深度学习指导的热裂褐藻角质酶工程用于促进PET解聚

doi:10.1016/S1872-2067(23)64470-5

催化学报 ›› 2023, Vol. 50: 229-238.DOI: 10.1016/S1872-2067(23)64470-5

深度学习指导的热裂褐藻角质酶工程用于促进PET解聚

孟帅奇^a^,¹, 李忠玉^b^,¹, 张鹏^a, Francisca Contreras^a, 季宇^a^,^*(), Ulrich Schwaneberg^a^,^c^,^*()

^a亚琛工业大学生物技术研究所, 德国
^b北京化工大学北京生物过程重点实验室, 北京 100029, 中国
^c莱布尼茨互动材料研究所, 德国

收稿日期:2023-03-16 接受日期:2023-06-15 出版日期:2023-07-18 发布日期:2023-07-25
通讯作者: *电子信箱: u.schwaneberg@biotec.rwth-aachen.de (U. Schwaneberg), yu.ji@biotec.rwth-aachen.de (Y. Ji).
作者简介:
¹共同第一作者.
基金资助:
中国留学基金委员会(201906880011)

Deep learning guided enzyme engineering of Thermobifida fusca cutinase for increased PET depolymerization

Shuaiqi Meng^a^,¹, Zhongyu Li^b^,¹, Peng Zhang^a, Francisca Contreras^a, Yu Ji^a^,^*(), Ulrich Schwaneberg^a^,^c^,^*()

^aInstitute of Biotechnology, RWTH Aachen University,Worringerweg 3, Aachen 52074, Germany
^bBeijing Bioprocess Key Laboratory, Beijing University of Chemical Technology, Beijing 100029, China
^cDWI-Leibniz Institute for Interactive Materials, Forckenbeckstraße 50, Aachen 52074, Germany

Received:2023-03-16 Accepted:2023-06-15 Online:2023-07-18 Published:2023-07-25
Contact: *E-mail: u.schwaneberg@biotec.rwth-aachen.de (U. Schwaneberg), yu.ji@biotec.rwth-aachen.de (Y. Ji).
About author:
¹These authors contributed equally to this work.
Supported by:
Ph.D. scholarship from the China Scholarship Council(201906880011)

摘要/Abstract

摘要：

有效和可持续的聚合物循环经济需要低二氧化碳排放和高效的回收工艺. 聚对苯二甲酸乙二醇酯(PET)的酶解聚是使PET聚合物成为聚合物循环经济一部分的第一步. 本文利用基于结构的深度学习模型来识别TfCut2中负责提高水解活性和增强稳定性的残基. 机器学习引导设计确定了新的有益位置(L32E, S35E, H77Y, R110L, S113E, T237Q, R245Q和E253H), 对其进行评估并逐步重组, 最终获得有益变体L32E/S113E/T237Q. 与TfCut2 WT相比, 后一种TfCut2变体表现出更好的PET解聚性能: 无定形PET膜, 2.9倍的改进; 结晶PET粉末(结晶度>40%), 5.3倍的改进. 就耐热性而言, 变体L32E/S113E/T237Q的半灭活温度(T₅₀⁶⁰)增加5.7 °C. 利用带耗散监测的石英晶体微量天平(QCM-D)实时监测PET水解过程, 以研究涂覆在金传感器上的PET解聚动力学. 构象动力学分析结果表明, 取代诱导了变体L32E/S113E/T237Q的构象变化, 其中主要构象使催化位点和PET之间的接触更紧密, 促进了PET水解. 总之, 本文展示了蛋白质工程中深度学习模型在识别和设计高效PET解聚酶方面的潜力.

