Can constraint network analysis guide the identification phase of KnowVolution? A case study on improved thermostability of an endo-β-glucanase

Francisca Contreras; Christina Nutschel; Laura Beust; Mehdi D Davari; Holger Gohlke; Ulrich Schwaneberg

doi:10.1016/j.csbj.2020.12.034

Can constraint network analysis guide the identification phase of KnowVolution? A case study on improved thermostability of an endo-β-glucanase

Comput Struct Biotechnol J. 2020 Dec 28:19:743-751. doi: 10.1016/j.csbj.2020.12.034. eCollection 2021.

Authors

Francisca Contreras¹, Christina Nutschel², Laura Beust¹, Mehdi D Davari¹, Holger Gohlke^{2

3}, Ulrich Schwaneberg^{1

4}

Affiliations

¹ Institute of Biotechnology, RWTH Aachen University, Worringerweg 3, 52074 Aachen, Germany.
² John von Neumann Institute for Computing (NIC), Jülich Supercomputing Centre (JSC) and Institute of Biological Information Processing (IBI-7: Structural Biochemistry), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany.
³ Institute for Pharmaceutical and Medicinal Chemistry, Heinrich Heine University Düsseldorf, 40225 Düsseldorf, Germany.
⁴ DWI-Leibniz Institute for Interactive Materials, Forckenbeckstraße 50, 52074 Aachen, Germany.

Abstract

Cellulases are industrially important enzymes, e.g., in the production of bioethanol, in pulp and paper industry, feedstock, and textile. Thermostability is often a prerequisite for high process stability and improving thermostability without affecting specific activities at lower temperatures is challenging and often time-consuming. Protein engineering strategies that combine experimental and computational are emerging in order to reduce experimental screening efforts and speed up enzyme engineering campaigns. Constraint Network Analysis (CNA) is a promising computational method that identifies beneficial positions in enzymes to improve thermostability. In this study, we compare CNA and directed evolution in the identification of beneficial positions in order to evaluate the potential of CNA in protein engineering campaigns (e.g., in the identification phase of KnowVolution). We engineered the industrially relevant endoglucanase EGLII from Penicillium verruculosum towards increased thermostability. From the CNA approach, six variants were obtained with an up to 2-fold improvement in thermostability. The overall experimental burden was reduced to 40% utilizing the CNA method in comparison to directed evolution. On a variant level, the success rate was similar for both strategies, with 0.27% and 0.18% improved variants in the epPCR and CNA-guided library, respectively. In essence, CNA is an effective method for identification of positions that improve thermostability.

Keywords: AU, absorbance units; CMC, carboxymethyl cellulose; CNA, Constraint Network Analysis; Cellulase; Constraint network analysis; EGLII, endoglucanase II; GH5 endoglucanase; HTS, high-throughput screening; KnowVolution; MD, molecular dynamics; MTP, 96-well microtiter plates; PCR, polymerase chain reaction; Protein engineering; SSM, site-saturation mutagenesis; Thermostability.