Screening of Enzymes

In order to expand the enzyme toolbox for the 1,6‐nucleophile addition, a set of PADs identified by a BLAST‐search17a with 48–75% sequence identity to FDC_Es (Supporting Information Figure S1, Table S1) possessing conserved catalytically relevant residues for hydration and carboxylation (Tyr19, Tyr21, Glu72 and Arg49)19 was tested.

Among the first group of nucleophiles (Table 1, entries 1–5), methoxyamine (2a), cyanide (2b) and n‐propanethiol (2c) were readily accepted by FDC_Es (Table 1), while 2d and 2e were less promising due to low reproducibility (2d) or due to spontaneous background reactions (2q). Hence, the former were chosen for the enzyme screening (Scheme 3, for the screening with 2c see the Supporting Information, Table S4). All tested decarboxylases catalyzed the formation of nucleophile adducts 3a and 3b beside the minor hydration product 4 (Table 2) with the single exception of PAD_Mc and 2a (Table 2, entry 6). However, distinct variations in conversion and optical purity of the products (S)‐3a and (S)‐3b were noted.

Open in a separate window

Scheme 3

Conjugate 1,6‐addition of methoxyamine 2a and cyanide 2b.

Table 2

Enzyme screening for the addition of 2a and 2b onto 1.^[a]

Nucleophile		2a						2b
Enzyme	Entry	3a [%]	ee [%]	4 [%]	ee [%]		Entry	3b [%]	ee [%]	4 [%]	ee [%]
FDC_Es	1	93	17	6	39		8	82	64	4	n.d.
PAD_Lp	2	48	3	6	10		9	5	68	<1	n.d.
PAD_Ba	3	11	rac	4	n.d.		10	25	88	1	n.d.
PAD_Ll	4	42	5	6	24		11	29	76	<1	n.d.
PAD_Ms	5	15	8	2	n.d.		12	23	85	<1	n.d.
PAD_Mc	6	<1	n.d.	3	n.d.		13	6	71	8	89
PAD_Ps	7	73	10	8	26		14	57	91	<1	n.d.

Open in a separate window

^[a] Abbreviations: FDC_Enterobacter sp. (Es), PAD_Pantoea sp. (Ps), PAD_Mycobacterium columbiense (Mc), PAD_Methylobacterium sp. (Ms), PAD_Lactobacillus plantarum (Lp), PAD_Lactococcus lactis (Ll) and PAD_Bacillus amyloliquefaciens (Ba). Screening conditions: lyophilized E. coli cells (20 mgmL⁻¹) containing the heterologously expressed FDC or PAD, 1 (10 mM), 2a or 2b (100 mM) in KP_i buffer (50 mM, pH 7.0); incubation for 24 h at 30°C and 700 rpm; n.d.=not determined due to low conversion.

In the addition of methoxyamine (2a), FDC_Es showed the best results, both in terms of conversion and enantioselectivity (17% ee) (Table 2, entry 1). PAD_Lp, PAD_Ll and PAD_Ps showed moderate conversions of 42–73% (Table 2, entries 2, 4 and 7) and poor optical purity of (S)‐3a (ee ≤10%). Hydration was largely reduced and showed only moderate ees of 4 (max. 39% with FDC_Es).

Similarly, addition of cyanide (2b) proceeded with moderate to good conversion, but stereoselectivities were generally much better with all enzymes (Table 2, entries 1–7 vs. entries 8–14). FDC_Es and PAD_Ps performed best in terms of conversion (Table 2, entries 8 and 14) and the latter enzyme also exhibited superior stereoselectivity in the addition of cyanide (91% ee, Table 2, entry 14) thus promoting it as a promising candidate for further investigations. Again, with 2b hydration was only a minor side reaction (max. 8% with PAD_Mc).

The absolute configuration of products 3a–3c was determined by comparison with authentic reference material (3a), comparison of optical rotation values (3b) and CD spectroscopy (3c) as described in the Supporting Information (Table S5). Overall, a strong preference for the formation of the (S)‐product is congruent for all enzymes.

Given the comparatively high sequence identity of 73% between PAD_Ps and FDC_Es (other PADs show 48–52% identity, Supporting Information, Table S1) it is plausible that these two enzymes also perform similarly in the addition of nucleophiles.

