar substrates, but both use their active pockets to recognize and catalyze substrates. They work because of matching structures, polar or non-polar interactions, and suitable substrate channels.
1. How AqAga binds and catalyzes tetramers efficiently?
AqAga uses 9 hydrogen bonds (most < 3.5 Å) to hold tetramers in its active pocket. Key residues (e.g., TRP133, ASP137) form strong polar interactions (for tight binding). Hydrophobic pockets (e.g., VAL631, PHE663) also help (via van der Waals forces), but some conflicts (e.g., with GLU677) mean the binding could be better.
Tetramers can pass through the whole active channel. The binding area has both hydrophilic and hydrophobic parts—this matches polysaccharide-degrading enzymes: hydrophilic parts help substrate binding and catalysis (e.g., GLU/ASP donates protons), while hydrophobic parts help position and stabilize.
2. How agaNash binds and catalyzes dimers?
agaNash and the dimer form a moderately strong polar interaction network via 8 hydrogen bonds, among which the key hydrogen bonds formed by ASN 149 and ARG 234 provide crucial anchoring for the binding. Meanwhile, the hydrophobic interface composed of aromatic residues such as TRP 88 and PHE 76 serves as the main driving force for binding through dense negative overlapping contacts.
This binding mode exhibits local steric clashes, most prominently in the regions of PHE 125 and THR 126; these clashes may affect conformational stability. Some hydrogen bond sites also show steric strain, indicating the need for coordination between polar interactions and stereomatching.
The binding region presents a typical feature of alternating hydrophilic-hydrophobic distribution, which meets the substrate recognition requirements of glycoside hydrolases. The polar network is responsible for substrate-specific recognition, while the hydrophobic region provides positional stabilization. The synergistic effect of the two endows the system with moderate binding affinity, which can support the binding and catalytic functions towards the dimer substrate.
CYCLE 3: Expressing Agarase and Neoagarobiose Hydrolase
Design
After obtaining the Sq-Ag5 strain with the highest enzyme activity, we hoped that Sq-Ag5 could integrate the exogenous enzymatic hydrolysis pathway and the endogenous fermentation pathway, and successfully produce the rare ginsenoside precursor squalene using agar. We first attempted to conduct fermentation experiments using 20 g/L glucose and 20 g/L agar.
Build
We inoculated the Sq-Ag5 strain on a nutrient-deficient plate, then carried out shake flask fermentation in YPD medium and YPA liquid medium respectively, and finally detected the squalene content in the fermentation broth using a HPLC.
Test
Since the fermentation temperature of yeast is 30°C, the medium with only agar added solidified, making it impossible to detect the product. Strangely, the group with only glucose added also had a very low squalene content, only 5.89 mg/L, which could not provide sufficient precursor substances for the subsequent conversion of ginsenosides.
Figure HPLC Detection Chromatogram of Sq-Ag5 Squalene Fermentation Broth
Learn
After communication with Dr. Xie, it was initially determined that the MVA pathway for squalene production in Saccharomyces cerevisiae was limited. Literature research showed that up-regulating the expression of genes encoding key enzymes in the MVA pathway can effectively increase the overall flux of this metabolic pathway.
CYCLE 4: Optimization of the MVA Pathway
CYCLE 4.1: Try to Produce Squalene Using Sq-Ag5
Design
To enhance the MVA pathway for improved squalene synthesis capacity, we inserted the tHMG1 and IDI1 genes[9] into the GAL80 locus of the genome of wild-type Saccharomyces cerevisiae CEN PK2-1D. tHMG1[10][11] is a rate-limiting enzyme in the MVA pathway, catalyzing the conversion of HMG-CoA to mevalonate (MVA) and determining the metabolic flux of the MVA pathway. IDI1 catalyzes the reversible conversion between isopentenyl pyrophosphate (IPP) and dimethylallyl pyrophosphate (DMAPP). Meanwhile, knockout of the GAL80 gene[12] not only helps to relieve the inhibition of galactose utilization in subsequent experiments but also enhances the expression intensity of galactose-regulated promoters.
