Letter reSeArCH
Formation of the fluorescein product was monitored over 30 min by absorbance Fluorescein 4-fluorophenylacetate (3). The crude product was purified by flash
at 500 nm using a CLARIOstar plate reader (BMG Labtech). The most active column chromatography (1:1 ethyl acetate: cyclohexane) to give the product as a
clones from each round were rescreened in triplicate. Expression and screening yellow oil (73 mg, 29%). Electron ionization mass spectrometry (EI-MS) m/z = 469
+
1
was performed as described above but from glycerol stocks prepared from the [M + H] . H NMR (400 MHz, CDCl3) δ 8.02 (1H, d, J = 7.2 Hz, CH), 7.64
original overnight cultures. (2H, m, 2 × CH), 7.35 (2H, dd, J = 8.4 Hz, 5.4 Hz, 2 × CH), 7.07 (4H, m, 4 × CH),
Steady-state kinetic assays. Kinetic assays were performed using fluorescein 6.76 (2H, m, 2 × CH), 6.67 (1H, d, J = 2.2 Hz, CH), 6.58 (1H, d, J = 8.7 Hz, CH),
13
2
2 3
-phenylacetate (5–100 µM) and enzyme (concentration of 1 µM for BH32, 6.51 (1H, dd, J = 8.4, 2.3 Hz, CH), 3.86 (2H, s, CH ); C NMR (100 MHz, CDCl )
BH32(His23Ala), OE1, OE1.1, OE1.3(Tyr45Phe) and OE1.3(Asn46Glu); 0.5 µM δ 169.9, 169.7, 162.2 (d, J = 245 Hz, C–F), 158.0, 153.0, 152.1, 151.8, 151.8, 135.3,
for OE1.2 and OE1.3; and 0.2 µM for CUT190) in PBS pH 7.4 with 3% DMSO as 130.9 (d, J = 8 Hz, C–F), 129.9, 129.2, 129.1, 128.6 (d, J = 3 Hz, C–F), 126.3, 125.1,
a co-solvent. Initial rates were measured at 500 nm using a Cary 50Bio UV-vis 124.0, 117.1, 116.7, 115.7 (d, J = 22 Hz, C–F), 112.6, 110.5, 110.2, 103.1, 82.9, 40.4.
spectrophotometer (Varian) at 22 °C, using a Δε500 (change in molar extinction Fluorescein 2-thiopheneacetate (4). The crude product was purified by flash
coefficient at 500 nm) of 43.5 mM−1 cm (as determined from standard curves column chromatography (1:1 ethyl acetate:cyclohexane) to give the product as
−1
+
1
of fluorescein and fluorescein 2-phenylacetate) to convert absorbance into prod- a yellow oil (116 mg, 48%). EI-MS m/z = 457 [M + H] . H NMR (400 MHz,
3
uct concentration. The data were corrected for the buffer-catalysed background CDCl ) δ 8.03 (1H, d, J = 7.2 Hz, CH), 7.65 (2H, m, 2 × CH), 7.28 (1H, m, CH),
reaction measured under the same conditions (Extended Data Fig. 5d). Assays with 7.12 (1H, d, J = 7.3 Hz, CH), 7.06 (2H, m, 2 × CH), 7.01 (1H, m, CH), 6.79
BH32, OE1 and variants were performed in triplicate and the averaged initial rates (2H, s, 2 × CH), 6.69 (1H, d, J = 2.3 Hz, CH), 6.60 (1H, d, J = 8.4 Hz, CH), 6.52
1
3
were fitted to the Michaelis–Menten equation using Origin software (Extended (1H, dd, J = 8.7, 2.4 Hz, CH), 6.30 (1H, br s, OH), 4.11 (2H, s, CH
2
); C NMR
Data Fig. 3). CUT190 assays were also performed in triplicate but the averaged (100 MHz, CDCl
3
) δ 170.0, 168.8, 158.1, 153.0, 152.1, 151.8, 151.7, 135.3, 133.8,
initial rates were fitted to a Hill equation using the Origin software (Extended 129.9, 129.2, 129.1, 127.3, 127.0, 126.3, 125.5, 125.1, 124.0, 117.1, 116.7, 112.6,
Data Fig. 3). Representative time courses for the hydrolysis of 100 µM fluorescein 110.4, 110.2, 103.1, 83.0, 35.5.
