1 Introduction
Synthetic polymers pervade all aspects of modern life, due to their low cost, high durability, and
impressive range of tunability. Originally developed to avoid the use of animal-based products, plastics
have now become so widespread that their leakage into the biosphere and accumulation in landfills is
creating a global-scale environmental crisis. Among them, synthetic plastic polymers Polyethylene
terephthalate (PET) is one of the most widely used forms of synthetic plastics in several industrial fields,
which has been reported spreading as particles less than 5 mm in environments [1] and being captured from
environments with low efficiencies. Thus, plastic waste may have a negative effect on human health and
reveal a serious threatening to environments [2]. In comparison to the other solving methods, enzymatic
treatment utilizing PET hydrolyzing enzymes is environmental-friendly because it avoids the use of hazardous
chemicals. However, the attempt to solve this problem via bio-enzymes is faced with multiple challenges. For
a long time, biodegradation of plastic PET with its degradation enzymes are highly limited by their low
degradation ability in mild temperature [3].
Fortunately, in 2016, Yoshida et al. [4] reported the discovery of the bacterium, Ideonella
sakaiensis
201-F6, which develops a two-enzyme system to deconstruct PET plastic to TPA and EG molecules, which could
be effectively catabolized as a carbon and energy sources for cells. Further research of this two enzymes
system reveals that enzyme IsPETase is a cutinase-like serine hydrolase that attacks the PET plastic
polymer
with the highest efficiency in mild temperature among all previously discovered PET degradation enzymes. The
metagenome-derived leaf-branch compost cutinase (LCC) shows higher activity on low crystallinity PET film
degradation [5]. However, LCC is still not generally applicable due to the cost and high applied optimum
temperature around 72 °C [5], which highly limit its further application in engineered bacteria
bio-degradation system. Furthermore, the percentage of PET degradation intermediates MHET generated via LCC
were near to 60% [6], which could not be removed by MHETase effectively due to the possible loss of its
enzymatic activity at LCC applied temperature. Compared to LCC, the MHET removing efficiency in IsPETase
and
MHETase coexisting system was near to 100% [7]. LCC reveals near 5.5 times and 4 times lower enzymatic
activity to PET film and high crystallinity PET substrate at 30 °C compared to IsPETase [4] [8]. IsPETase
was
considerably more active against PET film at low temperatures than other PET degradation enzymes including
TfH, LCC, and FsC [4]. And more efforts of IsPETase protein rational evolution are made and these
works
largely improved enzyme IsPETase thermostability with a Tm value that was increased by 8.81 °C and
reacting
activity by around 14-fold at 40 °C [8], whose activity is even higher than LCC and BhrPETase isolated from
bacterium HR29 [6][8]. Furthermore, in a latter research, highly synergistic relationship between IsPETase
and MHETase was discovered in the conversion of amorphous PET film to monomers [7], which effectively
accelerate PET degradation. However, this two enzymes system may be still limited due to enzymatic loss
caused by protein fusion, inhibition effects and diffusion of intermediates. And methods of rational
evolution of protein may not be applicable to further improve the overall turnover rate [9]. In detail, the
MHET molecules produced via PET degradation may be competitive to IsPETase active sites, which was
revealed
via calculation results shown in this work. Despite of this, MHET molecules diffusion in space may also be a
problem for MHETase remove MHET in time. Therefore, in this work, we designed an intricate multidomain
protein scaffold composed of short peptide tags of RIAD [11] and RIDD [12], enzymes of IsPETase and
MHETase
and protein hydrophobin [11], in which the enzymes constructed in near positions to each other may work with
highly synergistic relationship. All proteins involved have unique functions in this system. IsPETase
and
MHETase are two enzymes involved deconstructing polymer PET plastic to MHET and MHET to TPA molecules,
respectively. And hydrophobin protein, a small fungal protein, possess positive effects on altering the
physicochemical properties of PET surfaces and enzyme aggregation enhancement when it was fused with PET
degradation enzyme cutinase [13]. Here, hydrophobin protein are involved in our designed system with
possible functions of adhering to PET polymers and altering the physicochemical properties of PET for
degradation improvement [14]. The peptides of RIDD and RIAD originated from cAMP-dependent protein kinase
(PKA) and the A kinase-anchoring proteins (AKAPs), respectively. The RIAD peptide is capable of specifically
binds to the RIDD dimer with strong affinity [15]. The following two features make them ideal protein
binding modular for our system assembly: (1) the tiny size (44 and 18 amino acids, respectively), which
minimizes the disturbances to the structure and activity of the enzymes when fused with these peptides, (2)
the strong binding affinity (with a KD of 1.2 nM between RIDD dimer and RIAD peptide demonstrated in our
colleagues' previous work [15]) to ensure the stability of the whole enzyme complexes in the environment.