Build
We constructed a Cas9-sgRNA plasmid targeting the GAL80 locus, integrated the tHMG1 gene, IDI1 gene, and corresponding promoters and terminators as donor DNA, and finally performed gene editing on yeast. After primary screening and negative screening, the engineered strain S. cerevisiae Sq-0 was obtained, in which the genes were successfully integrated into the genome and the tags were removed.
Figure: IDI1 and tHMG1 Gene Modules
Test
The experimental procedure was consistent with CYCLE 4.1-Test. Under the culture condition of 20 g/L glucose, the squalene yield of strain Sq-0 was detected to reach 806 mg/L.
Figure: HP-LC Detection Spectrum of Squalene Fermentation Broth from Strain Sq-0
Learn
The detection results showed that the optimization of the MVA pathway by overexpressing the tHMG1 and IDI1 genes was highly effective, and the engineered strain Sq-0 could be used as the starting strain for subsequent engineering development
CYCLE 4: Optimization of the MVA Pathway
CYCLE 4.2: Construction of Engineered Strain Sq-0 with Overexpressed tHMG1 and IDI1
Design
According to the conclusion of CYCLE 4.1, we found that the excessively high concentration of red algae polysaccharides would increase the viscosity of the medium, which is unfavorable for the growth of strains. Therefore, it is important to explore the specific conditions for the liquefaction under the action of low-concentration hydrochloric acid: it is necessary to ensure the fluidity of the fermentation medium to facilitate the increase of microbial movement; it is also necessary to ensure that no monosaccharides are hydrolyzed during the liquefaction pretreatment.
Build
We used hydrochloric acid with a final concentration gradient of 0.001 to 0.01 M in YPA medium and carried out high-temperature treatment. To verify whether the polysaccharides were hydrolyzed during the liquefaction process, we planned to use the engineered strain Sq-Ag5 for fermentation experiments after finding the optimal liquefaction conditions to detect whether neoagarobiose and galactose were produced.
Test
The results showed that an extremely low acid concentration (0.005 M hydrochloric acid) could open the structure of agar to be in a fluid state, and the medium did not solidify during the entire shake flask culture of yeast. After liquefying the yeast medium with 0.005 M hydrochloric acid, no neoagarobiose or monosaccharides were detected by HPLC, and the medium could be used for yeast fermentation experiments after adjusting the pH to 6.0.
Figure Medium State During Fermentation
Learn
During the fermentation process, we monitored the state of the medium at different stages (0 h, 48 h, 120 h) and observed the effect of hydrochloric acid concentration on the fluidity of the medium. The above phenomena confirmed the effectiveness of this experimental system.
CYCLE 5: Fermentation Production of Squalene by the Engineered Strain Sq-Ag Using Red Algae Polysaccharides as Substrate
CYCLE 5.1: Determine Liquefaction Conditions
Design
There is a distribution limit of metabolic resources in Saccharomyces cerevisiae, so it is necessary to establish a dynamic balance between cell growth and target product synthesis[13]. In the early stage of the experiment, it was predicted by the Flux Balance Analysis (FBA) model that the flux of the Rh1 synthesis reaction is greater under the condition of mixed carbon sources. To optimize the carbon source utilization efficiency, we designed an orthogonal fermentation experiment to explore the optimal ratio of glucose to red algae polysaccharides. The strategy was to use low-concentration glucose to initiate microbial growth, while maximizing the efficiency of galactose as a carbon source.
Build
We introduced the plasmid p426-AqAga-agaNash into the engineered strain Sq-0 by lithium acetate transformation method to obtain the engineered strain Sq-Ag, and carried out orthogonal fermentation experiments in a series of media (liquefied with 0.005 M hydrochloric acid).
Test
Through detection, it was found that yeast could achieve the step-by-step degradation and saccharification of agar from scratch. The experimental results showed that 10 g/L glucose and 25 g/L agar could achieve the highest squalene yield of 667.02 mg/L.
