Plastics Concerns in SDGs

This year, our goal is to address the issue of plastic pollution. While biodegradable plastics have garnered interest, challenges arise due to the varying decomposition conditions required for different biodegradable plastic materials. Furthermore, even though plastics embedded with degrading enzymes show promise, most of these enzymes cannot withstand the high temperatures encountered during the plastic thermoforming process. Among biodegradable plastics, only polycaprolactone (PCL) becomes pliable at temperatures below 100°C. However, this near-boiling point still poses a challenge for all enzymes. Using the "engineering design cycle (design-build-test-learn)" framework, we aim to address these challenges in two phases:

PROBLEMS

1. PCL plastic products take too long to decompose, often requiring 2-3 years.
2. PCL-degrading enzymes (PCLase) cannot withstand boiling temperatures.

AIMS

1. Develop a thermostable enzyme-embedded PCL product, using GFP as a preliminary study.
2. Create a thermostable PCLase-embedded PCL product and evaluate its degradation properties.

SpyRing cyclization technique to enhance enzyme thermal resilience was clarified by Dr. Mark Howarth’s team¹. SpyRing harbors genetically modified SpyTag (13 amino acids) on the N-terminus and SpyCatcher (12kDa) on the C-terminus on the protein of interest. This context spontaneously reacts together through an irreversible isopeptide bond. SpyRing cyclization was demonstrated successfully to increase stress resilience of β-lactamase and some industrially important enzymes.

The mechanism for the spontaneous cyclization reaction involves Asp7 on the SpyTag at the N-terminus forming double hydrogen bonds with Glu77 on the SpyCatcher at the C-terminus. This interaction strengthens the creation of an irreversible isopeptide bond between Asp7 on SpyTag and Lys31 on SpyCatcher. As a result, the protein is seamlessly cyclized, enhancing its resistance to chemical, thermal, or enzymatic degradation^2.

Therefore, we designed a T7-promoter-based gene expression plasmid carrying GFP coding region flanked by a N-terminal SpyTag and a C-terminal SpyCatcher. When expressed in IPTG-induced E. coli BL21 cells, this plasmid is expected to produce GFP proteins with SpyRing, which should spontaneously cyclize and exhibit enhanced thermal stability.

We deleted start (ATG) codon and stop (TAA) codon, and incorporated SpyTag at the N-terminus and SpyCatcher at the C-terminus. This DNA fragment was synthesized by Integrated DNA Technologies, Inc. (IDT) and then cloned into pSB1C3 (SpyTag-GFP-SpyCatcher, Basic Part: BBa_K4652000). Then, the part was connected with a T7 promoter (Part:BBa_K1833999), a strong RBS (Part:BBa_B0030), and a double terminator (Part:BBa_B0015). This setup mirrored the context of pT7-eGFP (Part:BBa_K1833000)³, with the exception of the added SpyTag and SpyCatcher. The final construct was verified using colony PCR (Figure 1) and further validated through DNA sequencing. This resultant construct was designated as the improved BioBrick part, pT7-SpyTag-GFP-SpyCatcher (Composite Part:BBa_K4652002)⁴.

To assess whether the thermostability of GFP was enhanced by the addition of SpyTag and SpyCatcher, lysates from the transformed E. coli BL21, induced with IPTG, were subjected to a heat tolerance test at 90°C – a temperature known to degrade wild-type GFP. As illustrated in Figure 2, despite a pronounced decline in activity within the first minute of treatment, the fluorescence intensities remained relatively consistent up to 3 minutes. The retention of 22% GFP activity indicates a marked improvement in thermostability compared to the mere 1% observed for wild-type GFP at 90°C after 3 minutes (Figure 3).

Figure 2. Thermostability of cyclized SpyTag-GFPmut3b-SpyCatcher protein at 90°C. E. coli BL21 transformed with BBa_K4652002 was induced using 0.3 mM IPTG at 25°C for 20 hrs. Bacterial lysates were subjected to 90°C treatment for 3 min with an interval of 0.5 min each. Subsequently, the fluorescence of 100μl from each treated lysate was measured at Ex/Em = 483/513 nm. All values were divided by the average of the untreated control, with the resulting ratio representing the fluorescence fold change.