In addition to wild‐type enzymes, FDC_Es variants (Figure 1) were tested with nucleophiles 2a and 2b (Table 3). Mutants designed to provide more space in the active site led to considerably less conversion (L80A, L80A/V78A; Table 3, entries 3, 4, 12 and 13) or were inactive at all (L80A/V78A/W70V, L80A/V78A/ W70L, L80A/V78A/W70L/V46A; data not shown). However, exchange of Ile to Ala in position 41 appeared to be beneficial (Figure 1) since with both nucleophiles, the conversion was not adversely affected, but the ee was significantly enhanced (entries 2 and 11).

Open in a separate window

Figure 1

Active‐site of FDC_Es (PDB‐ID: 4UU3)28 with p‐vinylphenol (1) docked (green, docking performed with UCSF Chimera);29 residues targeted by mutagenesis for improvement of catalysis are highlighted in orange. (Putative) catalytic key residues are marked with an asterisk.

Table 3

FDC_Es mutants in the addition of 2a and 2b.^[a]

Nucleophile	2a						2b
FDC_Es variant	Entry	3a [%]	ee [%]	4 [%]	ee [%]		Entry	3b [%]	ee [%]	4 [%]	ee [%]
wt	1	95	17	5	39		10	82	64	<1	n.d.
I41A	2	96	36	4	rac		11	81	81	<1	n.d.
L80A	3	71	rac	3	rac		12	23	72	<1	n.d.
L80A/V78A	4	10	rac	<1	n.d.		13	53	85	<1	n.d.
Y27F	5	94	8	6	rac		14	55	82	<1	n.d.
Y39F	6	95	rac	5	12		15	16	72	<1	n.d.
Y19F	7	<1	n.d.	<1	n.d.		16	<1	n.d.	<1	n.d.
Y21F	8	<1	n.d.	<1	n.d.		17	3	n.d.	<1	n.d.
E72A	9	8	rac	<1	n.d.		18	14	14	<1	n.d.

Open in a separate window

^[a] Conditions: see Table 2; n.d.=not determined due to low conversion.

In order to evaluate the relevance of tyrosine residues 27 and 39, which are flanking the substrate, the corresponding Phe‐variants were prepared (Y27F, Y39F). With 2a, the activity and selectivity were not affected (entries 6 and 7) but with 3b conversion dropped significantly going in hand with enhanced selectivities (entries 14 and 15).

Changing Glu72 (responsible for CO₂‐activation in the carboxylation reaction) to alanine or either of Tyr19 and Tyr21 (for deprotonation of the phenolic OH)19b to phenylalanine completely abolished the activity for nucleophile addition as well as hydration (Table 3, entries 7–9 and 16–18), which underlines their crucial role in catalysis.

Preparative‐Scale Biotransformation

In order to fully characterize products 3a and 3b and to evaluate the applicability of this biotransformation for the preparative scale, reactions were performed with 50 to 100 mg substrate after optimization of the reaction conditions. Promising initial results (Table 1) and conversions of up to 73% in an enzyme screening (Supporting Information, Table S4) encouraged us to include also propanethiol (2c) in the up‐scales.

Given the heterogeneity of pK _a values of the nucleophiles and their pH‐dependent reactivity, a detailed pH‐study was performed and the maximum of substrate‐ and nucleophile‐loading was evaluated. Dimethoxyethane (DME) was identified as a suitable co‐solvent for water‐insoluble nucleophile 2c (Supporting Information, Section 5). Optimal results are listed in Table 4. After isolation and purification of the nucleophile adducts (3a–c) (Table 4), the absolute configuration of all products was determined to be (S) (Supporting Information, Table S5) and hence nicely matched the stereoselectivity of the biocatalytic hydration.17a

Table 4

Preparative‐scale biotransformation, isolation and characterization of products.^[a]

	FDC_Es variant	Yield [%]	Yield [mg]	ee [%]	[α]$\genfrac{}{}{0pt}{}{20}{\mathrm{D}}$
3a	I41A	71[b]	71	22 (S)	−6.2°
3b	wt	71[c]	44	85 (S)	−12°
3c	wt	56[d]	56	81 (S)	−156°

Open in a separate window

^[a] Conditions: 20 mgmL⁻¹ lyophilized E. coli whole cells with heterologously expressed FDC_Es variants in KP_i‐buffer (100 mM).