Figure Orthogonal Experimental Results of Squalene Synthesis by Glucose and Agar in Shake Flasks
Learn
When the concentration of red algae polysaccharides was ≤25 g/L, the squalene yield increased with the increase of red algae polysaccharide concentration. However, high-concentration polysaccharides (exceeding 25 g/L) began to inhibit the increase of squalene. The possible reason is that high-concentration red algae polysaccharides lead to excessively high medium viscosity, which affects oxygen transfer or cell metabolism. This reminds us that in the development of new carbon source fermentation processes, the selection of carbon sources requires multi-dimensional systematic evaluation.
CYCLE 5: Fermentation Production of Squalene by the Engineered Strain Sq-Ag Using Red Algae Polysaccharides as Substrate
CYCLE 5.2: Orthogonal Experiment
Design
Since the squalene yield under mixed carbon source conditions was still not ideal, we aimed to further improve the squalene yield through promoter engineering. Considering the large number of promoter combinations, we planned to calculate the optimal promoter combination based on wet lab experiments combined with biokinetic system analysis, so as to provide support and reference for the conclusions of the wet lab.
Build
Through literature research, 3 constitutive promoters (PTEF, PGPD, PTDH) and 3 inducible promoters (PGAL1, PGAL7, PGAL10) were identified, and 18 plasmids with different promoter combinations were constructed using the Gibson assembly method( the same as Cycle2-Build).
Figure: Schematic Diagram of Promoter Engineering
Test
We cultured the strains for 96 hours under the conditions of 5 g/L glucose and 25 g/L agar, and detected the squalene yield. The reason for choosing this carbon source combination was that, according to Cycle5.2-Test, when the glucose concentration increased from 2.5 g/L to 5 g/L, the squalene yield showed a jump; before the agar concentration reached 25 g/L, the squalene yield increased rapidly, and after exceeding 25 g/L, the increase rate of the yield became gentle and even showed a downward trend. To meet the growth needs of cells while maximizing the utilization rate of agar polysaccharides, 5 g/L glucose and 25 g/L agar were finally determined as the carbon source conditions for the promoter engineering experiment.
The detection results showed that the inducible promoter combination PGAL10 (regulating agarase expression) + PGAL7 (regulating neoagarobiose hydrolase expression) had the best effect, and the corresponding squalene yield reached 594.68 mg/L.
Figure: Experimental Verification Diagram of Different Promoter Combinations
Learn
The experiment found that inducible promoters had significant advantages in the presence of galactose. Inducible promoters can be induced and activated by galactose, thereby accurately regulating the expression of agarase and neoagarobiose hydrolase. Different from constitutive promoters that continuously express enzymes, inducible promoters can start enzyme synthesis at the appropriate time according to the presence of galactose, avoiding the problems of excessive enzyme expression or improper expression timing that may be caused by constitutive promoters, thus utilizing substrates more efficiently and reducing the metabolic burden of cells at the same time. To further verify the accuracy of the wet lab results, we constructed a promoter prediction model.
CYCLE 6: Promoter Engineering
CYCLE 6.1: Promoter Selection (Wet lab)
Design
This cycle aimed to construct a promoter prediction model to provide systematic guidance for the screening of optimal promoter combinations in product synthesis. During the design phase, the core goal of the model was clarified: to quantify the dynamic regulatory mechanism of different promoter combinations on product yield. A complete set of ordinary differential equation (ODE) model framework was constructed by integrating the Monod kinetic equation, carbon source allocation mechanism, cell growth inhibition term, and promoter expression intensity quantification method. At the same time, the types of wet lab data required for the model were planned, including mRNA expression level, green fluorescent protein (GFP) intensity, and product yield time series, which were used for parameter calibration and verification of the model in the subsequent steps. In terms of algorithm selection, the Runge-Kutta method (ode45) was used for numerical solution, and constrained nonlinear optimization (Sequential Quadratic Programming, SQP) was combined to realize parameter fitting, so as to ensure the accuracy and reliability of the model.