In our designing, RIAD or RIDD was fused to the N termini of hydrophobin and IsPETase linked by a
flexible
linker (GGGGS)3 respectively and another RIDD was fused to the C termini of MHETase by a flexible
linker
(GGGGS)3. MHETase and IsPETase are delicately constructed in near positions (as shown in
structure
calculations) via RIDD peptides strong affinity to accelerate MHET removement and reduce competitive
inhibition to IsPETase. Meanwhile, short peptides RIDD fused to enzymes effectively avoid
disturbances to
enzymes active sites. And for prudential and rational designing, the multicomplex enzymes system designed in
our work are evaluated via Molecular Dynamic simulations (MD) to ensure the structure designing
feasibilities. The components binding stability were measured and the whole system was revealed to maintain
its components binding stably with enzymes fixed in relative near positions to each other in liquid
environment. Furthermore, the key residues distances to Ca2+ ion in Ca2+ binding site
reveal maintenance of
structure stability. Then, we construct one plasmid that encoding the multicomplex enzyme system with liking
methods mentioned above and expressed it in the bacterium Escherichia coli with 1:1:1 theoretical
ratio for
each designed component. This ratio may delicately enhance our designed components assembling and avoid
possible mis-assembly. The assembled protein complex systems in the extracellular spaces were demonstrated
to effectively alter the physicochemical properties of PET and enhance the overall turnover rate of PET
degradation.
2 Results And Discussions
2.1 MHET may possess tendency limiting the efficiency of whole PET degradation process and a delicate
multidomain modular system is designed
In previous research, inhibition effect by MHET accumulating in the reaction medium has been demonstrated as
a key factor limiting the efficiency of the polyester hydrolases from Thermobifida fusca, whose
structure is
highly similar to IsPETase discovered in 2016. [4]. As the alignment shown in Figure 1 (a), the
structure of
the IsPETase is highly similar to polyester hydrolases from Thermobifida fusca [16] with same
key
residues
possessing similar enzymatic mechanisms in active site and may be inhibited by MHET as well. Results
calculated from Autodock Vina software [17] also reveal this probability, in which MHET, compared to PET
substrate, is shown to have the same binding affinity level to IsPETase enzyme active site and may be
competitive to IsPETase. Our later experiment results also reveal this probability. Based on these,
IsPETase
efficiency is considered to be limited and inhibited by MHET molecule and MHET need to be eliminated from
the reaction medium to expose occupied IsPETase active site to PET substrates. Despite of this, MHET
molecules diffusion in space may also be a limiting factor to PET degradation rate because MHETase may not
be able to effectively remove MHET from reaction medium in time. Meanwhile, as shown in Figure 1 (b), MHET
molecule reveals a stronger binding tendency to MHETase active site than to IsPETase active site and
this
stronger tendency reveals that MHET may be more likely to enter into MHETase active site rather than
occupying IsPETase active site when MHETase is near to IsPETase in liquid environment. Based
on these, for
the purposes of reducing MHET inhibition effect to IsPETase, eliminating MHET accumulation in the
reaction
medium and degrading PET film with higher speed, we designed a delicate multidomain proteins system as
mentioned above, in which the enzymes IsPETase and MHETase constructed in near positions to each
other work
with high synergistic relationship [7] to improve the whole PET plastic turnover rate. The hydrophobin
protein involved is expected to enhance enzymes aggregation, alter the physicochemical properties of PET
surfaces and improve PET decomposing rate.
Figure 1: (a) (a) Alignment of the polyester hydrolases from Thermobifida fusca and PETase from
Ideonella
sakaiensis and same key residues involved in enzymatic reaction. (b) Binding affinity energy of MHET to
PETase and MHET enzyme active sites calculated via Autodock Vina. The results reveal a strong entering
tendency of MHET to IsPETase active site and higher binding affinity to MHETase active site. These
reulsts
provide helpful references and insights of strong binding tendency of MHET to IsPETase and MHETase
active
sites. (c) Entrance of MHET molecule to IsPETase active site with strong binding affinity. (d) The
synergestic relationships of IsPETase and MHETase in PET degradation.
