Team:SCU-China/Design
PROJECT
DESIGN
Why Saccharomyces cerevisiae
As is provided in background information, we engineered yeast and created a delayed expression system to produce cordycepin automatically in factories at a cheaper price.
We have chosen Saccharomyces cerevisiae as the chassis for fermentation this year. According to FDA, S. cerevisiae is safe enough for fermentation and has been widely used in food and drug factory. And the enzymes worked for cordycepin synthesis are only found in fungi, indicating that cordycepin synthesis is more compatible to eukaryotic cells.
Synthesis of cordycepin
Cordycepin (3’-deoxyadenosine, COR) is naturally synthesized by Cns1 and Cns2 in Cordyceps militaris. The precursor, adenosine-3’-monophosphate (3’-AMP) is catalyzed by Cns2 and dephosphorylated to 2’-carbonyl-3’-deoxyadenosine (2’-C-3’-dA), and then converted to COR by Cns1. Then COR will be deaminated by adenosine deaminase (ADA), while this process can be inhibited by the ADA inhibitor like pentostatin (Figure. 1). (Xia et al., 2017)
Figure. 1 The pathway of COR synthesis
By using endogenous adenosine as the substrate, which can be converted to 3’-AMP in fungi, we transformed codon-optimized cns1 and cns2 into yeast to produce COR in our engineered yeast.
Because these two genes were newly found in C. militaris, we synthesized them with the sequence on NCBI and constructed the shuttle vectors pYES2-Cns1 and pYES2-Cns2 to verify their function. However, considering the possibility of losing plasmids during fermentation and that the intermediate product, 2’-C-3’-dA is hard to detect, we later constructed the plasmid pYES2-Cns1-Cns2 to express Cns1 and Cns2 separately.
After reviewing the paper, we were informed that Cns1 and Cns2 might work adjacently in vivo and fusion expression could increase the production of COR, which is consistent with the simulation in our modeling. (Figure. 2) Therefore, we removed the terminator of cns1 and promoter of cns2 in pYES2-Cns1-Cns2 and added some amino acid residues as a linker. We searched the articles and found some potential linkers. With the help of computational biological methods in modeling, we chose "SEAAAREAAAREAAAREAAAR" as the linker. As a result, we got the pYES2-Cns1-linker-Cns2 to express the fusion protein of Cns1-linker-Cns2.
Figure. 2 The fusion expression of Cns1 and Cns2 can increase COR production.
After validating the delayed expression system, we transformed this fusion gene in yeast and enabled COR expression at plateau stage with the protection of pentostatin during fermentation.
Synthesis of pentostatin
Pentostatin (2’-deoxycoformycin, PTN), an adenosine analog, is a chemotherapeutic drug used in the treatment for several forms of leukemia. Recently, a study of COR production in the fungi C. militaris reported that biosynthesis of COR is coupled with PTN production by a single gene cluster. We analyzed their results and concluded that the production of PTN is important for COR high production. Because they demonstrated that PTN can protect the stability of COR from deamination by inhibiting ADA activity (Xia et al., 2017). Taken together, the COR/PTN coupled biosynthesis strategy can extend to our eukaryotic COR producers to increase the production of COR.
The expression of Cns3 in yeast
To produce PTN, we considered cns1–cns3 gene cluster identified in C. militaris, which encodes enzymes involved in COR/PTN coupled biosynthesis. As we mentioned above, cns1 and cns2 genes are indispensable for COR biosynthesis. It is also confirmed that cns3 gene alone could enable some fungi (but no yeast) to produce PTN consuming adenosine. (Figure. 3) (Xia et al., 2017) This PTN biosynthetic pathway is simpler than the one in bacterium requiring at least three enzymes and is more compatible with yeast cells. Therefore, we initially intended to transform codon-optimized Cns3 to S. cerevisiae to produce PTN together with COR. Our laboratory work then prove that we are the first to realize successful Cns3 function in yeast.