2-phenylacetate are shown in Figs. 1b, 2c.
Fluorescein naphthalen-2-yl-acetate (5). The crude product was purified by flash
Kinetic assays were performed using single enantiomers of fluorescein 2-phe- column chromatography (1:1 ethyl acetate:cyclohexane) to give the product as
+
1
nylpropanoate (5–60 µM for OE1.3 and 1.25–60 µM for OE1.4) and enzyme a yellow oil (106 mg, 40%). EI-MS m/z = 501 [M + H] . H NMR (400 MHz,
concentration of 1 µM for OE1.3 and 0.5 µM for OE1.4) in PBS pH 7.4 with 3% CDCl ) δ 8.02 (1H, d, J = 6.7 Hz, CH), 7.85 (4H, m, 4 × CH), 7.62 (2H, m, 2 ×
DMSO as a co-solvent. Initial rates were measured at 500 nm at 22 °C using a CH), 7.50 (3H, m, 3 × CH), 7.07 (1H, d, J = 6.9 Hz, CH), 7.03 (1H, s, CH), 6.76
(
3
−1
−1
Δε500 = 47.0 mM cm (as determined from standard curves of fluorescein and (2H, s, 2 × CH), 6.64 (1H, d, J = 2.3 Hz, CH), 6.57 (1H, d, J = 8.7 Hz, CH), 6.52
1
3
fluorescein 2-phenylpropanoate) to convert absorbance into product concentra- (1H, br s, OH), 6.49 (1H, dd, J = 8.7, 2.5 Hz, CH), 4.05 (2H, s, CH
2
); C NMR
tion. Averaged initial rates for OE1.3 were fitted to a Michaelis–Menten equation (100 MHz, CDCl ) δ 170.0, 170.0, 158.1, 153.0, 152.0, 151.8, 151.8, 135.3, 133.4,
3
and the rates for OE1.4 were fitted to a Michaelis–Menten equation with substrate 132.6, 130.3, 129.9, 129.1, 129.1, 128.5, 128.2, 127.7, 127.7, 127.1, 126.3, 126.3,
inhibition (y = kcat × x / (K
M
i
+ x(1 + x/K ))) using Origin software (Extended 126.1, 125.1, 124.0, 117.2, 116.6, 112.6, 110.3, 110.3, 103.1, 83.1, 41.5.
Data Fig. 4).
General method for the synthesis of fluorescein esters from the corresponding
To determine the rate constant of fluorescein 2-phenylactetate hydrolysis cata- carboxylic acid. Acid (1 mmol, 1 equiv.) was dissolved in dry acetonitrile (10 ml),
lysed by the nucleophiles Me-His, DMAP and N-methylimidazole, reactions were dicyclohexylcarbodiimide (1.1 mmol, 1.1 equiv.) was added and the mixture was
performed using a fixed concentration of fluorescein 2-phenylactetate (50 µM) cooled to 0 °C. Fluorescein (1.2 mmol, 1.2 equiv.) was added and the reaction
and varying concentrations of nucleophile (0.5–1 mM). Linear fits of averaged mixture was stirred for 15 min at 0 °C, then at room temperature overnight. The
initial rates versus nucleophile concentration were used to derive bimolecular rate reaction was diluted with ethyl acetate (70 ml), washed with 1 M HCl (2 × 50 ml)
constants kMe-His, kDMAP and kNMI (Extended Data Fig. 5).
and then brine (1 × 50 ml). The organic layer was dried over MgSO4, filtered and
Substrate profiling of OE1 and OE1.3. Assays were performed using fluores- the solvent removed in vacuo to yield the crude product.
