Results

Results

---

Par I

Through literature and database search, we chose enzymes that could break down cellulose in pinewood, such as cellobiohydrolase I (KsCel7A) from Kocuria palustris, cellobiohydrolase II (TfCel6B) from Thermobifida fusca, synthesized based on ancestral endocellulase (LFCA_EG - Cel5A), and beta-glucosidase (CtBglA) from Clostridium thermocellum. These enzymes have high activity levels and can be epxressed from E. coli and B. subtilis. We predicted the tertiary structure of these enzymes using Alphafold. We found that in all proteins except CtBglA, the catalytic part and the carbohydrate-binding module formed an intact structure. Also, CtBglA could exist in a soluble form to effectively decompose cellobiose released from cellulose by CBHI, II, and LFCA-EG.

We tested several types of promoters to express the cellulose-degrading enzyme in Bacillus subtilis and Escherichia coli. Through literature review, we selected four constitutive promoters and two inducible promoters known to be strong in Bacillus. To confirm that these promoters are working properly, mScarlet was used as a reporter. As shown in Figure 2, trnQ showed the strongest transcriptional activity. In addition, transcription by the grac100 promoter was inhibited by Lacl.

For mass production of enzymes, we checked whether the IPTG inducible system in bacillus works normally. The bacillus having the plasmid DNA shown in Figure 3 was spread on an LB agar plate with or without IPTG. As a result, it was confirmed that the expression of the reporter gene was high in the condition with IPTG. These results show that our IPTG inducible system works normally in Bacillus subtilis.

On the other hand, the results were different in E. coli. The transcriptional activity of the P2069M promoter was higher than that of the trnQ promoter. As in Bacillus, the grac100 promoter was effectively inhibited by Lacl, but IPTG did not induce the expression of the reporter gene. Therefore, we used the previously known lac promoter of E. coli for production of cellulose degrading enzymes.

We tried to decompose cellulose by directly treating the media containing the enzymes on wood by simplifying the complex process of 'expression, purification, and decomposition' of the cellulose degrading enzyme and 'expression and secretion'. To this end, it was attempted to find a secretion signal through which each enzyme could be effectively secreted. A signal peptide library was prepared by cutting sp-library plasmid containing 12 secretion signals with NdeI and NcoI. The coding sequence of each enzyme was cut and purified with NcoI and XhoI, and PtrnQ and P2069M plasmids were prepared by cutting with NcoI and XhoI. Four ligation samples using PtrnQ as a backbone were transformed into Bacillus subtilis 1A976, and four recombinant DNAs using P2069M plasmid as a vector were transformed into E. coli Mach1. Transformants were spread on agar medium containing chloramphenicol and ampicillin, respectively. Each of 24 colonies was inoculated in a 96 well plate and incubated at 37°C for 12 hours. After centrifugation, the supernatant was spotted on a nitrocellulose membrane. After immunoblotting with an anti-his tag antibody, two clones showing a strong signal were selected and their plasmid DNA was isolated. Then, the secretion signal was identified through sequencing.

The secretion of KsCeli7A by epr and ywsB was the highest. Bacillus was most effectively secreted by bglC and ywsB.

In the case of KsCel7A, E. coli was secreted the highest by epr and ywsB. Bacillus was effectively secreted by bglC and ywsB.

LFCA_EG was enhanced by pel in both Escherichia coli and Bacillus subtilis, and also helped to secrete bglC and aprE.

Lastly, for CtBglA, phoB was the most frequent, and ywsB was observed in the rest.

The combination of 16 secretion signals and enzymes found previously was subcloned using the IPTG inducible system. These plasmids were transformed into Bacillus 1A976 and E. coli BL21 and filtered with a 0.2um CA syringe filter after induction with IPTG. Each sample was separated by size through SDS-PAGE. The resolved protein was transferred to a 0.45um PVDF membrane and then immunoblotted with an anti-his tag antibody. Finally, the blotted membrane was stained with amido black.

It was found that SP_ywsB-TfCel6B and SP_bglC-LFCA_EG were effectively expressed and secreted in BL21. Although the overall expression level was low in 1A976, SP_phoB-CtBglA and SP_ywsB-KsCel7A were stably expressed.

The activity of LFCA_EG was confirmed further in the CMC agar plate. It was observed that LFCA_EG secreted from both BL21 and 1A976 effectively decomposed CMC. However, according to the previous SDS-PAGE results, LFCA_EG proteins of the desired size were more present in BL21. Therefore, we used SP_bglC-LFCA_EG produced from BL21.Ultimately, we combined SP_ywsB-TfCel6B and SP_bglC-LFCA_EG produced in BL21 and SP_ywsB-KsCel7A produced in 1A976 to break down the cellulose.

We overexpressed the aforementioned plasmid in the E.coli BL21 and B. subtilis 1A976 for 12 hours at 30 degrees in the 0.4mM IPTG condition. After centrifugation, we obtained a supernatant containing the enzyme. Without purification, we mixed the combination mentioned in the figure 11, treated it with cellulose, and measured the amount of reducing sugar through the DNSA assay. As a result, KsCel7A failed to contribute much to the breakdown of cellulose. We thought this was due to the small amount of protein secreted. However, treating the 4 enzymes together resulted in the greatest amount of cellulose broken down. Therefore, we decided to use a mixture of the four enzymes

Additionally, we observed the degree of degradation over time to determine how long our cellulose degrading enzyme cocktail functions. As time passed, the activity of the enzyme gradually decreased, but cellulose was maximally degraded between 18 and 24 hours after the enzyme treatment. After 24 hours, the amount of glucose in the solution starts to decrease, which is thought to be the effect of the growth of bacteria.