Figure. 3 The pathway of PTN synthesis
The HisG Domain of Cns3
Subsequently, we communicated with the discoverer of COR in C. militaris, prof. Chengshu Wang in the Shanghai Institute of Plant Physiology and Ecology. Through communication, we learned that there are N-terminal nucleoside/nucleotide kinase (NK) and C-terminal HisG domains in Cns3 (Figure. 4). Consistent with their results, it was also confirmed that the HisG domain of Cns3 is indispensable for PTN production, while the function of NK domain of Cns3 is unclear (More details about this communication).Yeast cells transformed with either cns1–cns3 or cns1–cns2 can produce COR, but the expression of Cns3 in yeast is unclear (Xia et al., 2017). So, based on the information above, we decided to learn more about HisG domain of Cns3 instead of NK domain. A single HisG domain-containing enzyme Cns3 (called MF) or only a HisG domain (called ONLY) was introduced into S. cerevisiae to simplify the COR/PTN co-production system.
Figure. 4 The expression of HisG domain-containing Cns3 simplify the COR/PTN co-production system.
The Constitutive Promoters of Cns3
As delay expression was later designed in our project, we replaced the original inducible promoter that controls Cns3 expression with constitutive promoters. The constitutive expression of Cns3 can continuously produce PTN, which can accumulate enough PTN to protect the stability of COR by inhibiting ADA activity in yeast. We also tested constitutive promoters of different transcription levels to prevent engineered yeast from disrupted cell growth due to product toxicity, as well as the metabolic burden imposed by the redirection of carbon flux, redox cofactors, and ATP (Figure. 5) (Peng, Williams, Henry, Nielsen, & Vickers, 2015). The utilization of a promoter with appropriate strength ensures normal cell growth and protects the COR while avoiding the possible substrate competition between the biosynthesis of COR and PTN.
Figure. 5 PTN production controlled by appropriate strength constitutive promoters ensures normal cell growth and protects the COR while avoiding the possible substrate competition.
Delay expression system
Many fermentative processes need inducers to induce the microorganism to compound production during the fermentation. This is complicated and may cause some contamination in the fermentation system. If we add inducer at the beginning of fermentation, the inducer will induce microorganism synthesis heterologous protein instead of homologous protein, a phenomenon harmful to microorganism growth. Such drawbacks can lead to yeast cells’ inability to proliferate in quantity, thus an undesired productivity.
To solve these problems with a fermentative inducer, we designed the delay expression system to achieve automatic co-fermentation of COR and PTN. In this system, we can control the yeast to express the enzymes important for COR and PTN synthesis. The cells can proliferate to a suitable level, without adding inducer during the process.
The pGAL1 delay expression system
GAL4 system is opportune for us to construct the delay expression system. As a vital transcriptional factor in S. cerevisiae, GAL4 can activate the GAL1 promoter (pGAL1) to express downstream genes with other factors. GAL80 is an inhibitor of GAL4, which combines with GAL4 to prevent GAL4 from activating pGAL1. And galactose can interact with GAL80, disabling GAL80 to repress GAL4 activation, then GAL4 can activate pGAL1 with other factors (Figure. 6). But glucose is the optimal intake carbon resource for most yeasts. Galactose only can repress GAL80 and activate GAL4 indirectly, when there isn’t glucose in the medium (Traven, Jelicic, & Sopta, 2006).
Figure. 6 The mechanism of GAL4 and GAL80 activating transcription
At first, we constructed the pGAL1 delay expression system in S. cerevisiae BY4741, a strain concluding the original GAL4 and GAL80 gene in the genome. In this circuit, Cns1-linker-Cns2 coding sequence was inserted into the vector pYES2-NTA behind the pGAL1 (Figure. 7). With the constitutive expression of GAL4 and GAL80 in BY4741, we can control whenever we want the pGAL1 downstream to be expressed by changing glucose/galactose concentration in the medium. At the beginning of fermentation, yeasts metabolize glucose as carbon resource to grow. When yeasts proliferate in quantity, glucose is depleted in medium. Later on, galactose becomes the only carbon resource accessible and induces the yeasts to express Cns1 and Cns2 to synthesis COR (Figure. 8).
Figure. 7 The circuit of the pGAL1 delay expression system
Figure. 8 The mechanism of pGAL1 delay expression system
However, there are still some limitations to the pGAL1 delay expression. For example, we can’t delay the expression of Cns1 and Cns2 for a long time by adding a good deal of glucose, because high osmotic pressure may result in cell death. And galactose is more expensive than glucose, so it is not a desirable material in manufacturing.
