File Name : esi-figs1.tiff

Caption : figure s1 coomassie-stained 12% sds-page for q345f (a) in pcxp34h with an hexahistidine-tag at n-terminal and (b) in pet28b with the tag at c-terminal. changes of the conditions of production increased the amount and the purity of the enzymes. l: ladder; cf: crude fraction ; sf: soluble fraction; nrf: not retained fraction; wf1: first washing fraction with 10 mm imidazole- hcl; wf2: second washing fraction with 20 mm imidazole-hcl; efn: elution fraction with 250 mm imidazole-hcl, n = [0, 3]. elution fractions were deposed with a concentration of 2 μg of enzyme. (bacteria were grown overnight at 37°c in 10 ml lb- medium supplemented with 10% kanamycin (w/v). on the next day, 400 ml of lb autoinducible media containing 1% glucose (w/v) and the same antiobiotic concentration was incubated with 8 ml of overnight culture at 37°c for 6 h.)9

File Name : esi-figs2.tiff

Caption : figure s2 thermal denaturation assays of basp and its variants. thermostability depends on ph as showed in buffers mops-naoh 50 mm ph 8.0, citrate-naoh 100 mm ph 8.0, mops-naoh 50 mm ph 7.0 and hepes-naoh 25 mm ph 7.0 for (a) basp wt, (c) q345f and (e) q345f/p134d. thermostability depends on dmso concentration as showed in buffer mops-naoh 50 mm ph 8.0 for (b) basp wt, (d) q345f and (f) q345f/p134d. dashed lines are standard deviation curves. (3 μm enzyme, 5x sypro orange in buffers.)

File Name : esi-figs3.tiff

Caption : figure s3 hydrolytic activity of pnp-α-glc (plain lines) and pnp-β-glc (dashed lines) by basp wt, q345f and q345f/p134d. blanks contain no enzyme. the three enzymes are active on pnp-α-glc and inactive on pnp-β-glc, validating their stereoselectivy. (20 mm pnp-α-glc or pnp-β-glc, 3 μm enzyme in mops-naoh 50 mm ph 8.0.)