A Cu(II)–ATP complex efficiently catalyses enantioselective Diels–Alder reactions


Natural biomolecules have been used extensively as chiral scaffolds that bind/surround metal complexes to achieve stereoselectivity in catalytic reactions. ATP is ubiquitously found in nature as an energy-storing molecule and can complex diverse metal cations. However, in biotic reactions ATP-metal complexes are thought to function mostly as co-substrates undergoing phosphoanhydride bond cleavage reactions rather than participating in catalytic mechanisms. Here, we report that a specific Cu(II)-ATP complex (Cu2+·ATP) efficiently catalyses Diels-Alder reactions with high reactivity and enantioselectivity. We investigate the substrates and stereoselectivity of the reaction, characterise the catalyst by a range of physicochemical experiments and propose the reaction mechanism based on density functional theory (DFT) calculations. It is found that three key residues (N7, β-phosphate and γ-phosphate) in ATP are important for the efficient catalytic activity and stereocontrol via complexation of the Cu(II) ion. In addition to the potential technological uses, these findings could have general implications for the chemical selection of complex mixtures in prebiotic scenarios.


Artificial metalloenzymes (ArMs) are usually constructed by embedding metal cofactors into the chiral scaffolds of biological molecules that are used to expand the reaction types and unearth novel functions of the biomolecules. Over the past four decades, protein-based ArMs have been widely investigated to achieve a variety of valuable enantioselective transformations1,2,3,4,5,6. To precisely characterise the active centres and obtain insight into the reaction mechanisms of ArMs, simple scaffolds of peptides and amino acids have been employed to rationally design artificial metallo-peptides7,8,9,10,11,12,13 and metallo-amino acids14,15,16,17,18,19,20,21. In recent years, nucleic acids have aroused much interest among chemists for constructing diverse nucleic acid-based ArMs for enantioselective catalysis. Natural double-stranded DNA (dsDNA) was first employed as a chiral scaffold in a supramolecular assembly with an achiral copper(II) complex, realising the chirality transfer from dsDNA to the products22. Since then, many synthetic dsDNAs have been designed to covalently anchor metallic moieties to produce DNA-based ArMs23,24,25,26,27,28. These artificial designs enable the fine-tuning of the microenvironment and provide a deeper understanding of the origin of chiral induction. Through their tuneable structures, G-quadruplexes containing 21- to 69-mer nucleotides have been employed to construct G-quadruplex DNA metalloenzymes, which have been successfully applied to several enantioselective transformations and demonstrated to depend largely on the conformation of the non-canonical G-quadruplex structure29,30,31,32,33,34,35,36. In addition, a short single-stranded 11 nt DNA within a G-triplex structure was shown to bind to copper(II) ions and modestly promote an enantioselective Diels–Alder (D–A) reaction37.

In addition to DNA, a large number of ribozymes generated by in vitro selection exhibit catalytic activity with the assistance of metal ions38. Most importantly for the present work, a ribozyme was selected to catalyse a D–A reaction. The Diels–Alderase ribozyme shows a 20,000-fold rate enhancement and relies on the presence of divalent metal ions such as Mg2+ or Mn2+39. In a further work Jaeschke and co-workers developed a 49-mer Diels–Alderase ribozyme that could catalyse the D–A reaction of anthracene derivatives and maleimides with an enantiomeric excess (ee) of up to 95%40. Systematic experiments based on the anthracene derivative dienes and maleimide dienophiles suggest that the stereoselectivity of the reaction is mostly controlled by RNA–diene interactions and the hydrophobic side chain of the dienophile is responsible for RNA binding41. The crystal structure of the Diels–Alderase ribozyme shows that the Mg2+ ion is a structural cofactor that stabilises the wedge shaped pocket and the stereoselectivity is governed by the shape of the catalytic pocket42.