关键词: 塑料热解, 聚对苯二甲酸乙二醇酯, TfCut2, 机器学习, 带耗散监测的石英晶体微量天平, 定向转化

Abstract:

A responsible and sustainable circular economy of polymers requires efficient recycling processes with a low CO₂ footprint. Enzymatic depolymerization of polyethylene terephthalate (PET) is a first step to make PET polymers a part of a circular economy of polymers. In this study, a structure-based deep learning model was utilized to identify residues in TfCut2 that are responsible for improved hydrolytic activity and enhanced stability. Machine learning guided design identified novel beneficial positions (L32E, S35E, H77Y, R110L, S113E, T237Q, R245Q, and E253H), which were evaluated and stepwise recombined yielding finally the beneficial variant L32E/S113E/T237Q. The latter TfCut2 variant exhibited improved PET depolymerization when compared with TfCut2 WT (amorphous PET film, 2.9-fold improvement // crystalline PET powder (crystallinity > 40%), 5.3-fold improvement). In terms of thermal resistance the variant L32E/S113E/T237Q showed a 5.7 °C increased half-inactivation temperature (T₅₀⁶⁰). The PET-hydrolysis process was monitored via a quartz crystal microbalance with dissipation monitoring (QCM-D) in real-time to determine depolymerization kinetics of PET coated onto the gold sensor. Finally, conformational dynamics analysis revealed that the substitutions induced a conformational change in the variant L32E/S113E/T237Q, in which the dominant conformation enabled a closer contact between the catalytic site and PET resulting in increased PET-hydrolysis. Overall, this study demonstrates the potential of deep learning models in protein engineering for identifying and designing efficient PET depolymerization enzymes.

Key words: Plastic depolymerization, Polyethylene terephthalate, TfCut2, Machine learning, QCM-D, Directed evolution

孟帅奇, 李忠玉, 张鹏, Francisca Contreras, 季宇, Ulrich Schwaneberg. 深度学习指导的热裂褐藻角质酶工程用于促进PET解聚[J]. 催化学报, 2023, 50: 229-238.

Shuaiqi Meng, Zhongyu Li, Peng Zhang, Francisca Contreras, Yu Ji, Ulrich Schwaneberg. Deep learning guided enzyme engineering of Thermobifida fusca cutinase for increased PET depolymerization[J]. Chinese Journal of Catalysis, 2023, 50: 229-238.

×
扫码分享

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.cjcatal.com/CN/10.1016/S1872-2067(23)64470-5

https://www.cjcatal.com/CN/Y2023/V50/I7/229

图/表 5

Fig. 1. The deep-learning-based prediction for performance enhancement of TfCut2 based on the Mutcompute methodology. (a) Within the Mutcompute methodology [14] the local environment of TfCut2 was firstly analyzed based on its crystal structure. After extracting the features of the microenvironment, the data is sent to a series of convolution and pooling layers, and then proceed by fully connected layers. As the outputs, each amino acid was assigned a probability that reflecting the chemical congruency. Top ten ranked substitutions were experimentally generated by site-directed mutagenesis. (b) Shows the positions at which the beneficial substitutions are predicted. The catalytic trial of TfCut2 is shown in orange stick, and the selected positions were shown in cyan stick.

Fig. 2. The enzymatic depolymerization efficiency of PET (nanoparticles, film, and powders) by TfCut2 variants. (a) Enzyme activity of TfCut2 variants towards PET nanoparticles (size < 100 nm, amorphous). The depolymerization products were qualitied by absorbance measurement on 240 nm. All formed products (TPA, MHET, and BHET) were considered as TPA and quantified using the TPA standard curve. (b) Enzyme activity of TfCut2 variants towards PET film (0.25 mm thickness, amorphous, from Goodfellow) at 65 and 70 °C. The products were identified and quantified by HPLC analysis. (c) Enzyme activity of TfCut2 variant towards PET powders (crystallinity > 40%, maximum particle size: 300 microns, from Goodfellow). The products were identified and quantified by HPLC analysis.

Table 1 The melting temperatures (Tm) and half-inactivation temperatures (T5060) of TfCut2 variants. The Tm values were determined by Prometheus NT.48 (NanoTemper Technologies GmbH, München, Germany). The Tm of TfCut2 enzymes was measured on the heating ramp trace as the midpoint value at half-height. The T5060 was measured by residual activity measurements against pNPB, and was determined at half of the maximum activity.