cein esters 2–7 (100 µM) and purified enzyme in PBS pH 7.4 with 3% DMSO as Fluorescein 2-furanacetate (6). The crude product was purified by flash column
a co-solvent. Initial rates were measured at 500 nm using a Cary 50Bio UV-vis chromatography (1:1 ethyl acetate:cyclohexane) to give the product as a yellow
+
1
spectrophotometer at 22 °C, using a Δε500 to convert absorbance into product oil (24 mg, 6%). EI-MS m/z = 441 [M + H] . H NMR (400 MHz, CDCl3) δ 8.03
concentration. The data was corrected for the buffer-catalysed background reac- (1H, d, J = 7.1, CH), 7.66 (2H, m, 2 × CH), 7.42 (1H, d, J = 1.0 Hz, CH), 7.16 (1H,
tion measured under the same conditions. For details of enzyme concentrations, d, J = 7.5, CH), 7.10 (1H, s, CH), 6.80 (2H, d, J = 1.1 Hz, 2 × CH), 6.73 (1H, d,
extinction coefficients and the rates of background hydrolysis for each substrate, J = 2.4 Hz, CH), 6.65 (1H, d, J = 8.6 Hz, CH), 6.55 (1H, dd, J = 8.4, 2.4 Hz, CH),
13
see Extended Data Table 1.
2 3
6.37 (2H, m, 2 × CH), 3.96 (2H, s, CH ); C NMR (100 MHz, CDCl ) δ 169.6,
General method for the synthesis of fluorescein esters from the corresponding 167.5, 157.8, 153.0, 152.1, 151.8, 151.7, 146.5, 142.5, 135.2, 129.9, 129.3, 129.1,
33
acid chloride . Fluorescein sodium salt (0.53 mmol, 1 equiv.) and zinc chloride 126.4, 125.1, 124.0, 117.2, 116.8, 112.5, 110.9, 110.7, 110.3, 108.6, 103.1, 82.6, 34.2.
(
0.13 mmol, 0.25 equiv.) were stirred in dry DMF (10 ml) at room temperature. (R)- and (S)-Fluorescein 2-phenylpropanoate (7). The crude products were
The corresponding acid chloride (0.53 mmol, 1 equiv.) was added dropwise and purified by flash column chromatography (1:1 ethyl acetate:cyclohexane) to give
the reaction was stirred overnight at room temperature. The reaction mixture was (R)-7 (42 mg, 9%) and (S)-7 (51 mg, 11%) as yellow oils. EI-MS m/z = 465 [M +
+
1
diluted with ethyl acetate (70 ml) and washed with 1 M HCl (2 × 50 ml) and brine H] . H NMR (400 MHz, CDCl
3
) δ 8.01 (1H, d, J = 7.5 Hz, CH), 7.64 (2H, m, 2
(
1 × 50 ml). The organic layer was dried over MgSO4, filtered and the solvent × CH), 7.34 (5H, m, Ph), 7.07 (1H, m, CH), 6.96 (1H, m, CH), 6.76–6.63 (3H, m,
removed in vacuo to give the crude product.
3 × CH), 6.58 (1H, dd, J = 8.7 Hz, 1.2 Hz, CH), 6.51 (1H, m, CH), 3.99 (1H, q,
13
3 3
Fluorescein 2-phenylacetate (1). The crude product was purified by flash column J = 7.1 Hz, CH), 1.63 (3H, dd, J = 7.2, 1.5 Hz, CH ); C NMR (100 MHz, CDCl )
chromatography (1:1 ethyl acetate:cyclohexane) to give the product as a yellow δ 173.0, 169.9, 158.1, 153.0, 152.1, 151.9, 151.8, 139.5, 135.3, 129.9, 129.1, 129.0,
+
1
oil (118 mg, 49%). EI-MS m/z = 451 [M + H] . H NMR (400 MHz, CDCl3) 128.9, 128.9, 127.6, 127.5, 127.5, 126.3, 125.1, 124.0, 117.1, 117.1, 116.5, 112.6,
δ 8.02 (1H, d, J = 7.8 Hz, CH), 7.63 (2H, m, 2 × CH), 7.35 (5H, m, Ph), 7.09 (1H, 110.4, 110.2, 110.2, 103.1, 83.0, 45.6, 45.6, 18.4, 18.4.