For detailed process, please refer to Model Part 4.
Build
During the construction phase, the mathematical expression and preliminary implementation of the model were completed. Firstly, based on the wet lab data, the relative expression intensities of the three promoters PTEF, PGPD, and PTDH were normalized, and their weights were obtained as 0.265, 0.418, and 0.317 respectively. Furthermore, the synergy coefficients α and β for homologous and heterologous promoter combinations were defined. On this basis, an ODE system including glucose and agar consumption kinetics, cell growth equation, and product synthesis flux model was established, and a time-dependent carbon source allocation function r(t) was introduced to dynamically reflect the resource allocation strategy. To ensure the biological rationality of the model, a number of constraint conditions were set, including carbon source residual amount, lower limit of cell density, and lower limit of synthesis rate. The ODE solution and optimization algorithm code were written using MATLAB, and parameter training was conducted using data from 6 groups of known promoter combinations. To further expand the application scope of the model, inducible promoters such as PGAL1, PGAL7, and PGAL10 were also introduced, the Analytic Hierarchy Process (AHP) was used to evaluate their comprehensive scores, and the model was modified to incorporate galactose metabolism and induction delay effects.
Test
The performance of the model was comprehensively evaluated by comparing the model prediction results with independent wet lab data. For promoter combinations not involved in training (such as PTEF+PTDH, PGAL10+PGAL1), the model successfully predicted their squalene yields, and the predicted values showed good consistency with the experimental values. Statistical evaluation showed that the coefficient of determination R² of the model increased from the initial 0.7987 to 0.8384 after expansion; the K-S test result (p > 0.05) indicated that the predicted data and experimental data had the same distribution characteristics. It is particularly worth noting that the yield ranking predicted by the model was basically consistent with the experimental ranking, which proved its guiding value in practical screening. Through the visual analysis of scatter plots and confidence ellipses, it was further confirmed that the model had no significant systematic deviation, and the prediction results were reliable.
Learn
Through in-depth analysis in the learning phase, the effectiveness and improvement directions of the model were summarized. This ODE model could well capture the dynamic impact of promoter combinations on product synthesis and showed strong explanatory and predictive capabilities. However, the model had a slight underestimation phenomenon in some combinations, which might be related to the excessive setting of constraint conditions or incomplete parameter optimization. In the future, the generalization ability of the model will be improved by increasing experimental data of more promoter combinations, and machine learning methods will be introduced to assist parameter optimization, so as to further improve the fitting efficiency and prediction accuracy of the model.
CYCLE 6: Promoter Engineering
CYCLE 6.2: Promoter Selection (Model)
Design
Through literature research, it was found that inserting introns into promoters could further improve gene expression levels, thereby increasing squalene yield. Studies have shown that the RPS25Ai[27] intron in Saccharomyces cerevisiae is widely used in yeast genetic engineering and metabolic pathway modification, and has a significant effect. Therefore, we planned to chimerize the RPS25Ai intron with the PGAL10 and PGAL7.
Build
The RPS25Ai intron was inserted into the proximal regions of the PGAL10 and PGAL7. Both the PGAL10 and PGAL7 were connected to the RPS25Ai intron at the -1 position, and 4 promoter combinations containing chimeric introns were constructed.
Figure: Schematic Diagram of Chimeric Intron Construction
Test
The strains were cultured for 96 hours under the conditions of 5 g/L glucose and 25 g/L agar, and the squalene yield was detected. The experimental results showed that when the RPS25Ai intron was inserted into the proximal end of the PGAL10 and no intron was inserted into the PGAL7, the squalene yield was the highest, reaching 682 mg/L.
Figure: Squalene Yield of Four Promoter Combinations with Chimeric Introns
Learn
Inserting introns into promoters could enhance the expression of agarase and neoagarobiose hydrolase genes in Saccharomyces cerevisiae to varying degrees, thereby increasing squalene yield. After promoter engineering modification, the squalene yield was significantly improved, laying a foundation for obtaining higher yields of rare ginsenoside Rh1 in subsequent experiments.