2.2 Rational designing strategy calculations and evaluation for artificially designed system
In our designing, RIAD or RIDD was fused to the N termini of hydrophobin and IsPETase by a flexible
linker
(GGGGS)3 respectively and another RIDD was fused to the C termini of MHETase by a flexible linker
(GGGGS)3.
Flexible linker designing may drive three protein components to central RIDD-dimer-RIAD scaffolds to lower
the solvent surface area and physically fix IsPETase and MHETase in a near position to each other.
Such
designing may ensure that the reaction products of IsPETase effectively enter the near MHETase
enzymatic
site as new substrates, by which improves the whole reaction rate. Therefore, for the purposes of prudential
designing mentioned above, the binding stability of the whole system and enzymes active sites are calculated
to ensure that the modular enzymes components are fixed stably in near position and the whole structure is
well maintained as a complex in liquid environment.
For each component structure, IsPETase and MHETase are derived from previous X-ray diffraction
structures
and then RIDD is fused to enzymes termini as methods mentioned above [18] [19]. The structure of
hydrophobin-RIAD is not available in PDB dataset and is predicted by Phyre2 [20]. Then, with these raw
components structures, our team constructs the raw model of the multidomain proteins system (as shown in
supplementary Figure 2 (b)). The multidomain proteins components in our research are constructed via VMD
[19] and Z-dock [21] software.
Figure 2: The artificially designed complex system was applied with 150 ns explicit solvent MD simulation.
The last 60 ns MD simulation results revealed that the whole structure was well equilibrated and therefore
were considered to be sampled and analyzed. (a) Possible mature structure generated via MD calculations.
Compared to initial raw model of each component (shown in supplementary material Figure 1 (b)), each
component (IsPETase-RIDD, MHETase-RIDD, Hydrophobin-RIAD) are drawn toward to central RIAD and RIDD
dimer scaffolds to lower the solvent contact surface area. Components are bound to each other firmly and no
diffusion behaviors are observed in simulation. (b) RMSD values of simulated system. (c) Illustration of
components assembly process. (d) Last 60 ns simulation results of gyration radius of modular enzymes system
with total 150 ns simulation RMSD. Gyration radius values are fluctuated slightly, which reveal the
stability of the whole system density and structure in liquid environment. In the long time MD simulation,
component (IsPETase-RIDD, MHETase-RIDD and Hydrophobin-RIAD) is drawn to central scaffolds by (GGGGS)3
linker to lower the solvent contact surface area and maintained stably in near positions, whose results
meets our initial designing expectations.
Each component (PETase-RIDD, MHETase-RIDD and Hydrophobin-RIAD) is applied with 20 ns explicit solvent
simulation for structural optimization. Notably, for the purposes of obtaining an accurate structure of
predicted protein in aqueous environment, the structure of Hydrophobin predicted by Phyre2 is firstly
applied with 50 ns explicit solvent MD simulation before fusing to RIAD. Then, these components are
initially set in near positions to each other and spontaneously bound together as a complex via 20 ns
explicit solvent simulation to obtain a relative reasonable binding structure formed via affinity between
RIDD dimers and RIAD. This structure is highly consistent to the X-ray revealed binding complex formed by
free RIDD dimers and RIAD components [15]. With this result as initial structure, we applied it with 50 ns
explicit solvent MD simulation and then lengthen MD run with 100 ns for further optimization, equilibration
and sampling (All simulations totaling 1 μs). Atomistic MD simulations were performed using Amber 20 [22]
and the ff99SB force field [23] for proteins solvated by OPC water model with 310 K temperature setting. The
nonbonded interactions were truncated at a 10 Å cutoff. The simulation Truncated Octahedron box of about 18
nm × 18 nm × 18 nm size contained the system and 61888 water molecules. The size of the box was chosen to be
large enough so that no interactions through periodic boundary conditions can occur. By these, a new
delicate modular enzymes system model is established by processes mentioned above. The final calculated
artificially designed multidomain system structure is shown in Figure 2 (a). In the long scale simulation,
each component (PETase-RIDD, MHETase-RIDD and Hydrophobin-RIAD) is drawn toward to the central RIDD dimer
and RIAD scaffolds by (GGGGS)3 linker to lower the solvent contact surface area. This structural
change may
improve the stability of the whole complex and meet our designing expectations that fixing each component in
near positions. As shown in Figure 2 (d), the gyration radius of the whole system is relatively stable in
the last 60 ns simulation, which demonstrates that the strong binding affinities between RIDD and RIAD parts
are sufficient to drive three protein components fixed in near positions and the whole structure is
relatively stable. In the long time MD simulation, the components are bound to each other firmly and no
diffusion behavior is observed, as shown in the simulation video provided in the supplementary material.
