Protocol | Published:

Library design and screening protocol for artificial metalloenzymes based on the biotin-streptavidin technology

Nature Protocols volume 11, pages 835852 (2016) | Download Citation


Artificial metalloenzymes (ArMs) based on the incorporation of a biotinylated metal cofactor within streptavidin (Sav) combine attractive features of both homogeneous and enzymatic catalysts. To speed up their optimization, we present a streamlined protocol for the design, expression, partial purification and screening of Sav libraries. Twenty-eight positions have been subjected to mutagenesis to yield 335 Sav isoforms, which can be expressed in 24-deep-well plates using autoinduction medium. The resulting cell-free extracts (CFEs) typically contain >1 mg of soluble Sav. Two straightforward alternatives are presented, which allow the screening of ArMs using CFEs containing Sav. To produce an artificial transfer hydrogenase, Sav is coupled to a biotinylated three-legged iridium pianostool complex Cp*Ir(Biot-p-L)Cl (the cofactor). To screen Sav variants for this application, you would determine the number of free binding sites, treat them with diamide, incubate them with the cofactor and then perform the reaction with your test compound (the example used in this protocol is 1-phenyl-3,4-dihydroisoquinoline). This process takes 20 d. If you want to perform metathesis reactions, Sav is coupled to a biotinylated second-generation Grubbs-Hoveyda catalyst. In this application, it is best to first immobilize Sav on Sepharose-iminobiotin beads and then perform washing steps. Elution from the beads is achieved in an acidic reaction buffer before incubation with the cofactor. Catalysis using your test compound (in this protocol, 2-(4-(N,N-diallylsulfamoyl)phenyl)-N,N,N-trimethylethan-1-aminium iodide) is performed using the formed metalloenzyme. Screening using this approach takes 19 d.


In recent years, ArMs have attracted increasing attention as alternatives to homogeneous catalysts (such as Rh-based hydrogenation catalysts or Os-based dihydroxylation catalysts) and naturally occurring enzymes (including variants produced by standard mutations)1,2. Such hybrid catalysts result from the incorporation of an artificial cofactor within a host protein, and they combine attractive features of both systems. Several anchoring strategies have been pursued to ensure localization of the metal moiety within a well-defined protein environment: dative, covalent and supramolecular3.

In this context, the biotin-(strept)avidin system has been proven to be particularly versatile for the creation of ArMs4. Indeed, the hemispheric biotin-binding vestibule, which consists of loops between sheets β3,4, β4,5, β5,6 and β7,8, is ideally sized to harbor a biotinylated cofactor and its substrate. Inspired by a visionary publication by Wilson and Whitesides5, eight catalytic transformations have been implemented relying on ArMs derived from the biotin-(strept)avidin technology, which include hydrogenation5,6,7,8, transfer-hydrogenation (of ketones, imines and enones)9, allylic alkylation10, C–H activation11, metathesis12, sulfoxidation13,14, dihydroxylation15 and alcohol oxidation16. Although initial studies have relied on avidin as a host protein5,6, Sav is preferred in most cases because of its straightforward expression in Escherichia coli. To minimize laborious steps, mature Sav is preferred over core Sav, as the latter is expressed as inclusion bodies17,18. The biotin-Sav technology has been proven to be a versatile tool for the implementation of ArMs, offering the possibility to implement abiotic reactions within a biological environment. In this context, olefin metathesis (metathesase) is particularly attractive, as it is a bio-orthogonal reaction19 that has no equivalent in nature. Recently, it was shown that an artificial transfer hydrogenase (ATHase) based on the biotin-Sav technology could be combined with enzymes in cascades by preventing mutual inactivation20. Compared with other abiotic cofactor anchoring strategies3, supramolecular anchoring allows faster screening, as the ArMs are generated quantitatively and thus require no additional reagents, coupling steps or purification.

(Strept)avidin is a highly versatile protein that is often referred to as 'molecular Velcro'. In the past 40 years, it has found numerous applications in a variety of fields including protein purification, immobilization, interaction studies and diagnostic applications21,22,23. The protocol described herein may be applicable to any application of the biotin-Sav technology requiring fine-tuning of Sav properties.

A chemogenetic optimization strategy opens up the possibility to rapidly improve the catalytic performance of ArMs based on the biotin-Sav technology (Fig. 1). Systematic variation of the biotinylated spacer-ligand moiety can be combined with genetic modification of Sav to afford ArMs with improved characteristics (kinetics, stability, turnover number (TON) and selectivity). In view of the requirement of purified Sav samples to ensure activity of the abiotic metal cofactor, optimization efforts thus far have been limited to large-scale expression and purified Sav mutant libraries24,25. Thus far, mostly close-lying amino acid residues have been subjected to saturation mutagenesis. An attractive feature of this strategy is the possibility to exploit the same Sav libraries for a variety of ArMs. To fully capitalize on the potential of ArMs based on the biotin-Sav technology, a streamlined screening protocol relying on CFE or partially purified Sav is highly desirable (Fig. 2). Indeed, thus far, screening has been performed on purified Sav samples, thus seriously limiting the number of mutants accessible. The protocol described herein allows the screening of hundreds of Sav mutants within a reasonable time frame using CFE. With this goal in mind, we selected a 24-well-plate–based screening assay to streamline the entire screening process, thus significantly increasing the throughput.

Figure 1: Model reactions and structures of the ATHase cofactor [Cp*Ir(biot-p-L)Cl] 3 and the metathesase cofactor 6 (for a synthesis overview see Supplementary Figs. 1 and 2).
Figure 1

Upon incorporation of the biotinylated cofactor within Sav, an artificial transfer hydrogenase (see Step 28A) and an artificial metathesase (see Step 28B) are formed. The reaction conditions for the hydrogenation of 1-phenyl-3,4-dihydroisoquinoline and for the ring-closing metathesis reaction of 2-(4-(N,N-diallylsulfamoyl)phenyl)-N,N,N-trimethylethan-1-aminium iodide are displayed. MR buffer, metathesase reaction buffer; cat., catalyst.

Figure 2: Workflow used in this protocol.
Figure 2

Mutagenesis, expression and screening for the identification of genetically engineered ArMs variants based on the biotin-Sav technology. Step numbers are listed according to the PROCEDURE.

Experimental design


Here we detail a protocol for the creation of Sav isoforms, for their functional overexpression, and for the generation of ArMs; we include two complementary screening methods (Fig. 2). The protocol is broken down into the following sections: (i) basic primer design for QuikChange mutagenesis and generation of the Sav library (mutagenesis section); (ii) recombinant protein expression in 24-deep-well–plate format, from the glycerol stock culture to the final CFE containing the overexpressed soluble Sav (library expression section); (iii) the determination of the free biotin-binding sites present in the CFE using biotin-4-fluorescein (B4F; Fig. 2; Step 28A); and (iv) the optimization of the performance of ArMs for an artificial ATHase (Step 28A) and a metathesase (Step 28B) using CFE relying on two partial Sav purification schemes. For the ATHase, a diamide treatment is used to neutralize the deleterious glutathione (GSH)26, and for the metathesase a reversible immobilization on a Sepharose-iminobiotin resin is performed (Fig. 2; Step 28B)12,27.

Library design.

Precise primers are designed for each mutation. In this context, this strategy offers several advantages: (i) the introduced amino acid can be chosen freely (i.e., not dictated by the choice of degenerate primers)28, (ii) the screening effort is reduced as no oversampling is required and (iii) as each well contains a defined Sav mutant structure-activity relationships can be obtained for each experiment. For this purpose, a protocol for using site-directed mutagenesis29 in 96-well format is implemented. On the basis of previous screening efforts with a variety of ArMs, mature Sav (ref. 17) bearing an alanine mutation at position 121 as starting isoform for the library generation is selected. The sequence of pET-24a SavK121A is codon-optimized to reduce the G/C content from 68% (ref. 17) to 52%. This significantly improves the mutagenesis success rate without influencing Sav expression levels. The positions for mutagenesis are selected by identifying residues that lie within a 15 Å radius around the averaged position of the biotinylated metal moiety bound to Sav (Protein Data Bank (PDB) codes: 2QCB, 2WPU, 3PK2, 4GJS and 4GJV)25,27,30,31. This led to the identification of 21 amino acid positions, namely G48, A50, A65, D67, S69, R84, N85, A86, H87, L110, T111, S112, G113, T114, T115, A117, N118, A119, S122, T123 and L124. By examination of the X-ray structures, four further amino acid positions (R53, P64, G68 and E116) lying close to mutated areas, one (G98) within the backbone of loops between sheets β5,6 and β7,8, and two (K144, N150) within the nonresolved, flexible C termini from mature Sav are selected. In total, 28 positions are subjected to mutagenesis (Fig. 3).

Figure 3: Close-up view of the X-ray structure of an artificial transfer hydrogenase based on the biotin-Sav technology.
Figure 3

For clarity, only one cofactor [Cp*Ir(Biot-p-L)Cl] (compound 3) is displayed (as stick), and Sav (mutant S112A, displayed as surface/cartoon representation). The positions selected for mutagenesis are highlighted in orange and residue K121 is highlighted in cyan (PDB code: 3PK2).

Unique primers are designed for each Sav mutant according to the guidelines summarized in Box 1 (ref. 32). The following amino acids are substituted at the above positions in the template Sav K121A sequence (Fig. 4): A, V, L, D, E, Q, K, H, M, Y, S, P (or N if one of the targeted mutation is present at this position). It should be noted that the primer library used for the first round of mutagenesis may be used for the second round. However, this requires that the position to be mutated is distant enough from the mutation position from the previous round (typically six amino acids). If this is not the case, new primers need to be designed for use in the second round of mutagenesis. For the library production and the Sav expression, the E. coli strains TOP10 and BL21 (DE3) are used, respectively. The TOP10 strain is versatile for the site-directed mutagenesis steps, whereas the BL21 (DE3) strain is very efficient for the production of soluble Sav (ref. 17).

Box 1: Design of noncomplementary, site-directed mutagenesis primers • TIMING 1–2 d
Figure 4: Summary of designed mutants.
Figure 4

Matrix of Sav isoforms designed and produced for screening purposes. As a template for mutagenesis, the mutant Sav K121A is used. (a) Amino acid. (b) Position in the Sav template.