TfCut2 variants	T_m (°C)	T₅₀⁶⁰ (°C)
WT	72.8	64.8
L32E/H77Y	74.3	70.4
L32E/R110L	72.3	70.6
L32E/S113E	74.2	67.3
L32E/H77Y/R110L	74.5	68.9
L32E/S113E/T237Q	72.2	70.5

Fig. 3. The conformation dynamics of TfCut2 WT (a) and variant L32E/S113E/T237Q (b). Both two enzymes contain three metastable states. The percentage represents the equilibrium probability of the corresponding state. The arrows thickness represents the transition probability between each metastable, and the corresponding time on arrows indicates the transition time between metastable states. The zoom view showed the conformation of substrate PET2mer and the catalytic triad (S130, D176, H208).

Fig. 4. The substrate binding mode in the dominant conformation of TfCut2 WT (a) and variant L32E/S113E/T237Q (b). The left figure shown the substrate configuration in the TfCut2 enzyme, in which the residue S130 was shown in red. The right Fig. shown the enzyme-ligand interactions, in which the brick red spoked arcs represent the residues involved in hydrophobic interactions, and the green dotted lines indicate the hydrogen bond.

参考文献

[1]	R. Gao, H. Pan, J. Lian, Enzyme Microb. Technol. 2021, 150, 109868.
[2]	S. Ügdüler, K. M. Van Geem, R. Denolf, M. Roosen, N. Mys, K. Ragaert, S. De Meester, Green Chem. 2020, 22, 5376-5394.
[3]	Y. V. Soong, M. J. Sobkowicz, D. Xie, Bioenginerring, 2022, 9, 98.
[4]	M. MacLeod, H. P. H. Arp, M. B. Tekman, A. Jahnke, Science, 2021, 373, 61-65.
[5]	R. Wei, G. von Haugwitz, L. Pfaff, J. Mican, C. P. S. Badenhorst, W. Liu, G. Weber, H. P. Austin, D. Bednar, J. Damborsky, U. T. Bornscheuer, ACS Catal. 2022, 12, 3382-3396.
[6]	H. P. Austin, M. D. Allen, B. S. Donohoe, N. A. Rorrer, F. L. Kearns, R. L. Silveira, B. C. Pollard, G. Dominick, R. Duman, K. El Omari, V. Mykhaylyk, A. Wagner, W. E. Michener, A. Amore, M. S. Skaf, M. F. Crowley, A. W. Thorne, C. W. Johnson, H. L. Woodcock, J. E. McGeehan, G. T. Beckham, Proc. Natl. Acad. Sci. U. S. A. 2018, 115, E4350-E4357.
[7]	P. C. F. Buchholz, G. Feuerriegel, H. Zhang, P. Perez-Garcia, L. L. Nover, J. Chow, W. R. Streit, J. Pleiss, Proteins, 2022, 90, 1443-1456.
[8]	J. A. Verschoor, H. Kusumawardhani, A. F. J. Ram, J. H. de Winde, Front. Microbiol. 2022, 13, 821629.
[9]	Y. Yang, J. Min, T. Xue, P. Jiang, X. Liu, R. Peng, J. W. Huang, Y. Qu, X. Li, N. Ma, F. C. Tsai, L. Dai, Q. Zhang, Y. Liu, C. C. Chen, R. T. Guo, Nat. Commun. 2023, 14, 1645.
[10]	R.-J. Müller, H. Schrader, J. Profe, K. Dresler, W.-D. Deckwer, Macromol. Rapid Commun. 2005, 26, 1400-1405.
[11]	E. Herrero Acero, D. Ribitsch, G. Steinkellner, K. Gruber, K. Greimel, I. Eiteljoerg, E. Trotscha, R. Wei, W. Zimmermann, M. Zinn, A. Cavaco-Paulo, G. Freddi, H. Schwab, G. Guebitz, Macromolecules, 2011, 44, 4632-4640.
[12]	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. Guemard, M. Dalibey, J. Nomme, G. Cioci, S. Barbe, M. Chateau, I. Andre, S. Duquesne, A. Marty, Nature, 2020, 580, 216-219.
[13]	H. Lu, D. J. Diaz, N. J. Czarnecki, C. Zhu, W. Kim, R. Shroff, D. J. Acosta, B. R. Alexander, H. O. Cole, Y. Zhang, N. A. Lynd, A. D. Ellington, H. S. Alper, Nature, 2022, 604, 662-667.
[14]	R. Shroff, A. W. Cole, D. J. Diaz, B. R. Morrow, I. Donnell, A. Annapareddy, J. Gollihar, A. D. Ellington, R. Thyer, ACS Synth. Biol. 2020, 9, 2927-2935.
[15]	Z. Ding, G. Xu, R. Miao, N. Wu, W. Zhang, B. Yao, F. Guan, H. Huang, J. Tian, J. Hazard. Mater. 2023, 453, 131386.