Therefore, the binding force between RIDD and RIAD is considered to be sufficient to maintain the stability
of artificially designed complex. By these, three different proteins are designed to be fixed stably in near
positions with high synergistic relationship (as shown in Figure 2 (a)) to alter the physicochemical
properties of the PET surface and accelerate reaction turnover rate. And later experiment reveals successful
assemblies of the system and the stability of components binding as a complex. (More detailed information of
the raw model construction procedures and simulation results analysis are available in supplementary
material). Specifically, MHETase-RIDD Ca2+ binding site with its binding ions that may have
important
functions for structural stability is shown to be well maintained [22]. Four key residues that form the
binding site and function important roles in maintaining Ca2+ ion (Asp 588, Asp 591, Gly 592, Asp
599 and
Asn 600) are evaluated. These key residues geometric centers distances to Ca2+ ion in binding
sites
fluctuate slightly around 0.5 Å, revealing the high maintenance of Ca2+ binding site structure,
as shown in
Figure 2 (e).
These results successfully demonstrate the feasibility of our designing strategy as described above. Later
experiments further demonstrate our designed components enzymatic activity and functions in complex system.
References
[1] Chunguang Liu, Jia Li, Yilei Zhang, Lei Wang, Jie Deng, Yuan Gao, Lu Yua, Junjie Zhang, Hongwen Sun,
Widespread distribution of PET and PC microplastics in dust in urban China and their estimated human
exposure, 2019.
[2] Fusako Kawai, Takeshi Kawabata, Masayuki Oda, Current knowledge on enzymatic PET degradation and its
possible application to waste stream management and other fields, 2019.
[3] Xingxiang Xi, Kefeng Ni, Helong Hao, Yuepeng Shang, Bo Zhao, Zhen Qian, Secretory expression in Bacillus
subtilis and biochemical characterization of a highly thermostable polyethylene terephthalate hydrolase from
bacterium HR29, 2021.
[4] Shosuke Yoshida, Kazumi Hiraga, Toshihiko Takehana, Ikuo Taniguchi, Hironao Yamaji, Yasuhito Maeda,
Kiyotsuna Toyohara, Kenji Miyamoto, Yoshiharu Kimura, Kohei Oda, A bacterium that degrades and assimilates
poly (ethylene terephthalate), 2016.
[5] V. Tournier, C. M. Topham, A. Gilles, B. David, C. Folgoas, E. Moya-Leclair, E. Kamionka, M.-L.
Desrousseaux, H. Texier, S. Gavalda, M. Cot, E. Guémard, M. Dalibey, J. Nomme, G. Cioci, S. Barbe, M.
Chateau, I. André1, S. Duquesne, A. Marty, An engineered PET depolymerase to break down and recycle plastic
bottles, 2020.
[6] Xingxiang Xi, Kefeng Ni, Helong Hao, Yuepeng Shang, Bo Zhao, Zhen Qian, Secretory expression in Bacillus
subtilis and biochemical characterization of a highly thermostable polyethylene terephthalate hydrolase from
bacterium HR29, 2021.
[7] Brandon C. Knotta, Erika Ericksona, Mark D. Allenb, Japheth E. Gado, Rosie Graham, Fiona L. Kearns,
Isabel Pardo, Ece Topuzlu, Jared J. Anderson, Harry P. Austin, Graham Dominick, Christopher W. Johnson,
Nicholas A. Rorrer, Caralyn J. Szostkiewicz, Valérie Copié, Christina M. Paynec, H. Lee Woodcock, Bryon S.
Donohoef, Gregg T. Beckham, and John E. McGeehan, Characterization and engineering of a two-enzyme system
for plastics depolymerization, 2019.