Figure 3. Comparison of wild-type GFPmut3b and cyclized SpyTag-GPTmut3b-SypCatcher proteins. E. coli BL21 transformed with indicated plasmid was induced using 0.3 mM IPTG at 25°C for 20 hrs. Bacterial lysates were subjected to temperatures of 90°C for 3 min. The fluorescence of 100μl lysates was measured at Ex/Em = 483/513 nm. All values were divided by the average of the untreated control, with the resulting ratio representing the fluorescence fold change.

Furthermore, the lysates were also exposed to 100°C for durations ranging from 0.5 to 3 minutes. Remarkably, GFP activity demonstrated tolerance at boiling temperatures (Figure 4), maintaining levels comparable to those observed at 90°C (Figure 2). However, there was a more pronounced loss of activity at the 0.5-minute mark. The data is consistent with Dr. Schoene’s report where β-lactamase with SpyTag at N-terminus and SpyCatcher at C-terminus exhibited resilience to boiling temperature⁵.

Figure 4. Thermostability of cyclized SpyTag-GFPmut3b-SpyCatcher protein at 100°C. E. coli BL21 transformed with BBa_K4652002 was induced using 0.3 mM IPTG at 25°C for 20 hrs. Bacterial lysates were subjected to 100°C treatment for 3 min with an interval of 0.5 min each. The fluorescence of 100μl from each treated lysate was measured at Ex/Em = 483/513 nm. All values were divided by the average of the untreated control, with the resulting ratio representing the fluorescence fold change.

In addition, we prolonged the exposure of cyclized GFP to 100°C up to 30 minutes. As shown in Figure 5, the GFP signal retained approximately 45% of its activity within the first minute, 30-35% up to 3 minutes, 20-28% up to 5 minutes, 10% at 10 minutes, and dropped below 5% after 15 minutes. These results clearly highlight the potential thermal tolerance of a synthetic cyclized protein engineered using SpyTag and SpyCatcher elements.

Figure 5. Thermal tolerance of cyclized SpyTag-GFPmut3b-SpyCatcher protein at 100°C. E. coli BL21 transformed with BBa_K4652002 was induced using 0.3 mM IPTG at 25°C for 20 hrs. Bacterial lysates were subjected to 100°C treatment for up to 30 min at the indicated time points. The fluorescence of 100μl from each treated lysate was measured at Ex/Em = 483/513 nm. All values were divided by the average of the untreated control, with the resulting ratio representing the fluorescence fold change.

To investigate the impact of cyclization on protein thermostability, we created a linear version of SpyTag-GFP-SpyCatcher by introducing a mutation, substituting Asp7 with Ala7 on SpyTag, using site-directed mutagenesis by high-fidelity KOD-plus DNA polymerase with a primer sequence of 5'- AGCCCACATCGTGATGGTGGCAGCCTACAAGCCGACGAAG -3’. The mutated site was confirmed by DNA sequencing. This mutant (SpyTag (D7A)-GFP-SpyCatcher, Part:BBa_K4652001) was constructed in the same format, resulting in the expression plasmid of T7-RBS-SpyTag (D7A)-GFP-SpyCatcher-Tr (Part:BBa_K4652003)⁶.

Figure 6 clearly demonstrates that the linear version fails to withstand temperatures of 90°C beyond 1 minute, even though it maintained high activity up to 0.5 minutes. Meanwhile, the cyclized GFP exhibited fluorescence intensities of around 20-30%, when exposed to 90°C or even under boiling conditions (100°C) for as long as 3 minutes. These outcomes underscore the enhanced thermal tolerance conferred by protein cyclization.

To validate the protein structures resulting from SpyTag and SpyCatcher cyclization and its linear counterpart generated through mutation, we conducted SDS-PAGE followed by Coomassie Blue Staining on IPTG-induced bacterial lysates. In Figure 7, the linear mutant manifested at its anticipated size of 44.22 kDa. The mobility of the cyclized protein differed, likely due to changes in charge or conformation. This observation aligns with Dr. Schoene's findings¹ and supports the results from our heat tolerance fluorescence experiments. Notably, our protein analysis can partly account for the diminished activity observed during high-temperature exposure in Figure 6, possibly attributable to incomplete GFP cyclization. When GFP continues exposure at near boiling temperatures, the cyclized form exhibits marked resistance over an extended duration.