d, J = 6.9 Hz, CH), 7.03 (1H, d, J = 1.3 Hz, CH), 6.76 (2H, m, 2 × CH), 6.66 (1H, Crystallization, refinement and model building. Crystals of OE1 variants were
−
1
d, J = 2.4 Hz, CH), 6.57 (1H, d, J = 8.7 Hz, CH), 6.50 (1H, dd, J = 8.7, 2.4 Hz, CH), prepared by mixing 200 nl of 10 mg ml protein in PBS buffer pH 7.4 with
13
3
1
1
.89 (2H, s, CH
2 3
); C NMR (100 MHz, CDCl ) δ 170.0, 170.0, 158.1, 153.0, 152.1, equal volumes of precipitant. All trials were conducted by sitting-drop vapour
51.8, 151.8, 135.3, 132.9, 129.9, 129.3, 129.1, 129.1, 128.8, 127.5, 126.3, 125.1, diffusion and incubated at 4 °C. Protein crystallization conditions are given in
24.0, 117.2, 116.6, 112.6, 110.4, 110.3, 103.1, 83.0, 41.3.
Supplementary Table 2. All crystals were cryoprotected by the addition of 10% PEG
Fluorescein 4-methoxyphenylacetate (2). The crude product was purified by 200 to the mother liquor and flash-cooled in liquid nitrogen. Data were collected
flash column chromatography (1:1 ethyl acetate: cyclohexane) to give the product from single crystals at Diamond Light Source and subsequently scaled and reduced
+
1
as a yellow oil (45 mg, 18%). EI-MS m/z = 481 [M + H] . H NMR (400 MHz, with Xia2. Preliminary phasing was performed by molecular replacement in Phaser
CDCl3) δ 8.02 (1H, d, J = 7.1 Hz, CH), 7.64 (2H, m, 2 × CH), 7.29 (2H, d, J = 8.6 using a search model derived from wild-type BH32 (PDB code: 2UW6). Iterative
34
Hz, 2 × CH), 7.12 (1H, d, J = 7.3 Hz, CH), 7.04 (1H, d, J = 1.6 Hz, CH), 6.91 cycles of rebuilding and refinement were performed in COOT and Phenix.refine ,
(
2H, d, J = 8.6 Hz, 2 × CH), 6.76 (2H, m, 2 × CH), 6.69 (1H, d, J = 2.3 Hz, CH), respectively. Structure validation with MolProbity and PDBREDO were integrated
6
.61 (1H, d, J = 8.8 Hz, CH), 6.52 (1H, dd, J = 8.9, 2.3 Hz, CH), 5.65 (1H, br s, into the iterative rebuild and refinement process. Complete data collection and
13
OH), 3.82 (3H, s, CH
3
), 3.82 (2H, s, CH
2
); C NMR (100 MHz, CDCl
3
) δ 170.0, refinement statistics can be found in the Supplementary information. Coordinates
1
1
4
69.5, 159.0, 157.7, 153.0, 152.2, 151.9, 151.8, 135.2, 130.3, 129.9, 129.3, 129.0, and structure factors have been deposited in the Protein Data Bank under
26.4, 125.1, 125.0, 124.0, 117.3, 116.7, 114.2, 112.5, 111.0, 110.3, 103.1, 82.6, 55.3, accession numbers 6Q7N, 6Q7O, 6Q7P, 6Q7Q and 6Q7R. In silico mutagenesis
0.5. was performed using ICM Pro to derive model of engineered OE1.4.