Future Work

Our team has confirmed that it is possible to produce oil adsorbents by effectively decomposing cellulose in wood through experiments so far. However, the low expression and secretion of KsCel7A with cellobiohydrolase I activity is a problem to be solved. So, we cloned the coding sequence of KsCel7A into the pPICZalpha plasmid to confirm that a higher amount of protein can be obtained from KsCel7A in Pichia pastoris. In addition, in order to easily isolate positive clones from Pichia, T2A-self-cleaving peptide and sfGFP were fused behind the coding sequence of KsCel7A. When the KsCel7-T2A-sfGFP fusion construct is transcribed and translated by the AOX1 promoter to produce a long peptide, it is cleaved into KsCel7A and sfGFP by T2A. Therefore, positive clones can be easily distinguished by the presence or absence of fluorescence. In Figure 13, the plasmid was electroporated to pichia after linearization with PmeI. 20 colonies were patched on a minimal methanol plate containing methanol. As a result, two clones showing high fluorescence intensity were obtained. We plan to measure secretion and enzyme activity of KsCel7A by inducing overexpression of KsCel7A in Pichia.

Part II

Cellulose degradation in pine tree

For the effective elimination of cellulose from pine woods, we pulverized the wood waste to make it easier for enzymes to react. After decomposing pine tree powder using four types of enzymes for twelve weeks through our self-made reaction machine, we observed that the color of the powder turned darker brown compared to that before the reaction. Although there was a slight difference in color and brightness depending on the types of pine tree powder samples, the overall trend shows that the color of the powder got darker throughout the twelve weeks of reaction.

Contact Angle Analysis

In our pilot experiment, we observed the increase of hydrophobicity of pine wood powder that is decomposed by brown rot fungi when it is heated at 200℃ for 10 minutes. Also, since we know that the degree of hydrophobicity is directly proportional to the lignin composition of the powder, we attempted to increase the amount of lignin by using four types of enzymes that decompose cellulose from pine tree powder. To see if reaction time affects the hydrophobicity level, we collected samples from the week 1, 4, and 8 weeks and measure the contact angle. As a result, the contact angle of heated samples increases as reaction time gets longer. Although we couldn’t collect data from a longer time period due to the number limit of the reaction machine, there wasn’t a noticeable difference between weeks 12 and 15.

Change in Hydrophobicity Duration Time with Regards to Lignin Content

Here, we released five grams of pine wood powder that is heated at 200℃ and reacted with four types of enzymes on the water and measure the time it took for the powder to start sinking down. As we know that the longer reaction time with enzymes increases the lignin composition of the powder, it was reasonable to see that the lignin content of the initial sample increased from 27.5% to 45.6% on through the twelve weeks of reaction. Another interesting aspect of this experiment was that the duration of hydrophobicity also increased as the powder reacted with enzymes longer. In our observation, there was a huge increase in the duration of hydrophobicity after the eight weeks of reactions, and when the twelve weeks of reaction was completed, the sample floated on the water for 26 hours, which is three times as longer than the duration in week 8. While the lignin composition showed linear growth, the duration of hydrophobicity showed exponential growth throughout the twelve weeks of observation. Since the lignin content was 45.6% at the end of the experiment, Wwe concluded that the hydrophobicity of the pine wood powder increases significantly once it gets over 40% of the lignin composition. Therefore, we determined twelve weeks of reaction time to be the most effective time interval to induce hydrophobicity and named the pine tree powder from this reaction as “PineSorb”.

Hydrophobicity & Oleophilicity

To test the hydrophobicity and oleophilicity of PineSorb after twelve weeks of reaction time, we put PineSorb in a circle, flattened the surface using a wooden plank, and reacted with water and different types of solutions. While water contacted with PineSorb and formed circular waterdrops at the surface, other solutions that are relatively non-polar were absorbed extremely well by PineSorb.

After releasing the mixture of diesel oil and Nile Red dye and putting a tea bag with 4g of PineSorb, we can observe that Pinesorb absorbs almost every oil on the water in a short period of time.

After putting a beaker of water on the stirrer and releasing waste engine oil, we put 4g of PineSorb and started the stirrer. Despite the strong water current, the tea bag with PineSorb did not sink down and removes almost every waste engine oil on the water's surface. From this experiment, we can learn that PineSorb sustains its high hydrophobicity while absorbing oil efficiently.

In the water tank that imitates actual environments like oceans or lakes, we filled in water, released waste engine oil, and scattered the PineSorb powder. After a very short period of time, we pushed the powder with a stick and observed how it eliminates oil by forming a thin layer on the surface. Since the powder does not sink even after adsorbing oil, it is possible to completely remove spilled oil from the water.

Conclusion

When we activated four types of enzymes that decompose cellulose at two types of bacteria—Bacillus subtilis and Escherichia coli—and treated pine tree powder, it effectively decomposed cellulose through twelve weeks and increased lignin composition from 25~30% to 39~47%. The amount of time this process has taken is extremely shorter than when a pine tree is decomposed by brown rot fungi in a natural environment. Although we did not directly compare the same exact samples in our experiments, the fact that we increased the hydrophobicity and oleophilicity of pine trees in such a short period of time has enough value and merchantability in the eco-friendly oil adsorbent market. Based on this achievement we made through experiments, we are optimistic that we can even shorten the time it takes for pine wood to be decomposed by utilizing a fungi system and decomposing hemicellulose with cellulose together. Lastly, I believe that this study opened more possibilities for the use of lignin for industrial purposes.