Through a series of genetic and biochemical modifications, we successfully enhanced the thermostability of GFP protein. By removing its start and stop codons and integrating SpyTag and SpyCatcher at the N and C termini, respectively, we created a cyclized GFP protein with improved resilience to thermal stress. Experimental evidence from heat tolerance tests underscored the cyclized GFP's superior ability to retain activity under extreme temperatures compared to its linear counterpart and the wild-type GFP. Our findings are consistent with previous researches attempting cyclization of a protein, suggesting that the introduction of SpyTag and SpyCatcher not only provides a method to stabilize proteins but also has potential applications in synthetic biology requiring thermally stable proteins.

In Prof. Fan Li's laboratory, novel PCL-degrading enzymes, PCLase I and PCLase II, were identified and purified from Pseudomonas hydrolytica⁷. This discovery was made possible by cultivating the bacteria in a PCL-emulsified medium. The team conducted an in-depth study of the PCLase enzymes, examining aspects such as enzyme activity, the influence of pH and temperature, substrate specificity, degradation products, as well as the associated gene sequences and protein structures. Learned with this comprehensive data and the enzymes' impressive PCL-degrading efficiency, our aim is to enhance their thermostability.

Using SpyRing technique, we designed a T7-promoter-based gene expression plasmid carrying either PCLase I or PCLase II coding regions flanked by a N-terminal SpyTag and a C-terminal SpyCatcher. When expressed in IPTG-induced E. coli BL21 cells, these plasmids are expected to produce PCL-degrading enzymes with SpyRing, which should spontaneously cyclize and exhibit enhanced thermal stability.

PCLase I and PCLase II gene sequences with N-terminal SpyTag and C-terminal SpyCatcher were synthesized by Integrated DNA Technologies, Inc. (IDT) and then cloned into pSB1C3, respectively (SpyTag-PCLase1-SpyCatcher, Part:BBa_K4652008⁸;SpyTag-PCLase2-SpyCatcher, Part:BBa_K4652012⁹). Then, the parts were connected with a T7 promoter (Part:BBa_K1833999), a strong RBS (Part:BBa_B0030), and a double terminator (Part:BBa_B0015). The final construct was verified using colony PCR (Figure 8) and further validated through DNA sequencing. These resultant constructs were designated as T7-SpyTag-PCLase1-SpyCatcher (Part:BBa_K4652010)¹⁰ and T7-SpyTag-PCLase2-SpyCatcher (Part:BBa_K4652013)¹¹, respectively.

Figure 8. Verification of T7-SpyTag-PCLase1-SpyCatcher (Part:BBa_K4652010) and T7-SpyTag-PCLase2-SpyCatcher (Part:BBa_K4652013) using colony PCR. PCR was performed using a CmR-specific forward primer from the vector and a PCLase-specific reverse primer from the gene. The expected size of the amplified DNA fragments is 2395 bp for T7-SpyTag-PCLase1-SpyCatcher and 2561 bp for T7-SpyTag-PCLase2-SpyCatcher, respectively. The rightmost lane displays a 1 kb DNA ladder. The numbers correspond to selected colonies, with one control (lane 7) derived from a mock pick from a clear zone on the plate.

PCL-DEGRADING LIPASE COMPARISON

To compare the lipase activities of PCLase I, PCLase II, and other commercially available PCR-degrading enzymes such as BCLA¹² and CALB¹³, we conducted a pNPB assay. In this assay, a potential lipase breaks down the ester bond of p-nitrophenylbutyrate (pNPB), producing p-Nitrophenol. The concentration of p-Nitrophenol can be measured at 405nm, and these measurements are corresponding to the lipase activity.

Lysates from E. coli BL21, which carried the T7 promoter-driven expression plasmid with the indicated genes in the same context, were collected after being induced with 0.3 mM IPTG at 25°C for 20 hours. The lipase activities within these lysates were assessed using the pNPB assay¹². As shown in Figure 9, PCLase I exhibited the significantly highest readings at 405 nm. This suggests that under our experimental conditions, PCLase I is the most effective lipase, demonstrating potential activity in decomposing PCL through the hydrolysis of the ester bonds between polymers. Consequently, we chose to investigate the characteristics of PCLase I (hereafter referred to as PCLase for short) in terms of its thermostability, protein structure, PCL degradation capability, and its potential use in real-world products.

