Results

The hypothesis of our study is that polyamines can enhance the thermo-tolerance of S. cerevisiae. Therefore, we initially conducted exogenous spermidine addition experiments to preliminarily investigate the feasibility of our hypothesis.

We established different experimental groups with final spermidine concentrations ranging from 1nM to 1mM. After a 48-hour culturing at 38°C, we measured the growth of the wild-type strain (Figure 1).

The results indicated that spermidine did not significantly enhance the thermo-tolerance of the strain. This could be attributed to several potential factors. We have summarized the two primary possibilities: Firstly, if our hypothesis is correct, it is plausible that spermidine may not be able to enter the cell and thus fails to exert its effects. Moreover, extracellular spermidine may not impact cell growth. Alternatively, our hypothesis may not hold true, implying that spermidine cannot enhance the cell's thermo-tolerance capabilities.

According to literature reports, the presence of the endogenous OAZ1 gene in S. cerevisiae restricts the synthesis of polyamines, thereby limiting the direct precursor spermidine's production. To enhance the metabolic flux of spermidine synthesis in our strain, we initiated the knockout experiment of the endogenous OAZ1 gene. Our experiment consisted of three main phases: fragment construction, yeast transformation, and validation.

We obtained the plasmid backbone through PCR amplification, as shown in Figure 2A, and its size was as expected. Subsequently, we annealed to obtain the gRNA sequence. Following the extraction of yeast genomic DNA, we amplified the upstream and downstream sequences of the repair template by PCR, as depicted in Figure 2B, and their sizes were as expected. Finally, we generated the complete repair template through fusion-PCR, as illustrated in Figure 2C, and its size was as expected.

Following the acquisition of the aforementioned fragments, we proceeded with yeast competent cell preparation and transformation onto Delft agar plates. Subsequently, we randomly selected 10 individual clones from the plates for further expansion culture and genomic DNA extraction for preliminary validation. The length of PCR at the original locus was 1949bp, as shown in the Positive Control (PC) group in Figure 3A. The theoretically expected length after successful knockout should be 1011bp. Gel electrophoresis results indicated that colony 1-5 might be yeast strains with the OAZ1 gene knockout, as shown in Figure 3A. Therefore, we sequenced the PCR products from these colonies, and the results aligned with our expectations (Figure 3B and 3C).

SPE1 encodes an ornithine decarboxylase, catalyzing the decarboxylation of ornithine to form putrescine, and its activity is inhibited by the antizyme enzyme OAZ1. To enhance putrescine expression and consequently increase the supply of spermidine precursors, we constructed yeast strains with OAZ1 knockout and SPE1 knock-in for putrescine (and subsequently spermidine) overexpression. The basic experimental procedure is similar to the one described above.

After extracting yeast genomic DNA, we amplified the upstream and downstream sequences of the repair template by PCR, as well as the TEF1p, PRM9t (Figure 4A) and SPE1 (Figure 4B) fragments, and their sizes were as expected. Subsequently, we successfully prepared the repair template using a two-round fusion-PCR strategy. In the first round, we fused the first two fragments (upstream-2 and TEF1p) and the last two fragments (PRM9t and downstream-2), as shown in Figure 4C. In the second round, we fused three fragments (upstream-2-TEF1p, SPE1, and PRM9t-downstream-2), as shown in Figure 4D.

Then we tried to transform the fragments and validate the colonies. The length at the original locus was 1949bp, as shown in the Positive Control (PC) group in Figure 5A. The theoretically expected length after successful editing should be 3080bp. Gel electrophoresis results indicated that Colony2-1, Colony2-3, Colony2-4, and Colony2-7 might be successfully edited strains (Figure 5A). Therefore, we selected the PCR product from Colony2-1 for sequencing, and the results matched our expectations (Figure 5B).

Then we employed 5-FOA for counter-selection to obtain strains that had lost the gRNA editing plasmid.

And we performed HPLC test to ensure that our engineered strains have improved production of putrescine and/or spermidine. As shown in Figure 6, the OAZ1Δ strain produces 5.70 mg/L putrescine and 7.15 mg/L spermidine, and the OAZ1::SPE1 strain produces 9.5 mg/L putrescine and 15.2 mg/L spermidine. Compared to wild type strain produces 1.3 mg/L putrescine and 2.9 mg/L spermidine, the both engineered strain has a higher production of polyamines, which means that we successfully increased the flux of polyamine synthesis.

Subsequently, we conducted a thermo-tolerance growth test at 35°C for the aforementioned strains, as shown in Figure 7.

The experimental results indicated that the knockout of the OAZ1 gene and the overexpression of SPE1 did not have a significant impact on the thermo-tolerance of the strains.

After reaching this conclusion, our subsequent experiments will focus more on putrescine rather than its precursors (i.e., ornithine and spermidine). The AtACL5 gene, originating from Arabidopsis thaliana, encodes a thermospermine synthase capable of catalyzing the aminopropylation of dSAM to form thermospermine. We plan to construct a repair template with the sequence OAZ1_upstream-TEF1p-SPE1-PRM9t-TDH3p-AtACL5-DIT1t-OAZ1_downstream to simultaneously knockout OAZ1 and knock-in SPE1 and AtACL5 to achieve thermospermine overexpression.