The pMET3 delay expression system
To make the delay expression more flexible, we design pMET3 delay expression system. MET3 promoter (pMET3) is a strictly regulated promoter in methionine metabolism pathway of S. cerevisiae. The pMET3 regulates MET3 gene expression, an enzyme in the methionine biosynthetic pathway, and can be severely inhibited by a methionine concentration over 100 μM (Mao, Hu, Liang, & Lu, 2002). We can only add a small quantity of methionine to inhibit the pMET3. Therefore, it is suitable for regulating the transcriptional factor of the delay expression system.
The design of pMET3 delay expression system are shown as follows (Figure. 9). At the beginning, the concentration of methionine is high in the medium, so S. cerevisiae cannot express Cns1-linker-Cns2 to synthesize COR. After growth plateau, S. cerevisiae metabolize much methionine, so there is lower methionine concentration in the medium. It is worth noting that S. cerevisiae can express Cns1-linker-Cns2.
For the pMET3 delay expression system, we choose S. cerevisiae YM4271 as the chassis thanks to the advice from professor Liu Ke. GAL4 and GAL80 are knocked out in YM4271, so we can use GAL4 as a heterologous transcription factor and we do not have to use galactose to repress GAL80 and activate GAL4 indirectly. YM4271 can compound methionine by itself, so low concentration of methionine in the medium has no influence on YM4271 when the pMET3 is on operation (Reece-Hoyes, & Walhout, 2018).
Figure. 9 The mechanism of pMET3 delay expression system.
The integrated delay expression system
Finally, we combined the pGAL1 delay expression system with the pMET3 delay expression system to construct a better integrated delay expression system in S. cerevisiae YM4271.
The circuit of integrated delay expression system is pMET3-GAL4-pGAL1-Cns1-linker-Cns2 and we cloned this circuit onto the vector pYES2-NTA (Figure. 9). At the beginning of fermentation, we add a certain concentration of methionine into medium to inhibit pMET3 from transcribing downstream GAL4. As methionine is constantly consumed in the medium, the number of yeasts reaches a high level, while the yeasts also accumulate enough PTN to protect the stability of COR, which is suitable for expressing the enzymes producing COR. Meanwhile, because much methionine is depleted by yeasts, methionine concentration becomes lowered in the medium and its inhibition on pMET3 becomes removed. The yeasts express GAL4 inducing the pGAL1 to express Cns1-linker-Cns2 (Figure. 10). Therefore, we can delay the expression of Cns1-linker-Cns2 for a longer time with the integrated delay expression system. Moreover, we don’t need galactose as fermentative material. This system can also be applied to delay the expression of other heterologous proteins in yeast cells.
Figure. 10 The circuit of the integrated delay expression system
Figure. 11 The mechanism of integrated delay expression system
Modeling of delay system provides a simulation on delayed time before experimental measurements are made. Fermentation time from 10 to 360 hours with methionine concentraton from 1mM to 379mM (methionine max concentration in theory) were included in the simulation, indicating full flexibility.
Conclusion
This video describes the whole process of CORegulaTIN. It will help you to understand the project more easily.
Reference
Mao, X., Hu, Y., Liang, C., & Lu, C. (2002). MET3 Promoter: A Tightly Regulated Promoter and Its Application in Construction of Conditional Lethal Strain. Current Microbiology, 45(1), 37–40.
Peng, B., Williams, T. C., Henry, M., Nielsen, L. K., & Vickers, C. E. (2015). Controlling heterologous gene expression in yeast cell factories on different carbon substrates and across the diauxic shift: a comparison of yeast promoter activities. Microbial cell factories, 14, 91.
Reece-Hoyes, J. S., & Walhout, A. (2018). High-Efficiency Yeast Transformation. Program in Systems Biology Publications and Presentations. 141.
Xia, Y., Luo, F., Shang, Y., Chen, P., Lu, Y., & Wang, C. (2017). Fungal cordycepin biosynthesis is coupled with the production of the safeguard molecule pentostatin. Cell Chemical Biology, S2451945617303276.