Figure 9. Comparison between lipase activities of BCLA, CALB, PCLase I, and PCLase II using pNPB assay. E. coli BL21 was transformed using the indicated T7 promoter-driven gene expression plasmid. The bacteria were induced by 0.3 mM of IPTG at 25°C for 20 hours. Subsequently, the lysates were harvested using 0.1 mm Disruptor Beads (Scientific Industries, Inc). A 20 µL aliquot of these lysates was combined with 175 µL of Tris-HCl buffer (20mM, pH=8) and 5 µL of pNPB (40 mM dissolved in 2-methyl-2-butanol). Lipase activity was read at 405 nm based on p-Nitrophenol production. The obtained readings were normalized with the OD600 values at the time of bacterial lysate collection.

THERMOSTABILITY

To assess the thermostability of PCLase, the bacterial lysates, prepared as described earlier, were exposed to a 100°C treatment followed by a pNPB assay to measure lipase activity. PCLase activity decreased to 20% in 5 min but remained stable for up to 30 minutes (Figure 10). The phenomena were consistent with the observations made in lysates containing cyclized GFP (Part:BBa_K4652002)4, but, notably, PCLase exhibited much more prolonged heat resistance at the boiling temperature (i.e., cyclized GFP is tolerate for 5 min in Figure 5, while cyclized PLCase is for 30 min).

Figure 10. Thermostability of PCLase lipase activity. E. coli BL21 was transformed with T7-SpyTag-PCLase1-SpyCatcher (Part:BBa_K4652010). The bacteria were induced by 0.3 mM of IPTG at 25°C for 20 hours. Subsequently, the lysates were harvested using 0.1 mm Disruptor Beads (Scientific Industries, Inc). After treated at 100°C for the indicated time, a 20 µL aliquot of these lysates was combined with 175 µL of Tris-HCl buffer (20mM, pH=8) and 5 µL of pNPB (40 mM dissolved in 2-methyl-2-butanol). Lipase activity was read at 405 nm based on p-Nitrophenol production. All values were divided by the average of the untreated control, with the resulting ratio representing the lipase activity fold change.

To verify the protein structure and its correlation to thermal tolerance, in a similar approach on T7-RBS-SpyTag (D7A)-GFP-SpyCatcher-Tr (Part:BBa_K4652003)6, we engineered a SpyTag mutation of PCLase and designated as T7-RBS-SpyTag (D7A)-PCLase1-SpyCatcher-Tr (Part:BBa_K4652011)¹⁴. The lysates, as collected previously, were subject to SDS-PAGE and Coomassie Blue staining. In Figure 11, the result showed a distinct band corresponding to the linear form of SpyTag (D7A)-PCLase1-SpyCatcher, with an expected size of 46.53 kDa. However, the lysates from SpyTag-PCLase1-SpyCatcher displayed somewhat fuzzy bands representing incomplete cyclized or fully cyclized forms. This ambiguity might arise from aggregation or polymerization due to isopeptide bond formation¹⁵. It's worth noting that the presence of incompletely cyclized PCLase could partially account for the reduced lipase activity after heating observed in the experiment in Figure 10.

Figure 11. Protein structure analysis of cyclized form and linear form of SpyTag-PCLase1-SpyCatcher. 15 µg of IPTG-induced lysates were run on a 4-12% gradient SDS-PAGE (BIO-RAD) and stained with standard Coomassie Blue. The AceColor Prestained Protein Marker (10 - 180 kDa) was used for size reference. Lane 1 refers to the linear proteins derived from lysates of SpyTag (D7A)-PCLase1-SpyCatcher (BBa_K4652011). Lane 2 refers to the cyclized proteins derived from lysates of SpyTag-PCLase1-SpyCatcher (BBa_K4652010). The predicted molecular weight for the SpyTag-PCLase1-SpyCatcher protein is 46.53 kDa.