CYCLE 6: Promoter Engineering
CYCLE6.3: Engineered Promoter (Chimeric Intron)
Design
The synthesis of the rare ginsenoside Rh1 from squalene requires five exogenous genes[28] - PgDDS (Dammarenediol-II synthase gene，BBa_255LOGT6), CYP716A47 (Protopanaxadiol synthase gene，BBa_25RY2RMX), PgCPR1 (Cytochrome P450 reductase gene), CYP716A53v2 (Protopanaxatriol synthase gene，BBa_25NSM6TW) and UGTPg100 (Glycosyltransferase gene，BBa_255ROLW2). The vector was too large when five genes were placed on one plasmid. Therefore, we assigned the five genes to low-copy and high-copy plasmids, and then simultaneously transformed the high-copy and low-copy plasmids into the engineered strain S.cerevisiae Sq-0.
Build
The genes PgDDS, CYP716A47, PgCPR1 were placed on high-copy plasmids, the genes CYP716A53v2 and UGTPg100 was placed on low-copy plasmids. After the sequencing was correct, the plasmids were transformed into the engineered strain S.cerevisiae Sq-0 by lithium acetate conversion method to obtain the engineered strain S.cerevisiae Rh1.
High-copy plasmids
Low-copy plasmids
High-copy sequencing is correct
Low-copy plasmids
Test
The medium containing 20g/L glucose was used for fermentation for 144 hours. With the LC-16 high performance liquid chromatograph equipped with SPD-16 dual-wavelength ultraviolet detector, no rare ginsenoside Rh1 and any intermediate products were detected at multiple retention times.
Liquid phase detection graph
Learn
We conducted numerous repeated experiments. Despite verifying the correctness of the gene sequence and the copy plasmid sequencing, no rare ginsenoside products or any intermediate products were detected. We consulted relevant professors. Although the cause was not identified, after communication, the professor suggested that we integrate the gene into the genome of yeast cells for the experiment.
Cycle 7: Using red algae polysaccharides as the substrate to produce the rare ginsenoside Rh1
Cycle 7.1: Construction of the biosynthetic module of rare ginsenoside Rh1 and its compatibility with Saccharomyces cerevisiae cells
Design
In order to enable Saccharomyces cerevisiae to produce the rare ginsenoside Rh1 using squalene, we redesigned the experimental protocol and integrated five exogenous genes into the corresponding genomic sites of the engineered strain S.cerevisiae Sq-0[29] using CRISPR-Cas9 technology. An engineered strain S.cerevisiae Rh1-con capable of synthesizing the rare ginsenoside Rh1 was constructed. Integrate the gene PgDDS at the X-3 site, the gene CYP716A47 and the gene PgCPR1 at the XI-3 site, the gene CYP716A53v2 and the gene UGTPg100 is integrated at the LPP1 site. Among them, since the expression of the LPP1 gene will promote the flow of the precursor substance FPP of squalene towards the synthesis of farnyl alcohol, inserting the gene at the LPP1 site can increase the yield of squalene and its downstream products[30].
Build
We constructed the Cas9-sgRNA plasmid with sgRNA sequences targeting X-3, XI-3, and LPP1 sites. Then, after gene editing of each gene, high concentrations of TADH1-PgDDS-PGAL1,10-linear donor DNA, PGAL1,10-CYP716A47-TALT1-PGAL7-PgCPR1-TCYC1 linear donor DNA, and TADH1-CYP716A53v2-PGAL1,10-UGTPg100-TALT1 linear donor DNA were obtained respectively. These linear donor DNAs were successively integrated into the genome of yeast cells through gene editing. After primary screening and negative screening, the engineering strain that successfully integrated the gene into the genome and discarded the label was obtained. After all the gene editing was completed, the engineered strain S.cerevisiae Rh1-con was obtained.