^[b] 1 (20 mM), 2a (5 equiv.), pH 7.0.

^[c] 1 (10 mM), 2b (10 equiv.), pH 6.0.

^[d] 1 (10 mM), TAPS‐buffer pH 8.5, DME (10% v/v), 2c (10 equiv.).

Preparation of Biocatalysts: Cloning and Heterologous Expression

The genes encoding for the respective PADs and FDC were transformed in E. coli BL21(DE3) and heterologously expressed as described previously.17a

Site‐Directed Mutagenesis

Site‐directed mutagenesis was carried out with the QuikChange PCR mutagenesis kit from Stratagene using the respective primer sequences listed in the Supporting Information, Table S3.

Nucleophile Screening

Lyophilized E. coli whole cells containing heterologously expressed FDC_Es (20 mg) were rehydrated for 30 min at 700 rpm shaking in KP_i buffer (887 μL; 100 mM, pH 7.6) in 2.0 mL reaction vials. A stock solution containing the respective nucleophile in either buffer or 1,2‐dimethoxyethane (DME), depending on the compound's solubility (100 μL, 1M), was added to the cell suspension followed by short mixing and addition of the substrate 4‐vinylphenol 1 (13.4 μL of a 8.6% w/w solution in propylene glycol). The mixture was incubated for 24 h at 30°C and 700 rpm in an Eppendorf Thermoshaker. Then the mixture was split into two equal aliquots á 500 μL. One aliquot was extracted with ethyl acetate (2×500 μL) and after drying over MgSO₄ the organic phase was subjected to GC‐MS analysis. The other half was diluted with acetonitrile containing anisole as internal standard (10 mM) for quantification of substrate 1 and hydrate 4 using HPLC. Putative adducts were identified by means of MS fragmentation patterns.

Enzyme Screening

Lyophilized E. coli whole cells containing the heterologously expressed PAD (wild‐type or mutants, 20 mg) were rehydrated for 15 min at 700 rpm in KP_i buffer (50 mM) in 2.0 mL reaction vials containing 100 mM of methoxyamine 2a (1 mL, pH 7.0). Substrate 4‐vinylphenol 1 (13.4 μL of a 8.6% w/w solution in propylene glycol) was added and the mixture was incubated for 24 h at 30°C and 700 rpm in an Eppendorf Thermoshaker. The reaction was stopped by the addition of acetonitrile (1 mL) containing anisole as internal standard (10 mM), vortexing and centrifugation (14000 rpm, 10 min). An aliquot (1 mL) was withdrawn for quantitative aqueous HPLC analytics. The remaining aliquot was extracted with ethyl acetate (2×500 μL) and after drying over MgSO₄ the combined organic extracts were evaporated to dryness in an air stream, re‐dissolved in 2‐PrOH (100 μL), diluted with n‐heptane (900 μL) and subjected to chiral analytics with organic HPLC.

Preparative‐Scale Biotransformations

General procedure for the FDC‐catalyzed preparative‐scale addition of 2a–2c onto 1: Lyophilized E. coli whole‐cells containing the heterologously expressed FDC_Es wild‐type or I41A mutant (20 mgmL⁻¹) were rehydrated in KP_i buffer for 30 min and 120 rpm at 30°C in a cultivation shaker in a 50 mL Falcon tube. Nucleophiles 2a–2c were supplemented from a stock solution in reaction buffer or DME (1M) followed by the addition of substrate 1 (8.6% w/w solution in propylene glycol), the mixture was incubated at 30°C and 120 rpm for the denoted time. The reaction was stopped by extraction with ethyl acetate (3×20 mL), the combined organic extracts were dried over Na₂SO₄ and the solvent was removed under reduced pressure.