[8] Hyeoncheol Francis Son, In Jin Cho, Seongjoon Joo, Hogyun Seo, Hye-Young Sagong, So Young Choi, Sang Yup
Lee, Kyung-Jin Kim, Rational Protein Engineering of Thermo-Stable PETase from Ideonella sakaiensis
for Highly Efficient PET Degradation, 2019.
[9] Erickson E, Shakespeare TJ, Bratti F, Buss BL, Graham R, Hawkins MA, König G, Michener WE, Miscall J,
Ramirez KJ, Rorrer NA, Zahn M, Pickford AR, McGeehan JE, Beckham G, Comparative performance of PETase as a
function of reaction conditions, substrate properties, and product accumulation, 2021.
[10] Boneta S, Arafet K, Moliner V, QM/MM Study of the Enzymatic Biodegradation Mechanism of Polyethylene
Terephthalate, 2021.
[11] Carlson, C. R. et al. Delineation of type I protein kinase A-selective signaling events using an RI
anchoring disruptor. J. Biol. Chem. 281, 21535–21545, 2006. 24.
[12] Gold, M. G. et al. Molecular basis of AKAP specificity for PKA regulatory subunits. Mol. Cell 24,
383–395, 2006.
[13] Doris Ribitsch, Enrique Herrero Acero, Agnieszka Przylucka, Sabine Zitzenbacher, Annemarie Marold,
Caroline Gamerith, Rupert Tscheließnig, Alois Jungbauer, Harald Rennhofer, Helga Lichtenegger, Heinz
Amenitsch, Klaus Bonazz, Christian P. Kubicek, Irina S. Druzhinina, Georg M. Guebitz, Enhanced
Cutinase-Catalyzed Hydrolysis of Polyethylene Terephthalate by Covalent Fusion to Hydrophobins, 2015.
[14] Yuanzheng Wu, Jishun Li, Hetong Yang, Hyun-Jae Shin, Fungal and mushroom hydrophobins: A review,
2017.
[15] Wei Kang, Tian Ma, Min Liu, Jiale Qu, Zhenjun Liu, Huawei Zhang, Bin Shi, Shuai Fu, Juncai Ma, Louis
Tung Faat Lai, Sicong He, Jianan Qu, Shannon Wing-Ngor Au, Byung Ho Kang, Wilson Chun Yu Lau, Zixin Deng,
Jiang Xia, Tiangang Liu, Modular enzyme assembly for enhanced cascade biocatalysis and metabolic flux,
2019.
[16] Then, J. et al. Ca2+ and Mg2+ binding site engineering increases the degradation of polyethylene
terephthalate films by polyester hydrolases from Thermobifida fusca, 2015.
[17] O. Trott, A. J. Olson, AutoDock Vina: improving the speed and accuracy of docking with a new scoring
function, efficient optimization and multithreading, 2010.
[19] Humphrey, W., Dalke, A. and Schulten, K., J. Molec. Graphics, VMD – Visual Molecular Dynamics, 1996.
[20] Kelley LA, The Phyre2 web portal for protein modeling, prediction and analysis, 2015.
[21] P Brian, W Tong, Z Weng, M-ZDOCK: a grid-based approach for Cn symmetric multimer docking, 2005.
[22] Case, D. A., Darden, T. A., T.E. Cheatham, I., Simmerling, C. L., Wang, J., Duke, R. E., Luo, R.,
Walker, R. C., Zhang, W., Merz, K. M.; Roberts, B., Hayik, S., Roitberg, A., Seabra, G., Swails, J. Goetz,
A.W., Kolossvary, I., Wong, K. F., Paesani, F., Vanicek, J., Wolf, R. M., Liu, J., Wu, X.; Brozell, S. R.,
Steinbrecher, T., Gohlke, H., Cai, Q., Ye, X., Wang, J., Hsieh, M.-J., Cui, G., Roe, D. R., Mathews, D. H.,
Seetin, M. G., Salomon-Ferrer, R., Sagui, C., Babin, V., Luchko, T., Gusarov, S., Kovalenko, A., Kollman, P.
A. AMBER 20, 2020.
[23] Chen-Yang Zhou, Fan Jiang, Yun-Dong Wu, Residue-Specific Force Field Based on Protein Coil Library.
RSFF2: Modification of AMBER ff99SB, 2015.