The engineered strain S.cerevisiae Rh1-con
Test
The medium containing 20g/L glucose was used for fermentation for 144 hours. The amount of the corresponding product after each gene introduction was determined using an LC-16 high performance liquid chromatograph equipped with a SPD-16 dual-wavelength ultraviolet detector. The test results showed that the yield of Dammarenediol-II was 1 g/L, that of Protopanaxadiol was 273.8 mg/L, and that of Rare ginsenoside Rh1 was 170.8 mg/L.
The yield of Dammarenediol-II
The yield of Protopanaxadiol
The yield of Rare ginsenoside Rh1
Learn
Based on the experimental results, after integrating the Rh1 synthesis gene into the genome of yeast cells, a yeast strain capable of producing Rh1 was successfully obtained. We plan to further explore whether the introduction of the red algae hydrolase gene into yeast cells can utilize red algae to produce rare ginsenosides, and further investigate whether the modification of the MVA pathway remains effective after constructing a complete metabolic pathway from red algae to Rh1 in yeast cells.
Cycle 7: Using red algae polysaccharides as the substrate to produce the rare ginsenoside Rh1
Cycle 7.2: A recombinant Saccharomyces cerevisiae strain S.cerevisiae Rh1-con with stable accumulation of rare ginsenoside Rh1
Design
We used the lithium acetate conversion method to introduce the plasmid p426-AqAga-agaNash into the engineered strain S.cerevisiae Rh1-con to construct the engineered strain S.cerevisiae Rh1-Ag. Under the optimal mixed carbon source conditions, By comparing the Rh1 yield with that of the engineered strain S.cerevisiae Rh1-Con, it was verified whether the introduction of the red algae hydrolase gene into yeast cells could produce rare ginsenosides from red algae. We designed and constructed the strain S.cerevisiae SC0Rh1-Ag, which contains red algae hydrolase and integrates the Rh1 synthesis pathway but has not been modified by the MVA pathway. Under the optimal mixed carbon source conditions, the Rh1 yield was compared with that of the engineered strain S.cerevisiae Rh1-Ag To verify whether the modification of the MVA pathway remains effective after a complete metabolic pathway from red algae to Rh1 is constructed in yeast cells.
Build
Firstly, using the same method, five exogenous genes of the Rh1 synthesis pathway were integrated into the corresponding genome of wild-type Saccharomyces cerevisiae CEN PK2-1D by CRISPR-Cas9 technology to obtain the engineered strain S.cerevisiae SC0Rh1. Then, the plasmid p426-AqAga-agaNash was introduced into the engineered strains S.cerevisiae SC0Rh1 and S.cerevisiae Rh1-con by lithium acetate conversion method. The engineered strains S.cerevisiae SC0Rh1-Ag and S.cerevisiae Rh1-Ag were obtained.
The engineered strains S.cerevisiae SC0Rh1-Ag
The engineered strains S.cerevisiae Rh1-Ag
Test
The YPDA medium containing 10g/L glucose and 25g/L agar was shaken and fermented for 144 hours at 30℃ and 220 rpm. Samples were taken at regular intervals of 0, 12, 24, 48, 72, 96, 120 and 144 hours after inoculation to detect sugar consumption, bacterial growth and the yield of rare ginsenoside Rh1. The test results showed that compared with the control strain Rh1-con, the Rh1 yield of Rh1-Ag significantly increased in the later stage of fermentation (72-144 hours), reaching the maximum yield of 141.78 mg/L at 144 hours. The engineered strain S.cerevisiae SC0Rh1-Ag completely failed to detect the rare ginsenoside Rh1.
Liquid phase detection image of rare ginsenoside Rh1
A comparison chart of Rh1 yield between strains Rh1-con and Rh1-Ag
Learn
The test results show that the engineered strain can continuously utilize red algae polysaccharides to synthesize the rare ginsenoside Rh1. The synergistic utilization of red algae polysaccharides and glucose in the culture medium optimizes carbon flow distribution: glucose preferentially supports the growth and primary metabolism of the bacteria, while the degradation products of red algae polysaccharides drive the synthesis of Rh1. It has also been demonstrated that tHMG1 and IDI1 enzymes can enhance the MVA pathway in the yeast substrate and balance IPP/DMAPP, thereby increasing the yield of products such as cycloalene and downstream terpene rare ginsenoside Rh1.