(S)‐4‐(1‐Methoxyamino)ethylphenol (3a): Substrate 1 (803.4 μL stock solution, 0.6 mmol, 20 mM) was reacted with 2a (2.99 mL stock solution in buffer, 5 equiv., 100 mM) in KP_i buffer (50 mM, pH 7.0, 26.1 mL) for 26 h. Pure product 3a was obtained after column chromatography on silica gel [petroleum ether/ethyl acetate 7:3, R _f=0.30 (product 3a)] as an oil; yield: 70.6 mg (71%). ¹H NMR (300 MHz, CDCl₃): δ=7.24–7.16 (m, 2H), 6.78–6.73 (m, 2H), 4.09 (q, ³ J ₁=6.6 Hz, 1H), 3.50 (s, 3H), 1.36 (d, ³ J ₁=6.6 Hz, 3H); ¹³C NMR (75 MHz, CDCl₃): δ=155.3, 134.3, 128.6 (2C), 115.5 (2C), 62.5, 59.9, 19.6; GC‐MS (ESI⁺, 70 eV): t_R=8.70, m/z=167.1 [M⁺]; HR‐MS (CI⁺): m/z=168.10173 [MH⁺] (calcd. 168.10191); [α]$\genfrac{}{}{0pt}{}{20}{\mathrm{D}}$ : −6.2 (c 1, CHCl₃, ee=22%, [S]).

(S)‐2‐(4‐Hydroxyphenyl)propanenitrile (3b): Substrate 1 (457 μL stock solution, 0.33 mmol, 10 mM) was reacted with 2b (3.4 mL stock solution in buffer, 10 equiv., 100 mM) in KP_i buffer (50 mM, pH 6.0, 30 mL) for 40 h. Pure product 3b was obtained after column chromatography on silica gel [petroleum ether/ethyl acetate 7:3, R _f=0.57 (product 3b)] as a colorless oil that solidified upon storage at 4°C; yield: 43.8 mg (89%). ¹H NMR (300 MHz, CDCl₃): δ=7.22–7.18 (m, 2H), 6.86–6.81 (m, 2H), 5.58 (bs, 1H), 3.85 (q, ³ J ₁=7.2 Hz, 1H), 1.61 (d, ³ J ₁=7.2 Hz, 3H); ¹³C NMR (75 MHz, CDCl₃): δ=155.8, 128.8, 128.1 (2C), 122.1, 116.1 (2C), 30.6, 21.5; GC‐MS (ESI⁺, 70 eV): t_R=8.78 min, m/z=147.1 [M⁺]; HR‐MS (CI⁺): m/z=148.07555 [MH⁺] (calcd. 148.07569); [α]$\genfrac{}{}{0pt}{}{20}{\mathrm{D}}$ : −12.5 (c 1, CHCl₃, ee=85%, [S]).

(S)‐4‐[1‐(Propylthio)ethyl]phenol (3c): Substrate 1 (684 μL stock solution, 0.5 mmol, 10 mM) was reacted with 2c (5.0 mL stock solution in DME, 10 equiv., 100 mM) in TAPS buffer (100 mM, pH 9, 45 mL) for 24 h. Pure product 3c was obtained after column chromatography on silica gel [petroleum ether/ethyl acetate 7:3, R _f=0.57 (product 3c co‐elutes with 1!)] as a colorless oil; yield: 56 mg (58%). ¹H NMR (300 MHz, CDCl₃): δ=7.23–7.18 (m, 2H), 6.80–6.75 (m, 2H), 5.03 (bs, 1H), 3.91 (q, ³ J ₁=6.9 Hz, 1H), 2.36–2.21 (m, 2H), 1.56–1.45 (m, 5H), 0.91 (t, ³ J ₁=7.5 Hz, 3H); ¹³C NMR (75 MHz, CDCl₃): δ=154.4, 136.5, 128.6 (2C), 115.4 (2C), 43.5, 33.3, 22.8, 13.7; GC‐MS (ESI⁺, 70 eV): t_R=10.75 min, m/z=196.1 [M⁺]; HR‐MS (CI⁺): m/z=197.09602 [MH⁺] (calcd. 197.09946); [α]$\genfrac{}{}{0pt}{}{20}{\mathrm{D}}$ : −156.2 (c 1, CHCl₃, ee=81%, [S]).

Funding by the Austrian Science Fund FWF (project P 26863) is gratefully acknowledged. This work has been supported by the Austrian BMWFJ, BMVIT, SFG, Standortagentur Tirol and ZIT through the Austrian FFG‐COMET‐Funding Program. Byung‐Gee Kim (Seoul, South Korea) is cordially thanked for the generous donation of phenolic acid decarboxylase plasmids, Klaus Ditrich (BASF SE, Ludwigshafen, Germany) kindly provided reference material. We further thank Klaus Zangger, Tea Pavkov‐Keller and Kenneth Jensen for NMR, CD and HR‐MS measurements.