Structural insights of RmXyn10A – A prebiotic-producing GH10 xylanase with a non-conserved aglycone binding region

Hydrolysis of arabinoxylan (AX) by glycoside hydrolase family 10 (GH10) xylanases produces xylo- and arabinoxylo-oligosaccharides ((A)XOS) which have shown prebiotic effects. The thermostable GH10 xylanase RmXyn10A has shown great potential to produce (A)XOS. In this study, the structure of RmXyn10A was investigated, the catalytic module by homology modelling and site-directed mutagenesis and the arrangement of its five domains by small-angle X-ray scattering (SAXS). Substrate specificity was explored in silico by manual docking and molecular dynamic simulations. It has been shown in the literature that the glycone subsites of GH10 xylanases are well conserved and our results suggest that RmXyn10A is no exception. The aglycone subsites are less investigated, and the modelled structure of RmXyn10A suggests that loop ß6?6 in the aglycone part of the active site contains a non-conserved ?-helix, which blocks the otherwise conserved space of subsite +2. This structural feature has only been observed for one other GH10 xylanase. In RmXyn10A, docking revealed two alternative binding regions, one on either side of the ?-helix. However, only one was able to accommodate arabinose-substitutions and the mutation study suggests that the same region is responsible for binding XOS. Several non-conserved structural features are most likely to be responsible for providing affinity for arabinose-substitutions in subsites +1 and +2. The SAXS rigid model of the modular arrangement of RmXyn10A displays the catalytic module close to the cell-anchoring domain while the carbohydrate binding modules are further away, likely explaining the observed lack of contribution of the CBMs to activity. © 2017 The AuthorsVINNOVA Svenska Forskningsrådet Formas: 2015-769 VINNOVA Sixth Framework Programme: RII3/CT/2004/5060008 Vetenskapsrådet: 2014-5038This work was supported by VINNOVA via the Lund University Antidiabetic Food Centre (VINN Excellence Centre), by the Swedish Research Council (VR) [grant no. 2014-5038 ], and by the Swedish Research Council Formas [grant no. 2015-769 ]. Also to enable us to use the DESY-EMBL beamlines we are grateful for financial support from the European Community – Research Infrastructure Action under the FP6 ?Structuring the European Research Area Programme” contract number RII3/CT/2004/5060008. The GROMACS simulations were performed on resources provided by the Swedish National Infrastructure for Computing (SNIC) at LUNARC (SNIC-2017/1-361). Björn Stenqvist, Lund University, is thanked for his computational assistance. Appendix