[16]	N. Mohanan, Z. Montazer, P. K. Sharma, D. B. Levin, Front Microbiol. 2020, 11, 580709.
[17]	Y. Yang, J. Yang, L. Jiang, Science, 2016, 353, 759.
[18]	S. Meng, Y. Ji, L. Liu, M. D. Davari, U. Schwaneberg, ChemSusChem, 2021, 15, e202102434.
[19]	C. Roth, R. Wei, T. Oeser, J. Then, C. Föllner, W. Zimmermann, N. Sträter, Appl. Microbiol. Biotechnol. 2014, 98, 7815-7823.
[20]	L. Zheng, U. Baumann, J. L. Reymond, Nucleic Acids Res. 2004, 32, e115.
[21]	L. Pfaff, D. Breite, C. P. Badenhorst, U. T. Bornscheuer, R. Wei, Meth. Enzymol., 2021, 648, 253-270.
[22]	W. L. DeLano. Protein Cryst., 2002, 40: 82-92.
[23]	H. Land, M. S. Humble, Methods Mol. Biol. 2018, 1685, 43-67.
[24]	O. Trott, A. J. Olson, J. Comput. Chem. 2010, 31, 455-461.
[25]	Y. Duan, C. Wu, S. Chowdhury, M. C. Lee, G. Xiong, W. Zhang, R. Yang, P. Cieplak, R. Luo, T. Lee, J. Comput. Chem. 2003, 24, 1999-2012.
[26]	E. Krieger, R. L. Dunbrack, R. W. Hooft, B. Krieger, Comput. Drug Discovery Design, 2012, 819, 405-421.
[27]	D. Case, V. Babin, J. Berryman, R. Betz, Q. Cai, D. Cerutti, T. Cheatham III, T. Darden, R. Duke, H. Gohlke,AMBER 14, University of California, San Francisco, 2014, 1-826.
[28]	J. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman, D. A. Case, J. Comput. Chem. 2004, 25, 1157-1174.
[29]	A. Jakalian, D. B. Jack, C. I. Bayly, J. Comput. Chem. 2002, 23, 1623-1641.
[30]	P. Mark, L. Nilsson, J. Phys. Chem. A, 2001, 105, 9954-9960.
[31]	N. E. Siedhoff, U. Schwaneberg, M. D. Davari, Meth. Enzymol. 2020, 643, 281-315.
[32]	K. K. Yang, Z. Wu, F. H. Arnold, Nat. Methods, 2019, 16, 687-694.
[33]	U. T. Bornscheuer, B. Hauer, K. E. Jaeger, U. Schwaneberg, Angew. Chem. Int. Ed. 2019, 58, 36-40.
[34]	H. Cui, H. Cao, H. Cai, K. E. Jaeger, M. D. Davari, U. Schwaneberg, Chem. Eur. J. 2020, 26, 643-649.
[35]	N. Kress, J. M. Halder, L. R. Rapp, B. Hauer, Curr. Opin. Chem. Biol. 2018, 47, 109-116.
[36]	Z. Wu, S. B. J. Kan, R. D. Lewis, B. J. Wittmann, F. H. Arnold, Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 8852-8858.
[37]	A.-M. Illig, N. E. Siedhoff, U. Schwaneberg, M. D. Davari, bioRxiv, https://doi.org/10.1101/2022.06.07.495081.
[38]	J. A. Baath, K. Borch, K. Jensen, J. Brask, P. Westh, ChemBioChem, 2021, 22, 1627-1637.
[39]	B. Guo, S. R. Vanga, X. Lopez-Lorenzo, P. Saenz-Mendez, S. R. Ericsson, Y. Fang, X. Ye, K. Schriever, E. Bäckström, A. Biundo, R. A. Zubarev, I. Furo, M. Hakkarainen, P.-O. Syren, ACS Catal. 2022, 12, 3397-3409.
[40]	E. Erickson, T. J. Shakespeare, F. Bratti, B. L. Buss, R. Graham, M. A. Hawkins, G. Konig, W. E. Michener, J. Miscall, K. J. Ramirez, N. A. Rorrer, M. Zahn, A. R. Pickford, J. E. McGeehan, G. T. Beckham, ChemSusChem, 2022, 15, e202101932.
[41]	L. V. Sigolaeva, D. V. Pergushov, S. Y. Gladyr, I. N. Kurochkin, W. Richtering, Adv. Mater. Interfaces, 2022, 9, 2200310.
[42]	G. Yu, G. Liu, T. Liu, E. H. Fink, A. R. Esker, Biomacromolecules, 2023, 24, 566-575.
[43]	J. M. Sanchez-Ruiz, Biophys. Chem. 2010, 148, 1-15.
[44]	K. M. Polizzi, A. S. Bommarius, J. M. Broering, J. F. Chaparro-Riggers, Curr. Opin. Chem. Biol. 2007, 11, 220-5.
[45]	M. K. Scherer, B. Trendelkamp-Schroer, F. Paul, G. Perez-Hernandez, M. Hoffmann, N. Plattner, C. Wehmeyer, J. H. Prinz, F. Noe, J. Chem. Theory Comput. 2015, 11, 5525-42.
[46]	N. Plattner, F. Noe, Nat. Commun. 2015, 6, 7653.
[47]	M. M. Aboelnga, S. Kalyaanamoorthy, ACS Sustainable Chem. Eng. 2022, 10, 15857-15868.