Cycle 7: Using red algae polysaccharides as the substrate to produce the rare ginsenoside Rh1
Cycle 7.3: The recombinant strain was used to synthesize the rare ginsenoside Rh1 with red algae polysaccharides as the substrate
[1] Lin X Z, Tang Y, Zhang Y M, et al. Composition and structural spectral analysis of seaweed polysaccharides[J]. Chemistry Bulletin, 2005, (12): 33-39. [in Chinese]
[2] Zhou F, Tan H H, Sun H M, et al. Research progress on biotransformation based on red algae polysaccharides[J]. Food Science, 2021, 42(13): 326-334. [in Chinese]
[3] Aza P ,Molpeceres G ,Salas D F , et al.Design of an improved universal signal peptide based on the α-factor mating secretion signal for enzyme production in yeast[J].Cellular and Molecular Life Sciences,2021,78(7):1-17.
[4] Chulhong O ,Chamilani N ,Youngdeuk L , et al.Cloning, purification and biochemical characterization of beta agarase from the marine bacterium Pseudoalteromonas sp. AG4.[J].Journal of industrial microbiology & biotechnology,2010,37(5):483-94.
[5] Bokun L ,Guoyong L ,Shengkang L , et al.Draft genome sequence of the novel agarolytic bacterium Aquimarina agarilytica ZC1.[J].Journal of bacteriology,2012,194(10):2769.
[6] Bokun L ,Yan L ,Guoyong L , et al.An agarase of glycoside hydrolase family 16 from marine bacterium Aquimarina agarilytica ZC1.[J].FEMS microbiology letters,2017,364(4):fnx012.
[7] Aik-Hong T ,Hafizah N F ,Go F .Crystal structure of a neoagarobiose-producing GH16 family β-agarase from Persicobacter sp. CCB-QB2.[J].Applied microbiology and biotechnology,2020,104(2):633-641.
[8] Xu B R. Cloning, expression and enzymatic properties of thermostable chitinases from Cellvibrio sp. 79 and Cellvibrio sp. OA-2007[D]. Nanjing Agricultural University, 2020. [in Chinese]
[9] Jiangnan University. A Saccharomyces cerevisiae strain for squalene synthesis and its application: CN202110244286.8[P]. 2023-06-13. [in Chinese]
[10] Chikara O ,Masayoshi M ,Shusei O , et al.Production of geranylgeraniol on overexpression of a prenyl diphosphate synthase fusion gene in Saccharomyces cerevisiae.[J].Applied microbiology and biotechnology,2010,87(4):1327-34.
[11] Kenro T ,Masayoshi M ,Chikara O , et al.Overproduction of geranylgeraniol by metabolically engineered Saccharomyces cerevisiae.[J].Applied and environmental microbiology,2009,75(17):5536-43.
[12] Reddy V P ,Jayadeva P B ,Sharad B , et al.Systems biology of GAL regulon in Saccharomyces cerevisiae.[J].Wiley interdisciplinary reviews. Systems biology and medicine,2010,2(1):98-106.
[13] Yina S ,Lei G ,Juan Z , et al.[Effects of mixed carbon sources on glucose oxidase production by recombinant Pichia pastoris].[J].Sheng wu gong cheng xue bao = Chinese journal of biotechnology,2013,29(7):927-36.
[14] John B ,S H A .Promoter engineering: recent advances in controlling transcription at the most fundamental level.[J].Biotechnology journal,2013,8(1):46-58.
[15] Gao T, Ren Y, Li S, Lu X, Lei H. Immune response induced by oral administration with a Saccharomyces cerevisiae-based SARS-CoV-2 vaccine in mice. Microb Cell Fact. 2021 May 5;20(1):95.