In addition, we compared the thermostabilities between lipase activities in lysates with linear PCLase (SpyTag (D7A)-PCLase1-SpyCatcher) and cyclized PCLase (SpyTag-PCLase1-SpyCatcher) using the pNPB assay. After being treated at temperatures ranging from 70°C to 100°C for 10 min, the cyclized lipase retained higher activity compared to the linear form of PCLase (Figure 12). It maintained over 50% activity up to 90°C and retained up to 20% at 100°C. It's worth exploring whether purifying and concentrating the fully cyclized PCLase could enhance the lipase activity in the lysates for thermal testing. Taken together, the experimental data demonstrated successfully making a thermostable PCLase by cyclization through SpyRing technique.

Figure 12. Thermostability of PCLase lipase activities between cyclized PCLase (SpyTag-PCLase1-SpyCatcher, BBa_K4652010) and linear PCLase (SpyTag (D7A)-PCLase1-SpyCatcher, BBa_K4652011). E. coli BL21 was transformed with the indicated plasmid and then was induced by 0.3 mM of IPTG at 25°C for 20 hours. Subsequently, the lysates were harvested using 0.1 mm Disruptor Beads (Scientific Industries, Inc). After treated at 100°C for the indicated time, a 20 µL aliquot of these lysates was combined with 175 µL of Tris-HCl buffer (20mM, pH=8) and 5 µL of pNPB (40 mM dissolved in 2-methyl-2-butanol). Lipase activity was read at 405 nm based on p-Nitrophenol production. All values were divided by the average of the untreated control, with the resulting ratio representing the lipase activity fold change.

PCL GRANULE DECOMPOSITION

To evaluate PCLase's degradation potential on PCL plastic, we purchased polycaprolactone raw materials in granule form (Mn 80,000, #440744, SIGMA-ALDRICH). Measuring weight loss over days is a feasible approach to assess enzyme degradation efficiency¹⁶. Around 10 mg of PCL granules were placed in 5 ml of Tris-HCl at a pH of 8 and mixed with 500 µl of bacterial lysates containing PCLase. Incubate at 50°C for 24 hrs, the granules were taken out, oven-dried, and weighed. These granules were introduced again in the fresh buffer and fresh enzymes. This process was monitored over a span of 5 days. As shown in Figure 13, when compared to controls without any enzyme treatment, PCLase resulted in a daily weight loss of approximately 10-20%. By the final day of observation, the PCL granule had lost nearly half its weight. The findings suggest that the PCLase produced in our lab is effective in degrading PCL plastics.

Figure 13. PCL granule degradation by PCLase. PCL granules (Mn 80,000) were obtained from Sigma-Aldrich. Around 10 mg of the granules were immersed in 5 ml of Tris-HCL buffer (100mM, pH = 8) supplemented with 500 µl of PCLase or without any enzyme as a control. After a 24-hour incubation at 50°C, the granules were removed, dried and weighed, followed by replaced with fresh buffer and fresh enzymes. The cycle continued for 5 days. Weight loss was determined relative to the average weight on Day 1. Data represents the mean and standard error from three independent sets.

PCL NANOFIBER FILM DEGRADATION

Electrospun PCL nanofiber films are prevalent in medical care, especially for applications like wound dressings¹⁷. While PCL materials have FDA approval, their 2-3 year degradation time appears too long. Consequently, enzymes that degrade PCL have gained significant research interest. Prof. Hsiao-Chun Yang from the Department of Fiber and Composite Materials at Feng Chia University is a renowned expert in this field. We consulted him for guidance on our PCLase and obtained some PCL films for testing. To increase the PCLase protein concentration, we passed the sonicated bacterial lysates through 0.45 µm PES (Polyethersulfone) Syringe Filters (HYUNDAI MICRO CO.,LTD.) and subsequently concentrated the proteins using 30K Microsep™ Advance Centrifugal Devices (Pall Corporation). We then submerged 1-cm x 1-cm PCL films in 3 ml of Tris-HCl buffer (pH = 8) with 30 µl of the concentrated lysates, incubating at 50°C for 24 hrs. As demonstrated in Figure 14, PCLase significantly and efficiently degraded PCL film compared to controls without enzyme treatment or those with GFP. In sum, the PCLase we developed exhibits lipase activity in the pNPB assay and is also capable of breaking down PCL granules and films.

Figure 14. PCL electrospun film degradation by PCLase. PCL films were obtained from Prof. Yang’s laboratory. A 1-cm x 1-cm PCL films was immersed in 3 ml of Tris-HCL buffer (100mM, pH = 8) either supplemented with 30 µl of concentrated PCLase, 30 µl of concentrated GFP, or left enzyme-free as a control. After a 24-hour incubation at 50°C, the film decomposition was observed. The pictures were representative data from 3 independent experiments.