深度学习指导的热裂褐藻角质酶工程用于促进PET解聚

Deep learning guided enzyme engineering of Thermobifida fusca cutinase for increased PET depolymerization

RichHTML

PDF

Supporting Information

可视化

摘要/Abstract

引用本文

使用本文

扫码分享

图/表 5

参考文献

相关文章 5

编辑推荐

Metrics

[1]	陈宇卓, 王浩, 陆冰, 易倪, 曹亮, 王勇, 毛善俊. 用于高精度催化性能预测的精细结构敏感型深度学习框架[J]. 催化学报, 2023, 50(7): 284-296.
[2]	王立晶, 杨天一, 冯博, 许祥雨, 申玉莹, 李孜涵, Arramel, 江吉周. 基于机器学习和第一性原理计算构建双电子转移通道加速CO₂光还原[J]. 催化学报, 2023, 54(11): 265-277.
[3]	李晓天, 陈林, 商城, 刘智攀. 调控炔烃半氢化反应的催化选择性: 实验和理论的最新进展[J]. 催化学报, 2022, 43(8): 1991-2000.
[4]	陈乐添, 张旭, 陈安, 姚赛, 胡绪, 周震. 利用机器学习靶向设计先进电催化剂[J]. 催化学报, 2022, 43(1): 11-32.
[5]	杜治平;姚洁;王公应*. 不同催化剂上碳酸乙烯酯与对苯二甲酸二甲酯的酯交换反应[J]. 催化学报, 2005, 26(8): 637-638.