Expression of Sav mutants in 24-deep-well plates.

To maximize the amounts of soluble Sav in small-scale expression (i.e., >1 mg Sav, corresponding to 61 nmol of free binding sites (Sav FBS) or 15 nmol of tetrameric Sav in 6 ml of medium), a simple and robust protocol based on the Lac-operon induction system is applied. For this purpose, the ZYP-5052 medium is used. It contains defined amounts of glucose and lactose in the culture medium33. Upon using this medium, the recombinant protein's expression is induced by lactose in the exponential growth phase once glucose is consumed, thus requiring no manual induction by IPTG. The protein production is automatically turned on in each well at nearly the same optical density (OD). Relying on this simple and reproducible procedure, good yields of soluble and functional Sav are obtained within 24 h. Typically, final ODs of 6–7 are reached for each culture. A slight modification of the ZYP-5052 medium is required for a 24-h expression: compared with the original recipe, the amount of carbohydrates (lactose and glucose) and the MgSO4 concentration are doubled. To improve aeration within the wells, a pipette tip (0.1–10 μl) is added. This leads to an increase in OD and Sav yields. When this procedure is followed, Sav yields >1 mg (61 nmol Sav FBS) per well are consistently obtained. This is sufficient to perform one to two ArM screening experiments. Slightly higher Sav yields can be achieved if the expression time is increased to 48 h using the same culture volume. In this case, the original ZYP-5052 medium can be used and no pipette tip is required.

A technical issue concerns the necessary laboratory equipment. Although the protocol described here may be adapted to culture tubes, we strongly favor a 24-well-plate format. For this purpose, we recommend shaking-flask incubators that are equipped with sticky pads. In our experience, the 24-well plates can be conveniently fixed, and they are held securely up to at least a 250-r.p.m. shaking speed. This enables the cultivation of many plates in parallel in a single incubator. To save time, a plate centrifuge is recommended for harvesting and clarifying the supernatants. In our experience, adapting this protocol to culture tubes is inefficient and impractical. Indeed, we routinely observe lower cell densities, leading to higher deviations in catalysis. In addition, the handling of tubes versus 24-well plates is more time consuming.

Gratifyingly, upon cultivating ninety replicates in four 24-well expression plates, an 11% (Step 28A) and a 9% (Step 28B) s.d. in conversion for the ArM-catalyzed reaction is obtained, highlighting the excellent reproducibility of the protein expression protocol using this format.

Determination of Sav FBS in CFE.

Previous experience suggests that the concentration of the biotinylated cofactor (50 μM) should not exceed half of the Sav FBS (100 μM). Although each ArM behaves differently, it has been found that saturating all the biotin-binding sites with a metal cofactor may lead to substantial erosion of an ATHase's performance34. To determine the Sav FBS concentration within the CFE, a modified assay using B4F is implemented35,36. Gruber and colleagues35 and Mascotti and Waner36 demonstrated that B4F can be used for the Sav FBS determination in CFE using either fluorescence (excitation 485 nm; emission 520 nm) or absorbance at 493 nm. We adapted this assay to a 96-well-plate format measuring simultaneously the drop in absorption and in fluorescence of B4F in the presence of the Sav CFE. With this assay, multiple samples can be handled in parallel. The biotinylated fluorophore B4F (40 μM) is added to each Sav-containing well, and the resulting decrease in absorbance and fluorescence is determined. Depending on the expression levels, 30–60 μl of CFE is required to determine the Sav FBS concentration in triplicate. The single point determination yields a 10–20% error margin. However, when evaluating Sav FBS <10 μM, the error increases substantially. Importantly, the pET-24a empty vector CFE background must be subtracted from both the absorbance and fluorescence determinations.

ATHase for the reduction of cyclic imine.

Prochiral cyclic imines are versatile substrates for the production of enantioenriched amines. We have previously shown that ATHases resulting from the incorporation of a biotinylated pianostool complex within Sav are promising hybrid catalysts for this transformation (Fig. 1). The biotinylated three-legged iridium pianostool complex [Cp*Ir(Biot-p-L)Cl] (compound 3, Fig. 1) was synthesized according to a published procedure37 and added to Sav isoforms to afford an ATHase. Unfortunately, the recombinant Sav isoforms overexpressed in E. coli need to be purified by affinity chromatography, as traces of GSH, which are present in cellular debris38, irreversibly poison the precious metal cofactor. To circumvent this severe bottleneck, we recently identified that 1,1-azobis(N,N-dimethylformamide) (diamide) neutralizes the dendrimental effect of GSH on [Cp*Ir(Biot-p-L)Cl] (compound 3) by oxidizing it to GSSG26. This finding allows us to screen samples of Sav contained in E. coli CFEs and cell lysates.

Catalysis with immobilized Sav-mutants using iminobiotin-Sepharose beads.

As a model reaction with Sav using iminobiotin-Sepharose beads as a rapid purification tool27, the ring-closing metathesis of olefins was selected (Fig. 1). An artificial metathesase based on the biotin-Sav technology was previously reported by us12. As the ring-closing reaction using diallylsulfonamides works best at pH 4, an acetate buffer is selected as reaction medium. At this low pH, iminobiotin is protonated and the immobilized Sav is released from the iminobiotin-Sepharose beads. As the metathesis cofactor 6 (ref. 39) bears a biotin anchor, it binds tightly to Sav at this pH and catalyzes the ring-closing metathesis of substrate 4 to form product 5 (Fig. 1).


The main limitation of this protocol is the marked decrease in activity and reproducibility of ArMS when the Sav FBS concentration is low. For both reactions presented herein, cell cultures with Sav FBS >100 μM should be targeted. Indeed, past experience with ATHase of cyclic imines reveals that highly reproducible results require 3 M sodium formate in 0.6 M MOPS at pH 6 and >50 μM Sav FBS. The limited solubility of both formate and MOPS does not allow preparation of more than 2× concentrated stock buffer solutions. In contrast, the metathesase performs well down to 10 μM FBS Sav on purified Sav samples. However, the immobilization protocol requires >100 μM Sav FBS. Indeed, Sav FBS in the CFE must be present in excess compared with the loading capacity of the iminobiotin-Sepharose beads.

As illustrated by the ATHase and metathesase, each ArM behaves differently, and exploratory screens should be performed to identify the most suitable screening strategy. For example, diamide treatment is not suitable for the metathesase, as the ruthenium cofactor is inactivated by this reagent. Instead, iminobiotin-Sepharose beads are highly versatile for the metathesase, as the Sav is released from the beads at pH 4, which allows catalysis in solution to be performed. In our experience, the iminobiotin beads immobilization (Step 28B) is more versatile than diamide treatment (Step 28A). However, for reactions requiring neutral or basic conditions, an additional buffer-exchange step is necessary. For this purpose, the eluted fraction containing Sav at pH 4 can be back-titrated to the desired pH by the addition of concentrated base or a high-molarity reaction buffer.

Concerning evolution strategies, a combination of synergistic mutations may be envisaged28. To be able to reuse the initial design primer library, it is important to ensure that the fixed mutation (first mutation) does not overlap with the primer pair used for the mutation to be introduced (second mutation). For this purpose, the second mutation should be more than six amino acids away from the first mutation. If this is not the case, a new set of primers for the second mutation needs to be designed and used.



  • pET-24a empty vector (Novagen, cat. no. 69749, antibiotic: kanamycin resistance)

  • pUC-19 DNA (Clontech, cat. no. 3219, antibiotic: ampicillin resistance)

  • Sodium chloride, 99.8% (7647-14-5; Sigma-Aldrich, cat. no. 31434)

  • Yeast extract (8013-01-02; Merck, cat. no. 1.11926.1000)

  • Tryptone (91079-40-2; BD, cat. no. 211705)

  • Sodium hydroxide, 97% (1310-73-2; VWR, cat. no. 28240.292)

  • α-Lactose monohydrate, 99% (5989-81-1; Sigma-Aldrich, cat. no. L3625)

  • D-(+)-Glucose monohydrate, 99% (14431-43-7; Fluka, cat. no. 49159)

  • MgCl2, 99% (7791-18-6; Sigma-Aldrich, cat. no. 31413)

  • MgSO4, 99–101% (22189-08-8; Sigma-Aldrich, cat. no. 13143)

  • KH2PO4, 99% (7778-77-0; Acros Organics, cat. no. 447670010)

  • Na2HPO4, 98%, anhydrous (7558-79-4; Acros Organics, cat. no. 448140010)

  • Kanamycin (25389-94-0; AppliChem, cat. no. A1493)

  • Agarose (molecular biology grade, 9012-36-6; AppliChem, cat. no. A8963)

  • Agarose (bacteriology grade, 9002-18-0, AppliChem, cat. no. A0949)

  • DMSO, 99.9% (molecular biology grade, 67-68-5; Sigma-Aldrich, cat. no. 41640)

  • Glycerol, anhydrous p.A. (56-81-5; AppliChem, cat. no. A3552)

  • CaCl2 2H2O, 99.5% (10035-04-8; AppliChem, cat. no. A387)

  • RbCl, p.A. (7791-11-9; AppliChem, cat. no. A4240)

  • MnCl2·4H2O, 99.9% (13446-34-9; AppliChem, cat. no. A2087)

  • Potassium acetate, 99% (127-08-2; AppliChem, cat. no. A4279)

  • Lysozyme (9001-63-2; AppliChem, cat. no. A3311)

  • DNase I (Roche, cat. no. 10104159001, from bovine pancreas, grade II)

  • DpnI (NEB, cat. no. R0176)

  • Full-length Sav, 95% (Abcam, cat. no. ab78833)

  • Biotin-4-fluorescein (AnaSpec, cat. no. AS-60654)

  • DNA ladder, 1 kb (NEB, cat. no. N3232)

  • Q5 polymerase (NEB, cat. no. M0493)

  • Q5 polymerase reaction buffer, 5× (NEB, cat. no. B9027)

  • dNTP mix, 10 mM each (5 Prime, cat. no. 2900389, 10 mM of each dNTP)

  • Ethidium bromide, 10 mg ml−1 in H2O (1239-45-8; Sigma-Aldrich, cat. no. E1510)

  • Milli-Q water, 18.2 MΩ cm (hereafter MQ)

  • DMSO, ≥99.9% (67-68-5; Sigma-Aldrich, cat. no. 276855)

  • Dichloromethane (75-09-2; Baker, cat. no. 9410, HPLC quality)

  • n-Hexane (110-54-3; Baker, cat. no. 9304, HPLC quality)

  • 1,1-Azobis(N,N-dimethylformamide), 95% (10465-78-8; Sigma-Aldrich, cat. no. D3648-16)

  • Isopropanol (67-63-0; Biosolve, cat. no. 16260602, HPLC quality)

  • Diethylamine, 99.5% (109-89-7; Acros Organics, cat. no. 378370010)

  • 3-(N-Morpholino)propanesulfonic acid (MOPS), 99% (1132-61-2; Alfa Aesar, cat. no. 10171628)

  • Sodium formate, 99% (141-53-7; Alfa Aesar, cat. no. B05U045)

  • 1-Phenyl-3,4-dihydroisoquinoline prepared according to Lantos et al.40

  • Methanol (67-56-1; HPLC gradient grade from J.T.Baker, cat. no. 9093-02)

  • Acetonitrile for UPLC (75-05-8, ultra gradient; Romil, code H050)

  • Trifluoroacetic acid, >99.9% (76-05-1; Romil, code H853)

  • Sodium acetate, 98%, anhydrous (127-09-3; Fisher Scientific, cat. no. BP333-500)

  • Acetic acid, 100% (64-19-7; VWR, cat. no. 97064-482)

  • MgCl2·6 H2O, 99% (7791-18-6; Riedel-de Haen)

  • NaCl, 98% (7647-14-5; VWR)

  • Iminobiotin-Sepharose beads, 10 mg avidin per ml binding capacity (Affiland, product code IMI-H-4FF)

  • ATHase catalyst 3 (see the Supplementary Methods for the detailed synthesis)25,37

  • Metathesis catalyst 6 (see the Supplementary Methods for the detailed synthesis)39

  • 4-(2-Aminoethyl)benzenesulfonamide, 98% (35303-76-5; TCI, cat. no. A1363)

  • Boc anhydride (24424-99-5; Iris Biotech, cat. no. RL-1007)

  • Tetrahydrofuran (THF), 99.9% (109-99-9; Sigma-Aldrich, cat. no. 401757)

  • Dichloromethane (DCM, 75-09-2; Baker, cat. no. 9410, HPLC quality)

  • Ethylacetate (EtOAc), 99.7% (114-78-6; Sigma-Aldrich, cat. no. 34858)

  • MgSO4, 99–101% (22189-08-8; Sigma-Aldrich, cat. no. 13143)

  • Hexane, 95% (110-54-3; Sigma-Aldrich, cat. no. 650552)

  • Acetonitrile (ACN, 75-05-8, ultra gradient, Romil, code H050)

  • Allyl bromide (106-95-6; ABCR, cat. no. AB113353)

  • K2CO3, 99% (584-08-7; Sigma-Aldrich, cat. no. P5833)

  • Cyclohexane, 99% (110-82-7; Sigma-Aldrich, cat. no. 179191)

  • HCl, 4 M in dioxane (7647-01-0; Fluka, cat. no. 40427)

  • Diethylether (DEE), 99% (60-29-7; J.T.Baker, cat. no. 9259-02)

  • 1,2-Dimethoxyethane (DME), 99% (110-71-4; Sigma-Aldrich, cat. no. E27408)

  • Iodomethane, 99% (74-88-4; Sigma-Aldrich, cat. no. 67692)

  • [Cp*IrCl2]2 (99.95%, 12354-84-6; Umicore Precious Metals Chemistry, cat. no. 3000034538)

  • DMSO-d6, 99.8% (2206-27-1; Cambridge Isotope Laboratories, cat. no. DLM-7-100S)

  • CD3OD, 99.8% (811-98-3; Cambridge Isotope Laboratories, cat. no. DLM-24-10)

  • CDCl3, 99.8% (865-49-6; Cambridge Isotope Laboratories, cat. no. DLM-7-100)

  • Celite (68855-54-9; Sigma-Aldrich, cat. no. 419931)

  • 2-Chloro-4,6-dimethoxy-1,3,5-triazine, 98% (3140-73-6; Acros, cat. no. 26300)

  • N-Methylmorpholine, 99.5% (109-02-4; Acros, cat. no. 127151000)

  • Pd/C, 10 wt% loading (Sigma-Aldrich, cat. no. 205699)

  • Triethylamine (NEt3), 99% (121-44-8; Sigma-Aldrich, cat. no. T0886)

  • Hydrogen gas, ≥99.99% (1333-74-0; Sigma-Aldrich, cat. no. 295396)

  • Silicagel (112926-00-8; Sigma-Aldrich, cat. no. 419931; grade for column chromatography, particle size ≤0.063 mm)

  • N-Boc-ethylenediamine, 98% (57260-73-8; Acros, cat. no. 38990)

  • 4-Nitrophenylsulfonyl chloride, 97% (98-74-8; Sigma-Aldrich, cat. no. 272248)

  • Copper cyanide, 99% (544-92-3; Sigma-Aldrich, cat. no. 216305)

  • 1-Methyl-2-pyrrolidinone (NMP), 99.5% (872-50-4; Sigma-Aldrich, cat. no. 328634)

  • Lithium aluminum hydride (LiAlH4), 95% (16853-85-3; Sigma-Aldrich, cat. no. 323403)

  • 4-(Dimethylamino)pyridine (DMAP), 99% (1122-58-3; Sigma-Aldrich, cat. no. 107700)

  • Triethylorthoformate (HC(OEt)3), 98% (122-51-0; Sigma-Aldrich, cat. no. 304050)

  • Ammonium chloride (NH4Cl), 99.5% (12125-02-9; Sigma-Aldrich, cat. no. 213330)

  • Potassium hydroxide (KOH), 90% (1310-58-3; Sigma-Aldrich, cat. no. 484016)

  • Chloroform (CH3Cl), 99% (67-66-3; Sigma-Aldrich, cat. no. 372978)

  • Toluene, 99.8% (108-88-3; Sigma-Aldrich, cat. no. 244511)

  • Sulfuric acid (H2SO4), 95–98% (7664-93-9; Sigma-Aldrich, cat. no. 320501)

  • Dimethylformamide (DMF), 99.8% (68-12-2; Sigma-Aldrich, cat. no. 227056)

  • Biotin pentafluorophenol ester, 99% (120550-35-8; ABCR, cat. no. AB259902)

E. coli strains

  • All the strains used require a biosafety level I authorization. TOP10 (Invitrogen, cat. no. C4040): F mcrA Δ(mrr-hsdRMS-mcrBC) Φ80lacZΔM15 Δlac×74 recA1 araD139 Δ(araleu)7697 galU galK rpsL (StrR) endA1 nupG

  • BL21 (DE3; NEB, cat. no. C2527): E. coli B Fdcm ompT hsdS(rBmB) gal λ(DE3)


  • Tips, 0.1–10 μl × 34 mm (Eppendorf, cat. no. 0030 000.811)

  • 24-well plates, 10-ml round-bottom well (Qiagen, cat. no. 19583)

  • 96-well plates, 2-ml round-bottom well (GE Healthcare, cat. no. 7701-5200)

  • 96-well plates, 350 μl transparent, flat bottom (Brand, cat. no. 781602)

  • 96-well plate PCR applications

  • Permeable sealing membrane for multiwell plates (Sigma-Aldrich, cat. no. Z763624)

  • Millipore water purification system (Milli-Q (MQ) water)

  • Precision weighing balance

  • Weighing paper

  • Pasteur pipettes

  • Falcon conical tubes

  • HPLC vials

  • Teflon-coated magnetic stir bar (2 × 5 mm)

  • Plate with a magnetic stirrer

  • Volumetric flasks (1-ml volume)

  • Eppendorf tubes (1.5- and 2-ml volume)

  • Microplate reader (Tecan, Infinite M200)

  • Normal phase HPLC (Agilent 1100 with diode array detector and autosampler)

  • Illuminator at 302 nm

  • pH meter

  • Autoclave

  • Autoclaveable bags

  • VWR MiniSTar centrifuge

  • Well-plate centrifuge (Thermo Heraeus Multifuge 4KR, maximum 5,300g; if this speed cannot be reached, a lower one can be selected. In this case, the centrifugation time should be increased to ensure clarification of the supernatant)

  • Shaking incubator (shaking amplitude 2.5 cm, adapted with sticky pads)

  • Refrigerated table-bench centrifuge for 1.5 or 2 ml tubes (maximum 21,000g; if this speed cannot be reached a lower one can be selected. In this case, the centrifugation time should be increased to ensure clarification of the supernatant)

  • Laminar flow hood

  • Cryotubes (glycerol stocks)

  • Standard plastic box for cryotubes (133 mm × 133 mm × 45 mm)

  • Eppendorf pipetting tips

  • Pipettor (1 ml, 100 μl and 10 μl)

  • Pipettor for organic solvents or syringe

  • Acquity UPLC H-Class Bio (Waters)

  • BEH C18 column, 1.7 μm, 2.1 × 50 mm for UPLC

  • Chiral stationary phase HPLC Agilent (Chiralpak IC, 250 × 4.6 mm, 5 μm)

  • Chemical glass equipment: round-bottom flasks of different sizes, chromatography columns and test tubes

  • Magnetic stirring plates with a heating bath

  • Magnetic stirrers

  • Rotary evaporator

  • High-vacuum pump

  • Schlenk line


ZYP salts, 20×

  • Dissolve 1 M KH2PO4 (54.4 g), 1 M Na2HPO4 (56.8 g) and 0.5 M (NH4)2SO4 (26.4 g) in MQ water (final volume of 400 ml)33. Autoclave the solution (121 °C for 20 min). The sterile solution can be stored at room temperature (RT, 25 °C) for years.

ZYP sugar, 20×

  • Dissolve 10% (vol/vol) anhydrous glycerol (40 ml), 55.5 mM glucose monohydrate (4.4 g) and 4% (wt/vol) lactose XM (16 g) in MQ water (final volume of 400 ml)33. Filter-sterilize the solution using a 0.22-μm filter. The sterile solution can be stored at 4 °C for up to 1 year.

MgSO4, 100×

  • Dissolve 200 mM MgSO4 (2.46 g) in MQ water (50 ml). Autoclave the solution (121 °C for 20 min). The sterile solution can be stored at 4 °C for years.

Normal ZYP-5052 medium

  • Dissolve tryptone (8 g) and yeast extract (4 g) in 712 ml of MQ water, and then autoclave the solution (121 °C for 20 min)33. Add 20× ZYP salts (40 ml), 20× ZYP sugar (40 ml) and MgSO4 (8 ml) to the medium. The sterile solution (without antibiotic) can be stored at 4 °C for up to 1 year.

Modified ZYP-5052 medium

  • Dissolve tryptone (8 g) and yeast extract (4 g) in 664 ml of MQ water and autoclave the solution (121 °C for 20 min). Add 20× ZYP salts (40 ml), 20× ZYP sugar (80 ml) and MgSO4 (16 ml) to the medium. The sterile solution (without antibiotic) can be stored at 4 °C for up to 1 year.

Lysogeny Broth (LB)

  • Dissolve 1% (wt/vol) tryptone (10 g), 0.5% (wt/vol) yeast extract and 1% (wt/vol) NaCl in MQ water (final volume of 1 l). Adjust the pH to 7 using NaOH (ref. 41). For LB agar plates, use the same recipe but add 1.5% (wt/vol) agar (15 g, bacteriology grade). Autoclave the solution (121 °C for 20 min). To prepare agar plates, cool the plates to 55 °C, add the antibiotic, and then pour the medium into the plates. The sterile solution (without antibiotic) can be stored at 4 °C for up to 1 year. The plates can be stored at 4 °C for up to 3 months (for kanamycin).

Super-optimal lysogeny broth with catabolite repression LBSoc

  • Dissolve 2% (wt/vol) tryptone (2 g), 0.5% (wt/vol) yeast extract (0.5 g), 10 mM NaCl (0.05 g) and 2.5 mM KCl (0.019 g) in MQ water (final volume of 100 ml). Adjust the pH to 7 using NaOH (ref. 42). Autoclave the solution (121 °C for 20 min) and add filter-sterilized solutions (0.22-μm cutoff) of 1 M MgCl2 (1 ml), 1 M MgSO4 (1 ml) and 1 M glucose (2 ml). The sterile solution (without antibiotic) can be stored at 4 °C for up to 1 year.

Super-optimal lysogeny broth

  • LBSOB has the same composition as LBSoc but without glucose42. The sterile solution (without antibiotic) can be stored at 4 °C for up to 1 year.

RF1 buffer

  • Weigh RbCl (100 mM, 2.42 g), MnCl2·4 H2O (50 mM, 1.98 g), KOAc (30 mM, 0.59 g) and CaCl2·2 H2O (10 mM, 0.29 g), and fill up with 15% (vol/vol) glycerol (200-ml final volume). Set the pH to 5.8 using 0.2 M CH3COOH, and filter-sterilize the solution43. The solution can be prepared in advance and stored at 2–8 °C for up to 6 months.

RF2 buffer

  • Weigh RbCl (10 mM), CaCl2·2 H2O (75 mM) and MOPS (10 mM) and fill it up with 15% (vol/vol) glycerol in MQ water (50-ml final volume). Set the pH to 7 using NaOH, and filter-sterilize the solution43. The solution can be prepared in advance and stored at 2–8 °C for up to 6 months.

Reaction buffer for ATHase

  • Dissolve MOPS (12.556 g) and sodium formate (20.403 g) in MQ water (final volume 50 ml) to obtain final concentrations of 1.2 M MOPS and 6 M formate. Adjust the pH to 6 using NaOH. This solution can be prepared in advance and stored for up to 6 months at 2–8 °C.

ATHase substrate stock solution

  • Dissolve 1-phenyl-3,4-dihydroisoquinoline (abbreviated as compound 1 in Fig. 1; 165.8 mg, final concentration 400 mM) in DMSO (2 ml). The solution can be prepared in advance and stored for up to 6 months at 2–8 °C.

ATHase stock solution of the iridium catalyst

  • Dissolve [Cp*Ir(biot-p-L)Cl]37 (abbreviated as compound 3 in Fig. 1; 1.6 mg, final concentration of 2 mM) in DMSO (1 ml). This solution can be stored at 4 °C for 30 d.

Diamide stock solution

  • Dissolve 1,1-Azobis(N,N-dimethylformamide) (72.5 mg, final concentration 200 mM) in DMSO (2 ml).


    • This solution needs to be freshly prepared before each catalysis campaign.

Lysis buffer for ATHase or metathesis

  • Dissolve MOPS (ATHase, 522.9 mg, final concentration 50 mM) in MQ water (50 ml) or use PBS (metathesase, 12 mM phosphate, pH 7.5). Adjust the pH to 7.4 with NaOH. Dissolve lysozyme (20 mg) and DNase I (20 μg) in this buffer (20-ml final volume).


    • This solution needs to be freshly prepared before each catalysis campaign.

FBS assay buffer

  • Dissolve MOPS (ATHase, 2,092 mg, final concentration 100 mM) in MQ water (100 ml). Adjust the pH to 7.5 with NaOH. The solution can be stored at RT for up to 1 year.

Sav-stock solution for FBS determination

  • Prepare a pure 1 mM Sav FBS stock solution (use the molecular weight of the monomer 16,368 Da) in MQ water in a final volume of 1 ml. Dilute this in AB to a final concentration of 100 μM Sav FBS. The solution can be stored at −20 °C for 3 months if multiple freeze-thaw cycles are avoided.

B4F solution for FBS determination

  • First, prepare a biotin-4-fluorescein (B4F) stock solution (10 mM) in DMSO. Then, dilute this stock solution to 0.4 mM using FBS assay buffer. The solutions can be stored at −20 °C for 1 year if multiple freeze-thaw cycles are avoided.


    • As B4F is light-sensitive, protect the solutions with aluminum foil.

Iminobiotin binding buffer

  • Adjust the pH of a NaHCO3 buffer solution (50 mM) containing NaCl (0.5 M) to pH 9.8 by dropwise addition of NaOH (5 M stock solution). The buffer can be stored at RT for up to 1 year, but the pH should be controlled before use.

Metathesase reaction buffer

  • For this buffer, add MgCl2.6H2O (101.5 g) to an acetic acid stock solution (847 ml of a 100 mM solution) and a sodium acetate stock solution (153 ml of a 100 mM solution). Adjust the pH to 4 by dropwise addition of NaOH (5 M stock solution). The buffer can be stored at RT for up to 1 year, but the pH should be controlled before use.

tert-Butyl (4-sulfamoylphenethyl)carbamate (abbreviated as compound 8)

  • (Abbreviated as compound 8 in Fig. 5.) To a flask containing 4-(2-aminoethyl)benzenesulfonamide (abbreviated as compound 7 in Fig. 5; 5.01 g, 24.5 mmol, 1.00 equivalents (eq.)), add DCM (20 ml) and THF (40 ml). Next, add Boc anhydride (5.61 g, 25.7 mmol, 1.05 eq.) to the mixture and stir for 2 h. Dilute the mixture with 200 ml of EtOAc and wash it with 1 M aqueous HCl (2 × 100 ml) and brine (100 ml). Dry the organic phase over MgSO4 and remove the solvent under vacuum. Wash the white solid with hexane twice and dry it under high vacuum to obtain the desired product 8 as a white solid (6.20 g, 86%). MP: 181.5–182.5 °C. 1H NMR (400 MHz, DMSO) δ = 7.74 (d, 3JHH = 8.3 Hz, 2H), 7.38 (d, 3JHH = 8.3 Hz, 2H), 7.29 (s, 2H), 6.93 (s, 1H), 3.20–3.11 (m, 2H), 2.81–2.73 (m, 2H), 1.38 (s, 9H) (Supplementary Fig. 3). Use the compound immediately for the next synthesis step. No storage experience is available. The analytical data are identical to those shown in the literature44.

Figure 5: Overview of the synthesis of the diolefinic substrate 4.
Figure 5

tert-Butyl (4-(N,N-diallylsulfamoyl)phenethyl)carbamate (abbreviated as compound 9)

  • (Abbreviated as compound 9 in Fig. 5.) To a solution of tert-butyl (4-sulfamoylphenethyl)carbamate (8; 6.19 g, 20.6 mmol, 1.00 eq.) in ACN (100 ml), add allyl bromide (6.29 g, 4.53 ml, 51.5 mmol, 2.50 eq.) and K2CO3 (7.12 g, 51.5 mmol, 2.50 eq.). Stir the mixture at 85 °C overnight, and allow it to cool to RT. Next, filter and remove the solvent from the filtrate under reduced pressure. Purify the residue by flash column chromatography (silica gel, cyclohexane/EtOAc 2:1) to obtain 9 as a pale yellow oil (6.70 g, 86%). 1H NMR (400 MHz, CDCl3) δ = 7.75 (d, 3JHH = 8.4 Hz, 2H), 7.33 (d, 3JHH = 8.3 Hz, 2H), 5.67–5.56 (m, 2H), 5.19–5.10 (m, 4H), 4.62 (bs, 1H), 3.81 (d, 3JHH = 6.3 Hz, 4H), 3.44–3.35 (m, 2H), 2.88 (t, 3JHH = 7.0 Hz, 2H), 1.43 (s, 9H) (Supplementary Fig. 4). 13C NMR (100 MHz, DMSO) δ = 155.8, 144.3, 138.5, 132.6, 129.5, 127.4, 119.1, 79.5, 49.4, 41.3, 36.1, 28.4 (Supplementary Fig. 5). HRMS [ESI(+)TOF]: calculated for C19H28N2NaO4S [M]+ 403.1662; found 403.1665. Use the compound immediately for the next synthesis step. We have no storage experience with this compound.

Substrate for metathesase (abbreviated as compound 4)

  • (Abbreviated as compound 9 in Fig. 1.) Dissolve tert-butyl (4-(N,N-diallylsulfamoyl)phenethyl)carbamate (9; 3.91 g, 10.3 mmol, 1.00 eq.) in dioxane (15 ml) and add a solution of HCl in dioxane (15 ml, 4 M solution) and stir it at RT for 4 h until thin-layer chromatography (TLC) indicates complete conversion. Precipitate the hydrochloride salt by pouring the mixture into cold DEE (300 ml), collect it by filtration and dry it under high vacuum to obtain a colorless solid (2.91 g, 89%). Next, re-dissolve it in DCM (150 ml), add saturated aqueous NaHCO3 (200 ml) and vigorously stir it for 2 h. Collect the organic phase, extract the aqueous layer with DCM and dry the combined organic phases over MgSO4. Remove the solvent under vacuum to yield the amine as a pale-yellow oil. Take up a fraction of the amine (1.00 g, 3.16 mmol, 1.00 eq.) in DME (20 ml). Add iodomethane (4.49 g, 1.97 ml, 31.6 mmol, 10.0 eq.) and K2CO3 (2.62 g, 19.0 mmol, 6.00 eq.), and stir the mixture for 3 h at RT. Add DCM (80 ml) to re-dissolve the precipitated ammonium salt, and remove the carbonate by filtration. Concentrate the filtrate under vacuum, decant it with DEE and dry it under high vacuum to obtain the pure substrate (4) as a colorless solid (1.13 g, 79%). Use the standard Schlenk technique to prepare the catalyst45. Degas solvents by flushing these with a nitrogen stream for 30 min. Use oven-dried glassware (110 °C, overnight). Remove traces of air using three vacuum refills followed by nitrogen refills. Keep a small nitrogen overpressure throughout the entire reaction to prevent air from contaminating the flask. If you stir a reaction under reflux, place the tap connected to the Schlenk line at the top of the condenser, to prevent excessive evaporation. Use a rubber septum to add liquid via a syringe. Apply a gentle nitrogen overpressure before removing the septum to add a solid. When performing the ligand exchange or the biotinylation, remove the solvent using the Schlenk line equipped with a secondary cooling trap (not on a rotary evaporator). The catalyst should be stable in its solid form. However, it should always be kept in a freezer at −20 °C and under argon. Prepared aliquots of the catalyst should be used within 3 months. MP: 184–186 °C. 1H NMR (400 MHz, DMSO) δ = 7.84 (d, 3JHH = 8.4 Hz, 2H), 7.58 (d, 3JHH = 8.4 Hz, 2H), 5.60 (ddt, 3JHH = 16.3 Hz, 3JHH = 10.1 Hz, 3JHH = 6.2 Hz, 2H), 5.16 (d, 3JHH = 17.1 Hz, 2H), 5.14 (d, 3JHH = 11.0 Hz, 2H), 3.75 (d, 3JHH = 6.2 Hz, 4H), 3.65–3.55 (m, 2H), 3.22–3.12 (m, 11H) (Supplementary Fig. 6). 13C NMR (100 MHz, DMSO) δ = 141.7, 138.3, 132.8, 130.0, 127.3, 118.8, 65.0, 52.4, 49.3, 28.2 (Supplementary Fig. 7). HRMS [ESI(+)TOF]: calculated for C17H27N2O2S [M]+ 323.1788; found 323.1791.

Substrate stock solution

  • Dissolve the metathesase substrate 4 (90.1 mg) in the metathesase reaction buffer in a volumetric flask (10 ml). Metathesis catalyst (abbreviated as compound 6 in Fig. 1) aliquots are prepared as follows: charge a flask with 6 (4 mg) and add dichloromethane (1 ml). Portion this stock solution into aliquots (50 μl) inside small flasks, dispense into five test tubes and seal these with a rubber septum. Remove the solvent from the aliquots by applying a gentle vacuum through a needle. Once evaporated, flush it with nitrogen to obtain aliquots of 6 (0.2 mg). Aliquots should be stored at −20 °C and be used within 3 months.

Iminobiotin beads

  • The beads are sold as a suspension in 20% (vol/vol) ethanol. Adjust the total volume with 20% ethanol to twice the volume of the completely settled beads. Upon homogenization by thorough shaking, the binding capacity is 5 mg Sav per ml of suspension. The beads are stored at 4 °C, and their Sav binding capacity is maintained for several years (>5).

Internal standard solution

  • Dissolve benzyltriethylammonium bromide (272.2 mg) in MQ water in a volumetric flask (100 ml) and fill it up to the mark. The solution can be stored at 4 °C for several months.


Site-directed mutagenesis in 96-well format

Timing: 9–10 d

  1. Design oligos for site-directed mutagenesis (see Box 1 for more information). Spin the 96-well plate (4,000g for 5 min at RT) and either proceed immediately or store it at −20 °C. For a list of oligos used in our analysis, see Supplementary Table 1.

    Pause point

    • Primer solution can be stored at −20 °C indefinitely.

  2. Produce the mutagenesis template plasmid by cloning the SavK121A sequence into pET-24a (Supplementary Data). We recommend synthesizing the gene sequence commercially via a gene supplier where it can be precloned in the desired expression vector. In parallel, also prepare the pET-24a SavWT plasmid and the pET-24a empty vector control. For the Sav sequence, see Supplementary Data.

  3. Transform each plasmid into E. coli BL21(DE3). To do this, add 1−3 μl of plasmid (100 ng) to 50 μl BL21(DE3) on ice, and incubate the mixture for 30 min.

  4. Heat-shock the cells in a water bath (42 °C, 30 s), and then transfer the tubes back to ice for 2 min. Add LBSoc (250 μl) and incubate the tubes with a shaker (300 r.p.m. for 1 h at 37 °C).

  5. Mix the suspension by pipetting it and down. Plate out the suspension (250 μl) on LB agar plates containing the resistance marker kanamycin (50 μg ml−1). Incubate the plates overnight at 37 °C.


  6. Inoculate LB medium (5 ml) containing kanamycin (50 μg ml−1) with a fresh single colony bearing the template plasmid and incubate it overnight (37 °C at 200 r.p.m.).

  7. Prepare a glycerol stock for each by mixing 0.7 ml of overnight culture with sterile glycerol (0.3 ml) to obtain an end concentration of 30% (vol/vol) glycerol and incubate it on ice (30 min). Freeze it in liquid nitrogen and store it at −80 °C. Next, use a commercial MiniPrep plasmid isolation kit (e.g., MN NucleoBond) and follow the instructions of the supplier for high-copy plasmids. Use the remaining overnight culture for isolation. For the elution step, use sterilized MQ water (50 μl) to elute the template plasmid. For the DNA quantification, analyze the above solution (50 μl) with a NanoDrop ND-1000. The sample is pure if the absorbance ratio at 260 nm/280 nm is 1.8. Heat the plasmid solution (70 °C, 10 min) to deactivate any DNase present.

    Pause point

    • Plasmid solutions can be stored indefinitely at −20 °C.

  8. Prepare the QuikChange mutagenesis PCR master mix using the pET-24a SavK121A plasmid. The following is described for one 96-well plate. For QuikChange mutagenesis, a methylated circular DNA template (plasmid) is amplified using mutagenic primers (forward and reverse) and a high-fidelity polymerase. After PCR, a DpnI digestion is performed, which cleaves only the methylated template plasmid DNA. As the newly synthesized DNA is non-methylated, it is not digested by DpnI. This eliminates all of the background template DNA. No ligation of the resulting nicks in the newly synthesized DNA occurs during the PCR. A transformation into an efficient cloning strain, such as E. coli TOP10, is thus necessary to repair the nicks present in the newly synthesized DNA.

    CompoundVolume (μl)Final in 2,500 μl
    MQ water1,687.5 
    Q5 polymerase 5× buffer500
    Template plasmid pET-24a SavK121A (25 ng μl−1)500.5 ng μl−1
    dNTP mix 10 mM each50200 μM each
    Q5 polymerase (2 U μl−1)12.525 U
    Final volume for 100 reactions2,500 
    1. It is possible to apply an alternative high-fidelity, thermostable DNA polymerase. Mix the solution gently and spin down. Add polymerase slowly to avoid air bubbles.
  9. With a multichannel pipette, divide the mutagenesis PCR master mix (24 μl) into each well of a 96-well plate. To avoid unnecessary loss of the master mix, use PCR tubes (0.2 ml) as reservoirs. Pipette gently to avoid air bubbles.

  10. For each mutant, add the primer mix solution (1 μl of the 10 μM solution from Step 1) to the master mix solution (24 μl). Be sure to contact the meniscus with the tip and do not mix afterward. Close the plate tightly with an appropriate PCR lid.

  11. Place the plate into a 96-well-plate–compatible thermocycler. Program the following temperatures, cycles and durations:

    Cycle numberDenatureAnnealExtend
    195 °C, 2 min  
    2–1895 °C, 15 s60 °C, 15 s72 °C, 5 min (40 s kb−1)
    19  72 °C, 10 min
    204–8 °C  

    Critical step

    • Steps 8–11 should be performed without any interruption.

  12. After PCR, add DpnI (0.3 μl, 20 U μl−1) with a multichannel pipette to each well, seal the plate with a breathable membrane and incubate it for 2 h at 37 °C. Deactivate DpnI by heating (10 min at 70 °C). This step digests the methylated template DNA.

    Pause point

    • Digested PCR samples can be stored at −20 °C indefinitely if multiple freeze/thaw cycles are avoided.

  13. Prepare an agarose gel to check for the presence of the mutated PCR product. To do this, weigh agarose (1.5 g, molecular biology grade) in 0.5× TBE buffer (150 ml, 1% (wt/vol) final concentration). Heat it in a microwave until agarose dissolves. Add ethidium bromide (10 μl), mix it gently and pour it into the chamber. Add combs for 100 samples into the warm solution. Wait until the gel is solid (30 min).

  14. Place the gel into the running chamber containing 0.5× TBE and remove the combs. Use DpnI-digested PCR product (5 μl) and mix it with 6× loading buffer (1 μl). This mixture (5 μl) is then loaded into each lane. Use a 1-kb DNA ladder (3 μl) as a standard. Run the gel at 100 V/400 mA for 40 min. Analyze the gel under an illuminator at 302 nm and search for an amplified band at 5.7 kb. Figure 6 displays a typical gel resulting from one 96-well PCR mutagenesis plate with a strong amplified band of the vector in each lane. A band at the size of the plasmid is a good hint that the PCR was successful.


  15. Perform a transformation of each successfully mutagenized plasmid into competent E. coli TOP10 cells. Transfer the digested PCR product (3–5 μl) gently (do not pipette up and down) using a multichannel pipette into a 96-well plate containing chemically competent E. coli TOP10 cells (50 μl per well). Incubate the plate on ice for 30 min. For the preparation of chemically competent E. coli cells in 96-well plates, see Box 2. Use a positive control (e.g., pUC19 vector). Depending on the E. coli strain used (TOP10 typically yields more c.f.u. than BL21 (DE3)), one can typically expect 30–800 c.f.u.

    Critical step

    • Competent cells must be stored on ice the whole time until heat shock. Amplified plasmids are nicked: transformation into an efficient cloning strain is thus necessary.

  16. Heat-shock and plate the cells on 96 separate LB plates containing kanamycin (50 μg ml−1), as described in Steps 4–6. Incubate the plates overnight (37 °C at 200 r.p.m.).

  17. Pick one to three colonies per mutant and grow them overnight (37 °C at 200 r.p.m.) in 96-well plates containing LB with kanamycin (50 μg ml−1). Follow the manufacturer's protocol from the 96-well plasmid isolation kit for the preparation of overnight cultures and for the plasmid isolation procedure. Elute twice with the elution buffer (2 × 30 μl). Alternatively, in a 24-deep-well plate, LB can be used (6 ml per well) for overnight cultures to increase the final DNA content. Typical isolated plasmid yields are 50–100 ng μl−1.


  18. Sequence plasmids with the standard T7 promoter primer. Forward sequencing is sufficient to cover the whole Sav-encoding gene (<500 bp). To minimize costs, it is recommend to use the 96-well plate format sequencing to confirm the introduction of the desired mutations.


  19. Perform transformation into competent E. coli BL21(DE3) cells. Add the mutated Sav plasmids (1–3 μl, 100 ng) to chemically competent E. coli BL21 (DE3; 50 μl) in 24-deep-well plates on ice. Incubate the plates for 30 min. Repeat Steps 4 and 5.

    Pause point

    • The remaining plasmid solution can be kept at −20 °C for several years.

  20. Prepare overnight cultures in LB medium (5 ml) containing kanamycin (50 μg ml−1) from a single colony of each E. coli BL21 (DE3) Sav mutant (same as Step 6).

  21. Prepare glycerol stocks as described in Step 7.

    Pause point

    • Glycerol stocks can usually be stored for several years at −80 °C.

  22. To validate the glycerol stocks, repeat Steps 6 and 7 (glycerol stock preparation not necessary). Use the 96-well plasmid isolation kit (Step 17). Prepare overnight cultures directly from the stock.

    Critical step

    • Avoid thawing of the glycerol stocks by keeping them on dry ice.

    Pause point

    • The remaining plasmid solution can be kept at −20 °C for several years.

Figure 6: DNA analysis of the plasmid amplification by agarose gel electrophoresis.
Figure 6

Analytical agarose gel (1%) with a 1-kb marker (M) using the PCR product (5 μl) after DpnI digestion. DNA is visualized by ethidium bromide. Each lane corresponds to one mutation.

Box 2: Preparation of chemically competent cells in 96-deep-well plates • TIMING 1.5 d

Sav library expression using ZYP-5052 medium

Timing: 3–4 d

  1. Prepare a sterile 24-deep-well plate, and fill the wells with LB (2 ml) containing kanamycin (50 μg ml−1). Use one 24-deep-well plate for each amino acid position. Inoculate 12 wells with the twelve different mutants to be introduced at this position, three with pET-24a-empty vector controls, three with pET-24a SavK121A and three with pET-24a SavWT. The remaining three are sterile controls. Seal the plate with a permeable membrane. Incubate the plates overnight (210 r.p.m., 37 °C).

  2. Transfer the overnight culture (10 μl) into a new sterile 24-deep-well plate containing the modified ZYP-5052 medium (6 ml; for the 24-h expression) and kanamycin (50 μg ml−1). Seal the plates with a permeable membrane and incubate them with shaking (210 r.p.m., 30 °C).

    Critical step

    • To increase the collected Sav amount, it is recommended to set up two identical plates simultaneously. Both plates can be pooled for screening purposes. For the 24-h expression protocol, add a sterile pipette tip (0.1–10 μl) into each well and use the modified ZYP-5052 medium. For the 48-h expression protocol, addition of a pipette tip has no effect, and the normal ZYP-5052 medium should be used.

  3. Determine the OD600 of each well culture after 24 h; they should typically have an OD of 6–7. Centrifuge (4,500g, 10 min, 4 °C) and discard the supernatant. Set the plate on ice for 2 min. Invert the plate on a tissue to remove remaining liquid and wait for 5 min. Freeze the pellet at −20 °C.

    Pause point

    • The cell pellet containing Sav can be stored at −20 °C for several weeks.

  4. Thaw the cell pellets and resuspend them in 0.2 ml (ATHase, see Step 28(A)) or 0.4 ml (metathesase, see Step 28(B)) of lysis buffer. Do not use EDTA, as this inhibits ArMs. Incubate the pellets for 30 min at 37 °C and 200 r.p.m. Freeze them for 1 to 2 h at −20 °C. Thaw the cell lysate at RT.

    Critical step

    • The lysate should be a homogeneous suspension. If not, add more DNase I (1 μg ml−1 to the initial concentration) and incubate the mixture with shaking (30 min, RT).

  5. To prepare the CFE, clear the cell suspension by centrifugation of the plates (4,500g, 4 °C, 1–2 h). As an alternative, transfer the cell lysate into tubes (1.5 ml) and use a table centrifuge (21,100g, 20 min, 4 °C).

    Pause point

    • CFE containing Sav can be stored at 4 °C for several weeks or at −20 °C for up to 6 months.


ATHase or metathesase screening

  1. At this point in the procedure, you can proceed with ATHase (option A) or metathesase (option B) screening.

    1. ATHase screening

      1. Determine Sav FBS using the B4F absorption-fluorescence assay. One 96-well plate allows for the determination of the Sav FBS concentration of one 24-well plate containing the twelve mutants at one amino acid position of Sav. Perform all measurements, including the calibration curve, in triplicate. Use rows A-H and columns 1–3 of a 96-well plate for the calibration curve with commercial full-length streptavidin. Use the remaining wells to determine the Sav FBS of the twelve mutants for one amino acid position, as well as all controls (wild-type (WT) Sav, K121A Sav and empty vector, each in triplicate). First, prepare the B4F working solution by combining 5.6 ml of FBS assay buffer and 0.8 ml of 0.4 mM B4F solution.

        Critical step

        • As B4F is light-sensitive, protect the solutions with aluminum foil. Typically, a subsequent assay dilution of the samples by 5–10 is sufficient to remain in the range of the calibration curve under these conditions. By changing the amount of FBS assay buffer, other dilutions can be achieved.

      2. Use transparent 96-well plates (e.g., Brand, 350 μl per well, flat bottom). Measure each sample (final volume 100 μl) in triplicate. Follow the pipetting scheme for the calibration curve (rows A–H) displayed in the table below. Pipette the following into columns 1–3.

         Amount (μl)
        ComponentRow ARow BRow CRow DRow ERow FRow GRow H
        Final FBS Sav concentration (μM)05101520253040
        B4F working solution1010101010101010
        Sav stock (100 μM)05101520253040
        FBS assay buffer9085807570656050

        Critical step

        • Follow the table in order during pipetting to avoid air bubbles. Hold the pipette vertically, and contact the meniscus and the vessel wall with the tip when releasing the liquid.

      3. For measurements with the samples, pipette the B4F working solution (80 μl in each well) into the remaining 72 wells. Gently add 20 μl of the CFE into each well. Perform a triplicate determination for each sample. Shake the plate in the reader at 306 r.p.m. for 2 min. Incubate for 15 min. Determine the absorbance (493 nm) and the fluorescence (excitation 485 nm; emission 520 nm) at 25 °C.

        Critical step

        • Check the plate for air bubbles. Bubbles strongly influence the resulting absorbance, although the presence of air bubbles does not markedly affect the fluorescence determination.


      4. Typical standard curves are displayed in Figure 7. Data points yielding Sav FBS <10 μM or Sav FBS >35 μM should be repeated at higher or lower concentrations, respectively. Subtract the empty vector value and multiply by the assay dilution factor (5 when using of 20 μl of CFE, see Step 26).

      5. Diamide treatment. On the basis of the Sav FBS determination, add the appropriate amount of cell lysate or CFE to an HPLC vial with a magnetic stir bar. Each vial should contain a final Sav FBS = 100 μM. Add the diamide stock solution (5 μl) and preincubate the solution by stirring (>2 h). After dilution to 200 μl, the final diamide concentration should be 5 mM.

      6. Catalysis setup. Add 2× concentrated reaction buffer (100 μl) into the preincubated cell lysates or the CFEs.

      7. Dilute the samples with MQ water, so that the final volume after addition of the iridium and the substrate stock solutions reaches 200 μl.

      8. Add the iridium stock solution (5 μl) and incubate the solutions by stirring (5 min). After dilutions, the final iridium concentration [Cp*Ir(Biot-p-L)Cl] (compound 3) should be 50 μM.

      9. Initiate the reaction by adding the substrate stock solution (5 μl) and seal the HPLC vials. The final substrate concentration should be 10 mM.

      10. Stir the reaction at RT for 48 h (e.g., in a standard plastic box for cryotubes).

      11. Workup of catalysis reactions and analysis. Dilute the reactions by the addition of MQ water (500 μl).

      12. Transfer the reaction mixtures to a 2-ml polypropylene (PP) tube.

      13. Add NaOH (50 μl of 20% (wt/wt) solution).

      14. Extract the reaction mixture twice with dichloromethane (2 × 1 ml).

        Critical step

        • If a gel-like layer forms during extraction, especially when using cell lysates, spin (2,000g, 1 min, RT) the 1.5-ml PP tubes before separating the organic layer.


      15. Combine the organic phases, transfer them into a new 2-ml PP tube and dry them by adding a spatula tip of anhydrous Na2SO4.

      16. Spin down the tubes (2,000g for 5 min at RT).

      17. Transfer the supernatant into a HPLC vial using a Pasteur pipette.

      18. Prepare the isocratic HPLC elution mixture containing 99.44% hexane, 0.5% isopropanol and 0.06% diethylamine.

      19. Analyze the samples by HPLC using a Chiralpak IC column. Use a flow rate of 1 ml min−1 at 25 °C, and determine the absorbance at 265 nm.

      20. For comparison purposes, the retention times are as follows: (S)-1-phenyl-1,2,3,4-tetrahydroisoquinoline (S)-2, 7.6 min; (R)-1-phenyl-1,2,3,4-tetrahydroisoquinoline (R)-2, 10.5 min; and 1-phenyl-3,4-dihydroisoquinoline 1, 16 min.

      21. Determine the enantiomeric excess (ee) for each sample. The yield can be estimated using a response factor (substrate 1)/(product 2) = 12.535.

        Critical step

        • Determine the response factor experimentally by measuring at least three different substrate (1) to product (2) ratios.

      Timing: 3 d + 10 h of measurement time

    2. Metathesase screening

      1. Small-scale iminobiotin immobilization. Transfer the CFEs (400 μl in PBS buffer) from the 24-deep-well plates into a 96-deep-well plate. Add iminobiotin binding buffer (1 ml) to each well to adjust the pH to 9. Add the iminobiotin-Sepharose beads (100 μl, shake the suspension thoroughly before pipetting; because of the dilution, pipetting and mixing is straightforward). Incubate the 96-deep-well plate with rapid shaking (at 800 r.p.m. for 1 h at RT).


      2. Centrifuge the plate (5,300g, 15 min, RT) and discard the supernatant. Resuspend the beads in iminobiotin binding buffer (1 ml), centrifuge the beads (5,300g, 15 min, RT), and discard the supernatant. Repeat this washing step three times to avoid deactivation of the catalyst. Use the thoroughly washed beads for catalysis.

      3. Catalysis setup. To each well containing iminobiotin beads, add the metathesase reaction buffer (97 μl).

      4. Add a solution of the catalyst (6) in DMSO (3 μl of the 3.33 mmol stock solution) to each well (to obtain the 3.33 mmol solution, take a 0.2 mg aliquot and add 69 μl of DMSO) to obtain a final concentration of 50 μM.


      5. Add the solution of substrate (100 μl of the 20 mM stock solution) to each well to obtain a final substrate concentration of 10 mM, and place the 96-deep-well plate into a shaking incubator (37 °C, 200 r.p.m., 12 h).

      6. Workup of catalysis reactions and analysis. Remove the plate from the incubator and add the internal standard solution (100 μl) to each well and methanol (700 μl). Next, centrifuge the plate (5,346g, 25 min, RT). From each supernatant, transfer an aliquot (250 μl) into a new 96-deep-well plate and dilute with MQ water (750 μl).

      7. Schedule 10 h for an entire 96-well plate. Analyze the wells on an Acquity UPLC H-Class Bio from Waters (or a similar instrument, retention times may vary) using a BEH C18 column (1.7 μm, 2.1 × 50 mm). Use MQ water + 0.1% (vol/vol) TFA as solvent A and acetonitrile + 0.1% (vol/vol) TFA as solvent B. Apply a gradient method as follows: 0 min 80% A, 20% B; 1 min 80% A, 20% B; 3 min 10% A, 90% B; 3.5 min 80% A, 20% B; and 4.5 min 80% A, 20% B. Plan 5 min for each sample.

      8. For quantification purposes, integrate the peaks of the absorbance at 210 nm. The retention times are as follows: starting material 4, 2.15 min; internal standard, 0.47 min; product 5, 0.67 min.

      9. Determine the concentration of the product with a calibration curve. For this, record samples of different concentrations of the purified product, 500, 300, 200, 100, 80, 50, 30 and 10 μM, as well as constant concentrations of internal standard (250 μM). A linear regression is obtained by plotting area(product)/area(standard) against the calculated product concentration.

      10. The total TON is obtained by dividing the calculated concentration in the sample by the catalyst concentration (50 μM) and by multiplying it with the dilution factor from the workup (20).

      Timing: 1 d + 10 h measurement time for an entire 96-well plate

Figure 7: Typical calibration curve for Sav K121A using the B4F-assay.
Figure 7

Absorption and relative fluorescence is displayed at increasing Sav concentrations. Measured data points are displayed as triangles. (a,b) The solid line corresponds to the linear regression: absorbance (λmax = 493 nm, each symbol (circle, square and triangle) corresponds to one independent determination) (a), and relative fluorescence (λexcit = 485 nm and λemis = 520 nm, each symbol (circle, square and triangle) corresponds to one independent determination) (b). RFU, relative fluorescence units.


Troubleshooting advice can be found in Table 1. For mutagenesis, Sav expression and Sav FBS, the following guidance applies:

Table 1: Troubleshooting table.

After preparation of a fresh stock of the template plasmid, it is recommended to validate it for the right Sav sequence by sequencing before mutagenesis. Prepare a positive control of the PCR by using a primer pair (Supplementary Table 1), which will give you a specific amplification. If, after site-directed mutagenesis, no amplified vector is visible on the gel, reduce the annealing temperature. If several amplified fragments are visible and no mutant is obtained, increase the annealing temperature. The transformation efficiency of competent cells can decrease during storage because of their low viability. It is thus suggested to always use a positive-control transformation using the pUC19 vector. Because of the lower transformation efficiency of the BL21 (DE3), the transformation into this strain is less effective in 24-deep-well plates. If no colonies are obtained, a transformation in 1.5-ml tubes can lead to a few c.f.u. This Sav expression protocol is highly robust, but reduced Sav yields can exceptionally be obtained with no apparent reason. When this is the case, it is recommended to perform a new transformation into BL21 (DE3) and to prepare a new glycerol stock resulting from an overnight culture inoculated by a single colony. During expression of the Sav library, it is recommended to include controls on each 24-deep-well plate to validate the screening for hits and background. For this purpose, wild-type Sav, SavK121A and pET-24a empty vector should be included in each 24-deep-well plate. The B4F-absorption-fluorescence assay has a limited precision window ranging from 10 to 40 μM Sav FBS. It is thus recommend to re-determine data for samples with Sav FBS <10 μM or Sav FBS >35 μM after having adapted the concentration. It is important to subtract the background spectrum for the pET-24a empty vector (both for the absorbance and fluorescence determination). As residual biotin and biotinylated E. coli proteins may be present in cell lysates, the B4F assay may underestimate the Sav concentration: it is, however, highly reliable for the Sav FBS determination, which is the relevant concentration for catalysis with ArMs.


Steps 1–22, site-directed mutagenesis in 96-well format: 9–10 d

 Steps 1–16: 3–4 d

 Steps 17 and 18 (depending on shipping duration): 4 d

 Steps 19–22: 2 d

Steps 23–27, Sav library expression using ZYP-5052 medium: 3–4 d

 Step 23: 1 d

 Steps 24 and 25: 1–2 d

 Steps 26 and 27: 6 h

Step 28A, ATHase screening: 3 d + 10 h measurement time

 Step 28A(i–iv): 1–3 h

 Step 28A(v): 3 h

 Step 28A(vi–x): 2 d

 Step 28A(xi–xxi): 12 h for 18 samples

Step 28B, metathesase screening: 1 d + 10 h measurement time for an entire 96-well plate

 Step 28B(i,ii): 3.5 h

 Step 28B(iii–v): 19 h

 Step 28B(vi–x): 12 h for an entire 96-well-plate analysis

Box 1, primer design for noncomplementary, overlapping site-directed mutagenesis primer pair in 96-well plate format: 1–2 d

Box 2, preparation of chemically competent cells in 96-deep-well plates: 1.5 d

Anticipated results

Upon sequencing a single colony for each mutagenesis reaction, coverage of 80–90% of the targeted Sav mutants was achieved after PCR and transformation in E. coli TOP10. This value is achieved by sequencing one colony for each mutant. If amplification is visible on the analytical agarose gel, sequencing an additional one or two colonies from the transformation plate leads to nearly complete coverage of all targeted Sav mutants. By following this protocol, 335 Sav mutants from the targeted 336 Sav mutants are obtained.

Comparable ODs and Sav yields are obtained for each well when using the ZYP-medium expression protocol in 24-deep-well plates. Routinely, ≥1 mg Sav (corresponding to 15 nmol tetrameric Sav or 240 μM Sav FBS in 250 μl) is obtained per well after 24 h using the modified ZYP and adding a pipette tip. By following this procedure, 250 μl of CFE can be obtained. This is sufficient to determine the Sav FBS and to perform two catalytic runs either for the ATHase or the metathesase reaction (100 μM Sav FBS required). Upon increasing the expression time to 48 h, the Sav FBS typically increases by 40%. For this purpose, no pipette tip should be added to the well, and the standard ZYP medium is recommended.

Both the absorption and the fluorescence Sav FBS single-point determinations afford 10–20% errors. This is perfectly acceptable for the screening procedure, as catalysis is performed using twice the Sav FBS concentration versus the biotinylated catalyst, thus ensuring quantitative binding of the latter to Sav. The standard curves either for absorbance or fluorescence display high linearity with an R2 = 0.95–0.99 (Fig. 7). By using this dilution, the CFE samples typically yield Sav FBS = 20–30 μM. For CFE of the empty pET-vector, the background absorption or fluorescence corresponds to 1–5 μM Sav FBS.

Because of the high affinity of biotinylated probes for Sav, the ATHase or metathesase are immediately formed upon addition of the biotinylated catalyst to Sav isoforms. The success rate for catalysis depends on factors including activity, stability and expression level of the specific mutants. The excellent reproducibility of the protocol may allow the evaluation of NNK libraries, as hits of different screening plates can be compared. The success rate for catalysis is improved because of the normalization of the FBS Sav concentration present in each sample before catalysis: option A is normalized by the FBS determination and option B by the specific binding capacity of the applied beads if Sav is present in excess in the CFE.

Artificial transfer hydrogenase (option A)

The protocol allows highly reproducible preparation of cell lysates and corresponding CFEs containing the overexpressed Sav isoforms. Sav is produced in high concentrations, which allows one to set up two catalysis experiments (200-μl final volume) from a 6-ml cell culture (corresponds to 0.25 ml of cell lysate).

Preincubation of either cell lysates or CFEs for 2 h with 5 mM diamide, before addition of the transition metal catalyst, leads to the partial restoration of the activity of ArMs performed in the presence of purified Sav samples. Screening results from ATHase in the [Cp*Ir(biot-p-L)Cl] 3 K121A Sav are summarized in Table 2. As can be appreciated, upon addition of diamide, the catalytic activity is partially restored for samples containing either cell lysates or CFEs (Table 2).

Table 2: Expected results for the transfer hydrogenase (Step 28A).

Ring-closing olefin metathesis (option B)

CFEs and cell lysates contain unidentified catalyst poisons that inhibit ring-closing metathesis on substrate 4. Upon immobilization of Sav from CFEs on the iminobiotin-Sepharose beads, impurities from the CFEs can be washed away, thus partially restoring metathesase activity. Duplicate measurements performed without protein, with Sav K121A mutant and pET-24a empty vector CFE typically yield the results indicated in Table 3.

Table 3: Expected results for the metathesis (Step 28B).



  1. 1.

    et al. Protein design: toward functional metalloenzymes. Chem. Rev. 114, 3495–3578 (2014).

  2. 2.

    , & Artificial metalloenzymes in asymmetric catalysis: key developments and future directions. Adv. Synth. Catal. 357, 1567–1586 (2015).

  3. 3.

    Artificial metalloenzymes and metallopeptide catalysts for organic synthesis. ACS Catal. 3, 2954–2975 (2013).

  4. 4.

    Artificial metalloenzymes based on the biotin–avidin technology: enantioselective catalysis and beyond. Acc. Chem. Res. 44, 47–57 (2010).

  5. 5.

    & Conversion of a protein to a homogeneous asymmetric hydrogenation catalyst by site-specific modification with a diphosphinerhodium(I) moiety. J. Am. Chem. Soc. 100, 306–307 (1978).

  6. 6.

    , & Catalytic hydrogenation of itaconic acid in a biotinylated pyrphos–rhodium(I) system in a protein cavity. Tetrahedron Asymmetr. 10, 1887–1893 (1999).

  7. 7.

    Directed evolution of stereoselective hybrid catalysts. Top. Organomet. Chem. 25, 63–92 (2009).

  8. 8.

    et al. Artificial metalloenzymes for enantioselective catalysis based on biotin-avidin. J. Am. Chem. Soc. 125, 9030–9031 (2003).

  9. 9.

    , & Artificial metalloenzymes based on biotin-avidin technology for the enantioselective reduction of ketones by transfer hydrogenation. Proc. Natl. Acad. Sci. USA 102, 4683–4687 (2005).

  10. 10.

    et al. Artificial metalloenzymes for asymmetric allylic alkylation on the basis of the biotin–avidin technology. Angew. Chem. Int. Ed. 47, 701–705 (2008).

  11. 11.

    , , & Biotinylated Rh(III) complexes in engineered streptavidin for accelerated asymmetric C-H activation. Science 338, 500–503 (2012).

  12. 12.

    , , , & Artificial metalloenzymes for olefin metathesis based on the biotin-(strept)avidin technology. Chem. Commun. 47, 12065–12067 (2011).

  13. 13.

    , & Incorporation of biotinylated manganese-salen complexes into streptavidin: new artificial metalloenzymes for enantioselective sulfoxidation. J. Organomet. Chem. 694, 930–936 (2009).

  14. 14.

    et al. Artificial metalloenzyme for enantioselective sulfoxidation based on vanadyl-loaded streptavidin. J. Am. Chem. Soc. 130, 8085–8088 (2008).

  15. 15.

    et al. OsO4·streptavidin: a tunable hybrid catalyst for the enantioselective cis-dihydroxylation of olefins. Angew. Chem. Int. Ed. 50, 10863–10866 (2011).

  16. 16.

    , , & Aqueous oxidation of alcohols catalyzed by artificial metalloenzymes based on the biotin–avidin technology. J. Organomet. Chem. 690, 4488–4491 (2005).

  17. 17.

    & Expression of a cloned streptavidin gene in Escherichia coli. Proc. Natl. Acad. Sci. USA 87, 142–146 (1990).

  18. 18.

    , & Site-directed mutagenesis studies of the high-affinity streptavidin-biotin complex: contributions of tryptophan residues 79, 108, and 120. Proc. Natl. Acad. Sci. USA 92, 1754–1758 (1995).

  19. 19.

    & Bioorthogonal chemistry: fishing for selectivity in a sea of functionality. Angew. Chem. Int. Ed. 48, 6974–6998 (2009).

  20. 20.

    et al. Synthetic cascades are enabled by combining biocatalysts with artificial metalloenzymes. Nat. Chem. 5, 93–99 (2013).

  21. 21.

    , , & Genetically engineered avidins and streptavidins. Cell. Mol. Life Sci. 63, 2992–3017 (2006).

  22. 22.

    & The Strep-tag system for one-step purification and high-affinity detection or capturing of proteins. Nat. Protoc. 2, 1528–1535 (2007).

  23. 23.

    & Introduction to avidin-biotin technology. Methods Enyzmol. 184, 5–13 (1990).

  24. 24.

    , , , & Increasing the stability of an enzyme toward hostile organic solvents by directed evolution based on iterative saturation mutagenesis using the B-FIT method. Chem. Commun. 46, 8657–8658 (2010).

  25. 25.

    et al. Artificial transfer hydrogenases for the enantioselective reduction of cyclic imines. Angew. Chem. Int. Ed. 50, 3026–3029 (2011).

  26. 26.

    , , & Neutralizing the detrimental effect of glutathione on precious metal catalysts. J. Am. Chem. Soc. 136, 8928–8932 (2014).

  27. 27.

    et al. X-ray structure and designed evolution of an artificial transfer hydrogenase. Angew. Chem. Int. Ed. 47, 1400–1404 (2008).

  28. 28.

    , & Iterative saturation mutagenesis on the basis of B factors as a strategy for increasing protein thermostability. Angew. Chem. Int. Ed. 45, 7745–7751 (2006).

  29. 29.

    , , & Site-directed mutagenesis in one day with >80% efficiency. Strategies 9, 3–4 (1996).

  30. 30.

    et al. Chemo-genetic optimization of DNA recognition by metallodrugs using a presenter-protein strategy. Chemistry 16, 12883–12889 (2010).

  31. 31.

    et al. A dual anchoring strategy for the localization and activation of artificial metalloenzymes based on the biotin-streptavidin technology. J. Am. Chem. Soc. 135, 5384–5388 (2013).

  32. 32.

    , & An efficient one-step site-directed and site-saturation mutagenesis protocol. Nucleic Acids Res. 32, e115 (2004).

  33. 33.

    Protein production by auto-induction in high-density shaking cultures. Protein Expr. Purif. 41, 207–234 (2005).

  34. 34.

    et al. Structural, kinetic, and docking studies of artificial imine reductases based on biotin–streptavidin technology: an induced lock-and-key hypothesis. J. Am. Chem. Soc. 136, 15676–15683 (2014).

  35. 35.

    , , & Rapid estimation of avidin and streptavidin by fluorescence quenching or fluorescence polarization. Biochim. Biophys. Acta 1427, 44–48 (1999).

  36. 36.

    & A simple spectrophotometric streptavidin-biotin binding assay utilizing biotin-4-fluorescein. J. Biochem. Biophys. Methods 70, 873–877 (2008).

  37. 37.

    , & Organometallic chemistry in protein scaffolds. in Protein Engineering Handbook Vol. 3 (eds. Lütz, S. & Bornscheuer, U.T.) 215–238 (Wiley-VCH, Weinheim, 2012).

  38. 38.

    et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nat. Chem. Biol. 5, 593–599 (2009).

  39. 39.

    , , & Biotinylated metathesis catalysts: synthesis and performance in ring closing metathesis. Catal. Lett. 144, 373–379 (2014).

  40. 40.

    , & Synthesis of 1-phenyl-5H-2-benzazepines by ring expansion of 1-phenyl-1,2-dihydroisoquinolines. J. Org. Chem. 51, 4147–4150 (1986).

  41. 41.

    Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J. Bacteriol. 62, 293–300 (1951).

  42. 42.

    Studies on transformation of Escherichia coli with plasmids. J. Mol. Biol. 166, 557–580 (1983).

  43. 43.

    , & Techniques for transformation of E. coli. in DNA Cloning (Oxford University Press, 1995).

  44. 44.

    et al. Preparation of 1,4-diazepane-3,5-dione derivatives as chymase inhibitors and pharmaceutical use thereof. Japanese patent no. WO 2010053182A1 (2010).

  45. 45.

    et al. Organikum 19th edn. (Barth Verlagsgesellschaft, 1993).

  46. 46.

    Codon usage: nature's roadmap to expression and folding of proteins. Biotechnol. J. 6, 650–659 (2011).

Download references


T.R.W. thanks the Swiss National Science Foundation (grants 200020_162348 and the NCCR (National Centres of Competence in Research) Molecular Systems Engineering) and the Seventh Framework Programme Project METACODE (KBBE (Knowledge-Based BioEconomy), 'Code-engineered new-to nature microbial cell factories for novel and safety enhanced bioproduction') and the US National Institutes of Health (NIH; grant GM050781) for generous support. M.H. thanks the SNI (Swiss Nanoscience Institute) for a Ph.D. scholarship. The authors are happy to provide the library free of charge upon request to academic institutions.

Author information


  1. Department of Chemistry, University of Basel, Basel, Switzerland.

    • Hendrik Mallin
    • , Martina Hestericová
    • , Raphael Reuter
    •  & Thomas R Ward


  1. Search for Hendrik Mallin in:

  2. Search for Martina Hestericová in:

  3. Search for Raphael Reuter in:

  4. Search for Thomas R Ward in:


T.R.W. conceived and established the concept of ArMs using the Sav-biotin technology and planned the experiments. H.M. planned experiments, designed the mutant library and established the expression protocol. M.H. and R.R. designed and performed the screening catalysis experiments. T.R.W., H.M., M.H. and R.R. wrote the manuscript.

Competing interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to Hendrik Mallin or Thomas R Ward.

Integrated supplementary information

Supplementary information

PDF files

  1. 1.

    Supplementary Text and Figures

    Supplementary Figures 1–7, Supplementary Methods, Supplementary Table 1 and Supplementary Data

About this article

Publication history




By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.