APPLICATION OF PCLase-EMBEDDED PCL PRODUCT

Although many researches are working on lipase-embedded PCL materials^18,19,20, they often utilize existing PCL-degrading enzymes that are not able to endure the thermoforming process typical in standard PCL plastic production. We're wondering if our cyclized PCLase can address this issue. Traditionally, PCL granules are dissolved in organic solvents such as chloroform, dimethylformamide, or toluene²¹. Besides their inherent toxicity, these solvents often impair and deactivate any bioactive enzyme intended for mixing. To blend seamlessly with PCL without using organic solvents, we freeze-dried the concentrated PCLase into a powder form, facilitating its mixture with PCL powder (Figure 15).

Figure 15. PCL, GFP, PCLase materials in our studies. PCL granules were purchased from SIGMA-ALDRICH (Mn 80,000, #440744). PCL powder was purchased from MAGERIAL SCIENCE (100 mesh size, #24980-41-4). GFP powder and PCLase powder were prepared by ourselves using a freeze dryer (KINGMECH Freeze Dryers System #FD4.5-8P-L).

The PCL-PCLase composite containing 10 mg of PCL powder and 2 mg of PCLase powder in a glass bottle was heated to a pliable state at 100°C for 1min in a water bath. Once cooled, the PCLase-embedded PCL plastic resembled a raw fibrous membrane. Then, the bottle was filled with 3 ml of 100mM Tris-HCl buffer (pH = 8), followed by incubation at 50°C for 1 week. Compared to the GFP-embedded PCL plastic used as a control, the size of the PCLase-embedded plastics reduced to half by Day 3, to a tenth by Day 5, and had nearly vanished by Day 7 (Figure 16).

Notably, the buffer from the GFP-embedded membrane exhibits green fluorescence, likely due to excess GFP on the membrane's surface. It also could be a concern that GFP or any embedded protein might be released from the membrane, that warrants further refinement in protein embedding techniques. However, the membrane's distinct green fluorescence remains vivid under blue LED exposure even after 30 days of incubation in the buffer at 50°C (Figure 16, inset panel). This indicates that the protein retains its structural integrity and biological function in challenging conditions, making it promising for storage and transportation. Altogether, the results highlight the potential of thermostable PCLase-embedded PCL materials, suggesting their suitability for standard biodegradable plastic manufacturing processes worldwide.

Figure 16. GFP or PCLase-embedded PCL fibrous membrane. Each membrane was made by mixing and heating at 100°C for 1min with 10 mg of PCL powder and 2 mg of either GFP or PCLase protein powder. The membranes were submerged in 3 ml of Tris-HCL buffer (100mM, pH = 8) at 50°C and observed daily for a month. The images depict results from Days 1, 3, 5, and 7. The inset panel highlights GFP activity of GFP-embedded membrane on Day 30, captured under blue LED light. The pictures were representative data from 3 independent experiments.

Polycaprolactone (PCL), a widely-used biodegradable plastic, typically takes 2-3 years to decompose. To address this, our team identified PCL-degrading enzymes, PCLase, according to the research of Prof. Fan Li. Using the Dr. Mark Howarth’s SpyRing cyclization technique, we enhanced PCLase's thermostability, allowing it to retain activity at high temperatures. In practical tests, PCLase significantly reduced the weight of PCL granules within five days and similarly degraded PCL nanofiber films. Traditional PCL processing uses toxic solvents, but our freeze-drying method allowed PCLase to mix directly with PCL powder, leading to rapid degradation of the resulting composite. This implies that such a thermostable PCLase-PCL composite material is primed for integration into standard thermoforming plastic production processes in the real world.

Interestingly, the buffer containing the GFP-embedded membrane emitted a fluorescent signal, indicating potential protein leaching and underscoring the need for further refinement. However, our GFP-embedded membrane demonstrated remarkable stability of the embedded protein, enduring for 30 days or more in a challenging 50°C environment. In summary, our research presents a groundbreaking approach to faster biodegradable plastics, offering a sustainable solution to a global environmental challenge.