We initially obtained the short fragment (Figure 8A and 8B) through PCR. Subsequently, we amplified the SPE1 fragment (Figure 8C) from the previously used repair template. Following that, we obtained the AtACL5 fragment (Figure 8D) through fusion PCR. Finally, we acquired the complete repair template (Figure 7E) through fusion PCR.

However, numerous small colonies grew on the Delft plates after yeast transformation, as shown in Figure 9. These colonies could not be expanded in Delft media, and as a result, we never obtained positive clones. We suspect that this may be due to low transformation efficiency leading to a low positivity rate. Additionally, the use of uracil-containing medium during the preparation of competent cells and post-transformation recovery may have caused some cells that were not successfully transformed to maintain uracil for a period, allowing them to grow and divide, thereby leading to false positives.

Subsequently, we improved the transformation process to enhance its efficiency, as detailed in Engineering. Additionally, when fusing multiple short fragments into longer ones, we refined and optimized the details of fusion PCR, significantly increasing its efficiency, as elaborated in Engineering.

Using the OAZ1 knockout strain we previously constructed, we generated yeast expression plasmids to overexpress SPE1, AtACL5, as well as both SPE1 and AtACL5, to achieve overexpression of putrescine (and subsequently spermidine) and thermospermine. We initially obtained the plasmid backbone through PCR (Figure 10: 7-1 to 7-4) and the insert fragments (Figure 10: 7-5 to 7-7; 7-5 is TEF1p-SPE1-PRM9t, 7-6 is TDH3p-AtACL5-DIT1t, 7-7 is TEF1p-SPE1-PRM9t-TDH3p-AtACL5-DIT1t). Subsequently, we used homologous recombination to construct the complete plasmids and validated their successful construction through sequencing (Figure 12).

And we performed HPLC test to ensure that our engineered strains have improved production of putrescine, spermidine and thermospermine. As shown in figure 12, the strain contains plasmid that overexpress SPE1 has a higher production of putrescine and spermidine, namely 6.95 mg/L and 8.1 mg/L, respectively, compared to 5.00 mg/L and 6.7 mg/L in strain contains plasmid vector. And the strain contains plasmid that overexpress SPE1 and AtACL5 produces 5.2 mg/L spermidine, less than negative control group. As we didn’t get the standard of thermospermine, we didn’t get a directly evidence of thermospermine prodution. However, according to the decreased spermidine, we think it may indicated the improvement production of thermospermine.

Then we tested the growth of these strains under 35°C conditions (Figure 13). The results indicate that the expression of the AtACL5 gene did not have a significant impact on the yeast's thermo-tolerance capabilities.

Furthermore, after conducting additional literature research, we discovered that the expression of thermospermine not only depends on the level of putrescine but also on the level of dSAM (decarboxylated S-adenosylmethionine). And then we drew the detailed picture of polyamine synthesis in yeast(Figure 14). We hypothesized that even though we overexpressed the genes for putrescine and thermospermine, the insufficient intracellular dSAM level might have resulted in very low thermospermine expression. Additionally, the strong competitive nature of spermidine and thermospermine synthesis for the same precursors further reduced thermospermine levels.

Therefore, we plan to construct strains with simultaneous overexpression of dSAM to increase the levels of thermospermine. We designed a plasmid library to stably express SPE2 and tune the expression level of SPE3 and AtACL5 to fully use the spermidine substrate and examine the numerical relationship between spermidine and thermospermine.(Figure 15)

We firstly achieved all short single fragment by PCR from the genome of S. cerevisiae and plasmid(Figure 16). Then we connected all single expression device by fusion PCR(Figure 17).

However, the Gibson coloning results were not as expected. We performed the double enzyme digestion experiment and PCR to check whether the sequence of whole plasmids are right. And only one colony shows expected results(Figure 18), we guessed it may come from the unexpected similarity of homologous fragment. Due to time limited, we didn’t complete this part of work. However, we will keep on trying this design.

Though this season is end, our project keeps going on. We have many ideas about this project.

Firstly, the endogenous metabolism is important for any organism, which means any alteration should consider the negative effect on strain itself. In our project, we overlooked the flux from ornithine to arginine, and this may cause a decreased arginine production, which is harmful to our engineered strain. So, we plan to increase the production of ornithine to provided enough substrate for polyamine synthesis.

Secondly, not only thermospermine, as the information we collected from different articles, there are many kinds of long-chain polyamines and their derivatives may have an important effect on the thermo-tolerance of yeast strain. These unusual polyamines are our main targets for future.

Thirdly, polyamines play an important role in different kinds of stress-tolerance. We will use our strain to test whether it has other stress-tolerance, such as the tolerance to extreme pH, toxic product and so on.