[16] Bingyin P ,Chandra N B ,Zeyu L , et al.Engineering eukaryote-like regulatory circuits to expand artificial control mechanisms for metabolic engineering in Saccharomyces cerevisiae[J].Communications Biology,2022,5(1):135-135.
[17] Mi-Jin K ,Hyun B S ,Hyun-Joo P , et al.A new platform host for strong expression under GAL promoters without inducer in Saccharomyces cerevisiae[J].Biotechnology Reports,2022,36e00763-e00763.
[18] Miao B, Gou M, Chen D, et al. Evaluation of constitutive and inducible promoters of Saccharomyces cerevisiae under different sugar fermentation conditions[J]. Chinese Journal of Applied and Environmental Biology, 2019, 25(05): 1185-1191. [in Chinese]
[19] Peng B Y, Chen X, Shen Y, et al. Differential expression of xylulokinase under the control of different promoters and its effect on xylose metabolism in Saccharomyces cerevisiae[J]. Acta Microbiologica Sinica, 2011, 51(07): 914-922. [in Chinese]
[20] S M R ,Celia P ,F A R , et al.Comprehensive Analysis of the SUL1 Promoter of Saccharomyces cerevisiae.[J].Genetics,2016,203(1):191-202.
[21] Wang J J. Study on the expression of Saccharomyces cerevisiae glycerol synthesis key enzyme gene GPD1 and its promoter in tobacco[D]. Jiangnan University, 2008. [in Chinese]
[22] Wang J, Zhai H, Rexida R, Shen Y, Hou J, Bao X. Developing synthetic hybrid promoters to increase constitutive or diauxic shift-induced expression in Saccharomyces cerevisiae. FEMS Yeast Res. 2018 Dec 1;18(8).
[23] Jin L, Nawab S, Xia M, Ma X, Huo YX. Context-dependency of synthetic minimal promoters in driving gene expression: a case study. Microb Biotechnol. 2019 Nov;12(6):1476-1486.
[24] Kotopka, B. J., & Smolke, C. D. (2020). Model-driven generation of artificial yeast promoters. Nature Communications, 11(1), 2113.
[25] Rui L ,Lanqing L ,Xia L , et al.Engineering yeast artificial core promoter with designated base motifs.[J].Microbial cell factories,2020,19(1):38.
[26] LiS Y,Ma LZ,Fu WX,et al. Programmable syntheticlibraryfor fine-tuningupstream activatingsequencegeneexpression levels in Saccharomyces cerevisiae. AS SyntheticBiology,2022,11(3):1228-1239.
[27] S K S ,W M M ,H W Z V , et al.Promoter-proximal introns impact recombinant amylase expression in Saccharomyces cerevisiae.[J].FEMS yeast research,23(4), foad047.
[28] Zhang H. Analysis of the biosynthetic pathway of oleanane-type ginsenosides and establishment of their heterologous synthesis system[D]. Northeast Forestry University, 2022. [in Chinese]
[29] Shi Y S, Wang D, Li R S, et al. Construction of Saccharomyces cerevisiae cell factory for fermentative production of ginsenoside Rh2[J]. China Journal of Chinese Materia Medica, 2022, 47(03): 651-658. [in Chinese]
[30] Tan W ,Tong S ,Gao Y , et al.Rational spatial rewiring of key enzymes enhances α-santalene production in Saccharomyces cerevisiae.[J].Bioresource technology,2025,436133027.
[31] Shimizu, T., Kobayashi, T., Ba-Thein, W., Ohtani, K., & Hayashi, H. (1995). Sequence Analysis of Flanking Regions of the pfoA Gene of Clostridium perfringens: β- Galactosidase Gene (pbg) Is Located in the 31-Flanking Region. Microbiology and Immunology, 39(9), 677 - 686.
[32] Ariga, O., Okamoto, N., Harimoto, N., & Nakasaki, K. (2014). Purification and Characterization of α-Neoagarooligosaccharide Hydrolase from Cellvibrio sp. OA-2007. Journal of Microbiology and Biotechnology, 24(1), 48–51.
References: