Proximity-enhanced protein crosslinking through an alkene-tetrazine reaction

Abstract

The inverse electron demand Diels–Alder (iEDDA) reaction between a tetrazine and a strained alkene has been widely explored as useful bioorthogonal chemistry for selective labeling of biomolecules. In this work, we exploit the slow reaction between a non-conjugated terminal alkene and a tetrazine, and apply this reaction to achieving a proximity-enhanced protein crosslinking. In one protein subunit, a terminal alkene-containing amino acid was site-specifically incorporated in response to an amber nonsense codon. In another protein subunit, a tetrazine moiety was introduced through the attachment to a cysteine residue. Fast protein crosslinking was achieved due to a large increase in effective molarity of the two reactants that were brought to close proximity by the two interacting protein subunits. Such a proximity-enhanced protein crosslinking is useful for the study of protein-protein interactions.

Graphical Abstract

1. Introduction

Proximity-enhanced protein modifications through non-canonical amino acid (ncAA) mutagenesis has emerged as a useful tool to study and manipulate protein function [1]. These ncAAs contain mild electrophiles (e.g., haloalkanes [2–4], fluorosulfate [5], isothiocyanate [6], and vinylsulfonamide [7]) that only react with nearby nucleophilic residues (e.g., cysteine and lysine). The reaction rate increase benefits from an elevated effective molarity of reactants, which is commonly observed in enzyme-catalyzed reactions. Based on the redox chemistry of sulfur, our group also developed a probe that allowed an oxidation-induced and proximity-enhanced protein–protein crosslinking via an in situ formed mild Michael acceptor [8].

The current focus of the field is on reactions between exogenously introduced electrophiles and endogenous nucleophiles within target proteins. We seek to expand the toolbox for proximity-enhanced protein modifications beyond nucleophilic substitution reactions. To this end, we turn our attention to cycloaddition reactions, which display excellent bioorthogonality. Recently, we reported the use of terminal alkene (i.e., styrene) and tetrazine chemistry for protein modifications [9]. In comparison to the highly reactive strained alkenes [10–25], unstrained styrene has much slower reaction rate with tetrazine, but represents an intriguing alternative since styrene shows good cellular stability and is easy to synthesize. In addition to styrene, non-conjugated terminal alkenes have also been examined for their reaction with tetrazine, displaying even slower kinetics [26]. While such a slow reaction is not suitable for protein labeling at low reagent concentrations for biological studies, it provides an intriguing opportunity to develop a proximity-induced bioconjugation strategy. In this work, we demonstrate that the reaction between a non-conjugated terminal alkene and tetrazine can be used for proximity-enhanced protein crosslinking.

2. Results and discussion

2.1. The overall design and the synthesis of reagents.

The reaction between non-conjugated terminal alkenes and tetrazines have been studied before [27–30]. Most of these reactions proceed at much slower rates than those using strained alkenes. We examined a model reaction between pent-1-ene-4-ol (1 mM) and di-pyridyl-s-tetrazine (1 mM) in acetonitrile/H₂O (1:1). No reaction was detected after an hour and only trace amount of product was formed with 24 h of incubation. We envision that even less reaction product can be formed under biological settings where the concentrations of biomolecules and/or reagents are low. In contrast, this reaction can be potentially suitable for the study of biomolecular interactions (e.g., protein-protein), where fast reaction can only happen when the two reactants, which are separately grafted on two macromolecules, are brought to close proximity via specific interactions. To demonstrate such an application, our strategy is to install one reactant onto a target protein through non-canonical amino acid (ncAA) mutagenesis and to attach the other reactant onto a partner protein through the chemical modification of a cysteine residue.

The chemical structure of the ncAA, named as K-Alkene, is shown in Scheme 1. The synthesis of K-Alkene started with a commercially available Fmoc-lysine (1, Scheme 1). The free carboxylic acid group of Fmoc-lysine was first protected as a methyl ester by using thionyl chloride in methanol solution. Then methyl ester 2 reacted subsequently with pent-4-en-1-ol in the presence of a carbamatyl condense reagent, 1,1’-carbonyldiimidazole (CDI), and triethylamine to yield compound 3. After a final deprotection step with 1 equivalent NaOH in THF and water, K-Alkene sodium salt was obtained through lyophilization with an overall yield of 40.3%.

Scheme 1.

(a) MeOH, SOCl₂, reflux, 4h; (b) CDI, NEt₃, CH₂Cl₂; (c) NaOH, THF/H₂O.

To install a tetrazine group onto a protein, we used bifunctional molecules that contain both tetrazine and a cysteine-targeting maleimide moiety. The structures of the two tetrazine reagents, Tet-1 and Tet-2, are shown in Scheme 2A. They contain linkers with different length between tetrazine and maleimide. The overall length of the molecule can potentially be important for efficient crosslinking of two interacting proteins. Tet-1 is commercially available and Tet-2 was synthesized. The synthesis of Tet-2 started with thiocarbohydrazide (4, Scheme 2B), which reacted with iodomethane to give a methylated product 5. The resulting compound 5 went through a cyclization reaction with tri-methyl orthobenzoate to yield a thiol-benzyl-1,4-dihydrotetrazine intermediate. After oxidation with (diacetoxyiodo)-benzene at low temperature, 3-(methylthio)-6-phenyl-s-tetrazine (6) was formed as a red crystal with an overall yield of 47% for the first 3 steps. In the next step, boc-4-aminomethyl-phenyl boronic acid was coupled with compound 6 using Pd(dppf)Cl₂ and Ag₂O as catalysts. The resulting product 8 was deprotected and subsequently converted into Tet-2. While Tet-1 and Tet-2 contain both alkene and tetrazine functional groups, we do not expect that the alkene of maleimide can react with tetrazine at a significant rate since a fast iEDDA reaction between alkene and tetrazine requires an electron-rich dienophile. To verify this notion, we monitored the potential self-reaction of Tet-1 and Tet-2 by NMR. No significant reaction was observed within 24 h of incubation.

Scheme 2.

(A) Chemical structures of Tet-1 and Tet-2. (B) (a) MeI, EtOH, 60 °C; (b) (1) Trimethyl orthobenzoate, DMF, 50 °C; (2) tributyl phosphine, H₂O; (3) (diacetoxyiodo)-benzene, CH₂Cl₂, rt; (c) Pd(dppf)Cl₂, Ag₂O, 60 °C; (d) TFA, CH₂Cl₂, rt; (e) Et₃N, CH₂Cl₂, rt.

2.2. Genetic incorporation of K-Alkene

Since K-Alkene is a lysine derivative (Scheme 1), we screened a collection of reported pyrrolysine tRNA-synthetase (PylRS) mutants that are known to recognize a variety of lysine analogs as their substrates. In the screening, a GFP mutant (sfGFP-Asn149TAG) that contains an amber nonsense codon at position Asn149 was used as the reporter [31, 32]. The activity of PylRS mutants on charging K-Alkene onto amber suppressor tRNA was therefore directly linked to the amber suppression efficiency and the expression level of sfGFP-Asn149TAG. Higher fluorescence intensity would indicate higher PylRS activity on K-Alkene. As shown in Figure 1A, among all PylRS variants examined, the best three that supported the efficient synthesis of full-length sfGFP in the presence of K-Alkene are AbkRS [33], ACPKRS [34], and DizPKRS [35]. Furthermore, no significant expression of sfGFP was observed in the absence of K-Alkene. The ACPKRS (L274A, C313V, Y349F) was chosen for the following work. This synthetase variant displayed the best efficiency towards K-Alkene and is encoded by plasmid pBK-ACPKRS.

Figure 1.

Genetic incorporation of K-Alkene. (A) Fluorescence readings of E. coli cells expressing the indicated PylRS mutants and an sfGFP mutant that contains an amber mutation at position N149. K-Alkene was included in culture media at either 0 mM (blue bar) or 1mM (filled orange bar). Fluorescence intensity was normalized to cell growth. (B) Tandem mass spectrometry analysis. Both the y ion and b ions confirmed the K+112 modification at the desired peptide fragment in the peptide sequence LEYNYNSH-(K-Alkene)-VYITADK.

To further evaluate the fidelity and the site-specific incorporation of K-Alkene, we purified sfGFP-Asn149K-Alkene mutant and conducted analysis with tandem mass spectrometry (MS/MS; Figure 1B). The peptide fragment, LEYNFNSH(K+112)VYITADK, that covered position 149 in sfGFP has a K+112 (K represents a natural lysine residue; K-Alkene has a molecular weight that is 112 higher than lysine.) modification at the intended incorporation site. The modification is consistent with the molecular weight of K-Alkene. No incorporation of any canonical amino acids was detected at this site. The mass spectrometry result confirmed that K-Alkene was site-specifically incorporated into sfGFP with high fidelity.

2.3. Protein crosslinking

We used glutathione S-transferase (GST) as the model to examine a proximity-enhanced protein crosslinking via the terminal alkene-tetrazine reaction. According to the crystal structure of GST dimer (PDB:1Y6E; Figure S1), the distance between C84 in one GST monomer and D61 in the other monomer is 12.24 Å. In our experimental design, K-Alkene was incorporated at position D61 and tetrazine reagents were attached to C84. The length of K-Alkene (12.14 Å) and tetrazine reagents (Tet-1, 21.22 Å and Tet-2, 15.13 Å) should enable sufficient interaction of the two reactants.

To genetically incorporate K-Alkene into position 61 of GST, plasmid pLei-GST-D61TAG was constructed. This plasmid contains a GST variant with an amber mutation at position 61. As a control, we further modified pLei-GST-D61TAG to yield pLei-GST-D61TAG-C84S, in which an additional C84S mutation was introduced. This C84S mutation would prevent the attachment of tetrazine reagents to position 84 of GST and should not lead to protein crosslinking. Protein expression experiments were conducted by co-transforming E. coli cells with pBK-ACPKRS and pLei-GST-D61TAG (or pLei-GST-D61TAG-C84S as a control). Proteins, GST-D61K-Alkene and GST-D61K-Alkene-C84S, were purified by affinity chromatography using Ni-NTA resin (Figure S2). Again, GST mutants could only be expressed in the presence of K-Alkene, which confirmed the high-fidelity incorporation of K-Alkene by ACPKRS.

To examine the efficiency of crosslinking, we first conducted an experiment by incubating GST-D61K-Alkene and Tet-1 for over 20 h and analyzed the formation of GST dimer at multiple timepoints. As shown in Figure 2A, significant crosslinking product was detected at 5 min and the amount continued to grow over time. The GST dimer formation reached over 50% and 95% after 40 min and 20 h of incubation, respectively. Next, we examined crosslinking of three GST variants (wild-type GST, GST-D61K-Alkene, and GST-D61K-Alkene-C84S) in the presence of two different tetrazine reagents (Tet-1 and Tet-2). Since the wild-type GST does not contain an alkene functional group and the GST-D61K-Alkene-C84S mutant does not have C84 for the attachment of tetrazine reagent, we should not see any crosslinking. Indeed, as shown in Figure 2B, no apparent crosslinking product was detected with the wild-type GST and the GST-D61K-Alkene-C84S mutant (Lanes 1–3 and 7–9). Between the two tetrazine reagents, much more crosslinking product was formed in the presence of Tet-1 in comparison to Tet-2. According to their structures (Scheme 2 A), Tet-1 is longer and more flexible, which likely enabled a better interaction with K-Alkene on the other GST monomer. Since the product of the reaction between terminal alkene and tetrazine is fluorescent [9], we also sought to detect fluorescence signal of crosslinking products. As shown in the bottom panel of Figure 2B, a detectable fluorescence signal was indeed observed for the crosslinking of GST-D61K-Alkene in the presence of Tet-1. We further conducted a protein crosslinking experiment and monitored emission signals in a time-dependent manner. As shown in Figure S4, a detectable and steady increase in fluorescence intensity was observed over time.

Figure 2.

Protein crosslinking. (A) SDS-PAGE analysis of crosslinking of GST-D61K-Alkene in the presence of Tet-1. (B) Crosslinking of three GST variants in the presence of either Tet-1 or Tet-2. The reaction was run for 60 min. The top panel shows Coomassie blue stained gel and the bottom panel shows the fluorescent image (λ_ex = 302 nm; λ_em = 580 nm) of the same gel before Coomassie blue treatment. A dashed line in the top panel was drawn to align the relative position of GST dimer under each condition. For both (A) and (B), the concentration of GST-D61K-Alkene was 1 mg/mL (37 μM) and the concentration of Tet-1 was 1 mM.

3. Conclusions

We have shown that the slow reaction between terminal alkene and tetrazine can be used for a proximity-enhanced crosslinking of an interacting protein pair. The fluorogenic nature of this reaction can be especially useful for the study of protein-protein interactions. When alkene and tetrazine are separately grafted on two proteins, fluorescence signal can only be detected if the two proteins interact to each other and bring two functional groups to a close proximity. While the fluorescence signal in our demonstration is not very strong and the emission wavelength may not be ideal for certain biological studies, the fluorescence property of the bioconjugation product between terminal alkenes and tetrazines can certainly be improved by modifying the substituents of tetrazine. This will also make the fluorescence signal of the bioconjugation product more distinguishable from any background fluorescence within the system. We are currently working in this regard and will report results in due course.

4. Experimental section

4.1. Materials and General Methods.

All starting materials and reagents unless otherwise noted were obtained from commercial suppliers without further purification. Reactions were carried out under argon with freshly distilled solvents. CH₂Cl₂ was distilled from calcium hydride under argon. DMF was dried and distilled over calcium hydride under vacuum and stored over 4 angstrom sieves under argon. Flash chromatography was performed using silica gel 60 Å (234–400 mesh) obtained from Sigma-Aldrich. ¹H-NMR and ¹³C-NMR spectra were obtained using a Bruker Avance III 300 MHz or a Bruker Avance III 400 MHz spectrometer. Chemical shifts (δ) for NMR spectra were reported in parts per million (ppm) with deuterated solvents as internal standards (CDCl₃: H 7.26, C 77.0; DMSO-d₆: H 2.50, C 39.5; D₂O: 4.79; CD₃OD: H 3.35, 4.78, C 49.3).

4.2. Synthesis of K-Alkene

Methyl (((9H-fluoren-9-yl)methoxy)carbonyl)-L-lysinate hydrochloride (2)

To a solution of anhydrous methanol (50 mL) at 0 °C in ice-water bath, thionyl chloride (1.2 g, 713 μL, 10 mmol) was added dropwise with stirring. The resulting solution was warmed to room temperature. (((9H-fluoren-9-yl)methoxy)carbonyl)lysine hydrochloride salt (4g, 10 mmol) was added to the solution portion-wise. After the reaction mixture was stirred at room temperature overnight, solvent was evaporated under vacuum. The residue was re-dissolved in methanol and evaporated again to afford a white solid of 2 (quantitative yield). ¹H NMR (400 MHz, CD₃OD): δ 7.81 (d, J = 5.6 Hz, 2H), 7.67 (d, J = 7.2 Hz, 2H), 7.39–7.43 (m, 2H), 7.31–7.35(m, 2H), 4.42–4.46 (m, 1H), 4.34–4.38 (m, 2H), 3.74 (s, 3H), 2.91–2.95 (m, 2H), 1.45–1.91 (m, 6H).

Methyl N⁶-((pent-4-en-1-yloxy)-carbonyl)-lysinate (3)

To a solution of pent-4-en-1-ol (430 mg, 5 mmol) in 15 mL dry dichloromethane (DCM), a solution of 1, 1’-carbonyldiimidazole (CDI, 890 mg, 5.5 mmol) in 15 mL dry DCM was added at 0 °C in an ice-water bath with stirring. The resulting solution was warmed to room temperature and stirred for an additional 3 h. Compound 2 and triethylamine (NEt₃, 767 μL, 5.5 mmol) in 15 mL dry DCM was then added dropwise. The resulting reaction mixture was stirred at room temperature overnight. The reaction progress was monitored by TLC. After the completion of the reaction, 20 mL of DCM was added and the resulting mixture was sequentially washed with 1 N HCl (1 × 50 mL), saturated NaHCO₃ (1 × 50 mL), and brine (2 × 50 mL). The organic layer was dried over Na₂SO₄ and concentrated under vacuum. The crude product was further purified by silica gel flash chromatography (hexane:ethyl acetate = 5:1) to afford 1.2 g of the desired product as pale yellow solid with a 48.5 % yield. ¹H NMR (400 MHz, CDCl₃): δ 5.79–5.86 (m, 1H), 4.99–5.07 (dd, J = 17.2, 1.6 Hz, 2H), 4,72 (br, 1H), 4.0(t, J = 6.4 Hz, 2H), 3.74 (s, 3H), 3.44–3.47 (m, 1H), 3.18–3.20 (m, 2H), 2.12–2.15 (m, 2H), 1.43–1.74 (m, 8H).

Sodium N⁶-((pent-4-en-1-yloxy)carbonyl)-L-lysinate (K-Alkene sodium salt)

Compound 3 (1.1 g, 4 mmol) was dissolved in tetrahydrofuran (THF, 10 mL) and cooled in an ice-water bath. Sodium hydroxide solution (4 mL, 0.4 mmol, 1N) was added dropwise with stirring. The resulting mixture was stirred at room temperature overnight. The reaction mixture was then washed with diethyl ether (3 × 10 mL), frozen, and lyophilized to afford 930 mg K-Alkene sodium salt with a 83 % yield. ¹H NMR (400 MHz, H₂O/D₂O = 9:1, with water suppression): δ 5.78–5.85 (m, 1H), 3.95 (t, J = 6.4Hz 2H), 3.11–3.15 (m, 1H), 2.99–3.03 (m, 2H), 2.00–2.04 (m, 2H), 1.20–1.63 (m, 8H). 13C NMR (400 MHz, H₂O/D₂O = 9:1): δ 183.85, 168.39, 139.46, 114.81, 61.29, 56.24, 40.71, 34.69, 31.91, 30.87, 29.56, 22.46. MS (ESI) m/z: [M + Na]⁺ calculated for C₁₂H₂₁N₂Na₂O₄, 303.14, found 303.13.

4.3. Synthesis of Tet-2

Methylthiocarbohydrazide iodide (5)

Thiocarbohydrazide (5.3 g, 50 mmol) was added to absolute ethanol (75 mL) in a round bottom flask with stirring under Argon. Iodomethane (3.1 mL, 50 mmol) was slowly added to the reaction. The resulting mixture was heated to 60 °C and stirred overnight. White solid was precipitated out during the stirring. The reaction was cooled to room temperature, and hexane (100 mL) was slowly added to precipitate additional white solid. The mixture was filtered and the solid was washed with hexane (3 × 75 mL) and subsequently dried under vacuum to afford 5 as a white solid (11 g, 91%). The product was used in the next step without purification.

3-(Methylthio)-6-phenyl-1,2,4,5-tetrazine (6)

Compound 5 (4.95g, 20 mmol) and pyridine (3.55 mL, 40 mmol) was dissolved in dry DMF (20 mL) and stirred under Argon at 50 °C. Trimethyl orthobenzoate (8.4 mL, 40 mmol) was added dropwise over 1 h. The reaction mixture was stirred at 50 °C to give a mixture of 1,4-dihydrotetrazine and tetrazine products. After being cooled to room temperature, tributyl phosphine (5 mL, 20 mmol) and water (360 μL) were added to reduce tetrazine. The mixture was stirred for 30 min, and then diluted with 250 mL of DCM. The resulting solution was washed with saturated NaHCO₃ solution (1 × 50 mL), water (5 × 50 mL), and brine (1 × 50 mL). The organic layer was separated and dried over Na₂SO₄. The crude product was pre-absorbed onto silica, and solvent was removed under vacuum. The 1,4-dihydrotetrazine product was eluted by DCM:hexane (1:1) as a spongy solid (3.9g, 15 mmol).

The purified 1,4-dihydrotetrazine (3.9 g) was dissolved in DCM (100 mL) and cooled to 0 °C. (Diacetoxyiodo)-benzene (6.4 g, 20 mmol) was slowly added and the resulting solution was stirred at room temperature for 2 h. The solvent was then removed under vacuum and the crude product was purified by flash chromatography (DCM:Hexane = 1:4) to afford 1.9 g of 3-(methylthio)-6-phenyl-1,2,4,5-tetrazine (6) as a red crystal (47 % yield). ¹H NMR (400 MHz, CDCl₃): δ 8.54 (dd, J = 8,1.2 Hz, 2H), 7.60–7.62 (m, 3H), 2.82 (s, 3H). ¹³C NMR (400 MHz, CDCl₃): δ 175.32, 162.36, 132.31, 131.63, 129.25, 127.51, 13.43.

tert-butyl (4-(6-phenyl-1,2,4,5-tetrazin-3-yl)benzyl)carbamate (7)

Compound 6 (204 mg, 1 mmol), (4-(((tert-butoxycarbonyl)- amino)-methyl)-phenyl)-boronic acid (300 mg, 1.2 mmol), [1,1-Bis-(diphenylphosphino)- ferrocene]-dichloropalladium (II) (109mg, 0.15 mmol), and silver(I) oxide (577 mg, 2.5 mmol) were added to a round bottom flask with a stir bar under Argon. N,NDimethylformamide (10 mL) was then added into the flask and the mixture was stirred at 60 °C for 20 h. After being cooled to room temperature, solvent was removed under vacuum. The residue was redissolved in DCM and washed with brine, dried with Na₂SO₄. The crude product was pre-absorbed onto silica and solvent was removed under vacuum. The product was eluted by ethyl acetate:hexane (1:5) to afford 300 mg (83 %) tert-butyl (4-(6-phenyl-1,2,4,5-tetrazin-3-yl)-benzyl)-carbamate 7 as a purple solid. ¹H NMR (400 MHz, CDCl₃): δ 8.64–8.68 (m, 4H), 7.64–7.66 (m, 3H), 7.55–7.57 (m, 2H), 5.02 (s, 1H), 4.48 (d, J = 4.8 Hz, 2H),1.51 (s, 9H). MS(ESI) m/z: [M + H]⁺ calculated for C₂₀H₂₁N₅O₂, 363.17, found 364.18.

4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)-N-(4-(6-phenyl-1,2,4,5-tetrazin-3-yl)benzyl)butanamide (Tet-2)

Compound 7 (20 mg, 0.055 mmol) was dissolved in anhydrous DCM (1 mL) and trifluoroacetic acid (11.4 mg, 0.1 mmol) was slowly added. The resulting solution was stirred at room temperature for 5 h. After the solvent being removed by rotary evaporator for an extended time, the residue was redissolved in anhydrous DCM and triethylamine (10.1 mg, 0.1 mmol) was slowly added. Next, 2,5-dioxopyrrolidin-1-yl 4-(2,5-dioxo-2,5-dihydro-1H-pyrrol-1-yl)butanoate (20 mg, 0.07 mmol) was added and the resulting solution was stirred at room temperature overnight. The crude product was flushed through a silica column to obtain the desired product as a purple solid (18 mg, 76.5%). ¹H NMR (400 MHz, DMSO): δ 8.49–8.54 (m, 5H), 7.70–7.72 (m, 2H), 7.56–7.58 (m, 2H), 7.01–7.02 (m, 2H), 4.04 (d, J = 4.0 Hz, 2H), 3.43–3.47 (m, 2H), 2.18–2.21 (m, 2H), 1.80–1.81 (m, 2H). ¹³C NMR (400 MHz, DMSO) δ 171.97, 171.57, 163.78, 163.75, 145.15, 134.96, 133.06, 132.40, 130.78, 129.96, 128.58, 128.07, 128.01, 42.37, 37.36, 33.01, 24.64. MS(ESI) m/z: [M + Na]⁺ calculated for C₂₃H₂₀N₆O₃Na, 451.16, found 451.15.

4.4. Stability yest of Tet-1 and Tet-2

Tet-1 (or Tet-2) was dissolved in deuterated DMSO and ¹H-NMR was taken immediately. Next, the NMR sample was incubated at 37 °C for 72 h and analyzed again (Figure S3).

4.5. Plasmid construction

pLei-sfGFP-N149TAG

The sfGFP-N149TAG gene was amplified by overlapping PCR using wild-type sfGFP as the template. The digested PCR product was inserted into a pLei vector [32] behind a T5 promoter to afford plasmid psfGFP-N149TAG. The sfGFP sequence and the amber mutation were confirmed by DNA sequencing.

pLei-GST-wt

The wild-type GST gene was PCR amplified from a commercial pGEX plasmid. The GST gene was subsequently digested with NdeI and SacI, and ligated into a pLei vector [32] that was treated with the same set of restriction enzymes, to afford pLei-GST-wt. Plasmid pLei-GST-wt was confirmed by DNA sequencing.

pLei-GST-D61TAG

The GST-D61TAG gene was amplified by overlapping PCR using pLei-GST-wt plasmid as the template. The GST-D61TAG gene was subsequently digested with NdeI and SacI, and ligated into a pLei vector [32] that was treated with the same set of restriction enzymes, to afford pLei-GST-D61TAG. Plasmid pLei-GST-D61TAG was confirmed by DNA sequencing.

pLei-GST-D61TAG-C84S

The GST-D61TAG-C84S gene was amplified by overlapping PCR using pLei-GST-D61TAG plasmid as the template. The GST-D61TAG gene was subsequently digested with NdeI and SacI, and ligated into a pLei vector [32] that was treated with the same set of restriction enzymes, to afford pLei-GST-D61TAGC84S. Plasmid pLei-GST-D61TAG-C84S was confirmed by DNA sequencing.

4.6. Screening of PylRS variants for the genetic incorporation of K-Alkene.

E. coli GeneHogs strain that contains pLei-sfGFP-N149TAG and a PylRS variant of interest was inoculated into 1 mL of LB media with kanamycin (Kan, 50 mg/L) and chloramphenicol (Cm, 34 mg/L). Cells were cultured at 37 °C with shaking overnight. The seed culture (80 μL) was used to inoculate two tubes of fresh LB media (0.8 mL) containing Kan (50 mg/L), Cm (34 mg/L), and IPTG (0.25 mM). One tube contained K-Alkene and one tube did not contain K-Alkene. Following cultivation at 37 °C with shaking for 24 hours, cells were collected by centrifugation, washed with PBS buffer, and resuspended in 0.8 mL PBS for fluorescence and OD₆₀₀ measurements using a Synergy H1 Hybrid plate reader. The fluorescence of sfGFP was monitored with λ_Ex = 480 nm and λ_Em = 510 nm. The cell density was estimated by measuring the sample absorbance at 600 nm. Fluorescence intensities were normalized to cell growth. A total of 19 PylRS mutants were screened.

4.7. Protein expression and purification

E. coli GeneHogs strain that expresses ACPKRS and a target of protein (sfGFP or GST variants) was cultured in 100 mL LB media containing Kan (50 mg/L) and Cm (34 mg/L) at 37 °C with shaking. The protein expression was induced at OD₆₀₀ of 0.6 by the additions of IPTG (0.25 mM) and K-Alkene (0.5 mM). Following an additional 16 h of cultivation, cells were collected by centrifugation at 5,000g and 4 °C for 15 min. Harvested cells were resuspended in lysis buffer containing potassium phosphate (20 mM, pH 7.4), NaCl (150 mM), and imidazole (10 mM). Cells were subsequently disrupted by sonication. Cellular debris was removed by centrifugation (21,000g, 30 min, 4 °C). Protein purification using Ni Sepharose 6 Fast Flow resin (GE Healthcare) followed manufacturer’s instructions. Protein concentrations were determined by Bradford assay (Bio-Rad). Purified protein was desalted.

4.8. Mass spectrometry analysis of sfGFP mutant

The band containing sfGFP-N149K-Alkene was cut out from SDS-PAGE gel stained by Coomassie blue. After in-gel digestion with trypsin, the protein sample was dried down and redissolved in 120 μL of aqueous solution with 2.5% acetonitrile and 0.1% formic acid. A 5 μL of the digest sample was run by a RSLCnano system using a 1 h gradient on a 0.075 mm × 250 mm C18 column feeding into a Q-Exactive HF mass spectrometer. MS/MS data was analyzed using Mascot, which was set up to search a database that was customized with the provided protein sequence. Mascot search has a fragment ion mass tolerance of 0.060 Da and a parent ion tolerance of 10.0 ppm. Scaffold was used to validate MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 99.0% probability by the Peptide Prophet algorithm [36] with Scaffold delta-mass correction. Protein probabilities were assigned by the Protein Prophet algorithm [37].

4.9. Protein crosslinking

All protein concentrations were adjusted to 1 mg/mL (37 μM). A protein stock solution (9 μL) of interest was mixed with 1 μL of a tetrazine reagent (Tet-1 or Tet-2). The resulting mixture was incubated at 37 °C with slow shaking for the indicated time. The reaction was quenched with protein loading buffer (2X). The sample was heated at 95 °C for 10 min and then loaded onto a lab-made SDS-PAGE gel for electrophoresis. Detection of fluorescence signal was conducted before Coomassie blue staining. Bio-Rad Molecular Imager ChemiDoc XRS+ was used for gel imaging.

Highlights.

• Proximity-enhanced terminal alkene-tetrazine reaction

• Proximity-dependent protein crosslinking in the presence of a genetically incorporated alkene-containing non-canonical amino acid and a tetrazine reagent

• Fluorogenic nature of this reaction can be especially useful for the study of biomolecular interactions