In addition to in vitro-selected ribozymes such as the Diels–Alderase, synthetic double-stranded RNAs (dsRNAs) were shown to interact with copper(II) complexes to form RNA-based ArMs, giving rise to high reactivity yet modest enantioselectivity in a Friedel–Crafts reaction43. Another RNA-based ArMs containing either dsRNA or hairpin RNA exhibited very low enantioselective induction in a D–A reaction compared with the corresponding DNA-based ArMs44. The current nucleic acid-based ArMs always contain several tens to hundreds of nucleotides to achieve chiral scaffolding. In most cases, the precise location of the catalytic metal species is unclear, and high resolution structures of these ArMs are lacking. Recently, our group reported a cyclic dinucleotide (c-di-AMP)-based artificial metalloribozyme that catalyses a Friedel–Crafts reaction with high enantioselectivity45. Furthermore, a phosphine-modified deoxyuridine coordinating a palladium species enables an enantioselective allylic amination46, suggesting that the chirality of the sugar could be transferred to the product. Therefore, the design of nucleic acid-based ArMs with only a few nucleotides as the scaffold appears to be a promising approach for obtaining minimal systems that might be better suited for gaining accurate structural information and providing deeper insights into the reaction mechanisms. ATP is a well-known energy-storing molecule that participates in many processes in living organisms. In most enzyme-catalysed reactions, ATP acts as a co-substrate undergoing phosphoanhydride bond cleavage reactions. However, ATP has been demonstrated to specifically bind metal ions with high affinity and could therefore function as a chiral scaffold participating in enantioselective catalysis mediated by the complexed metal ions.

In this work, it is found that ATP interacts with Cu2+ ions to form a Cu(II)-ATP complex (Cu2+·ATP) that efficiently catalyses enantioselective D–A reactions (Fig. 1). From the ATP analogues experiments and spectroscopic characterisations, the purine moiety, β- and γ-phosphates in ATP are revealed as vital residues to coordinate Cu2+ ion for exerting enantioselective catalytic activity. The theoretical calculations further support a fine coordination structure of Cu2+·ATP, in which the Cu2+ ion binds to N7 atom from adenine and two oxygen atoms from β- and γ-phosphates, together with a hydrogen bond between 6-amino and γ-phosphate oxygen. This work demonstrates that a single nucleoside polyphosphate is sufficient for chiral induction in enantioselective reactions and the catalytic function of ATP-metal complexes might implicate the chemical selection in primordial chemistry.

Fig. 1: Schematic representation of enantioselective Diels–Alder reactions catalysed by a Cu(II)-ATP complex.

The Cu(II)–ATP complex (Cu2+·ATP) is composed of copper(II) trifluoromethanesulfonate (Cu(OTf)2) and ATP. The coordination structure of Cu2+·ATP is supported by spectroscopic characterisations and theoretical calculations, in which Cu2+ ion binds to one nitrogen atom (N7) from adenine, one oxygen atom from β-phosphate, one oxygen atom from γ-phosphate and one trifluoromethanesulfonate anion (OTf-) and simultaneously an intramolecular hydrogen bond is formed between a hydrogen atom from 6-amino and an oxygen atom from γ-phosphate. Typical Diels–Alder reactions of azachalcones (1a–h) and cyclopentadiene (2) are selected in this study, and the detailed reaction procedure is described in “Methods” section.


Enantioselective Diels–Alder reactions catalysed by Cu2+·ATP

A benchmark D–A reaction of azachalcone (1a) and cyclopentadiene (2) was employed to test the catalytic performance of ATP–metal ion complexes. The initial attempt using either ATP or a Mg2+·ATP complex resulted in a very low conversion and ee value (Table 1, entries 1 and 2). To our delight, when copper(II) nitrate was added to ATP as a metal cofactor, the corresponding D–A reaction gave a conversion of 90%, an ee of 3a (exo) of 74% and an ee of 3a (endo) of 65% (Table 1, entry 4). These results indicate that ATP and Cu2+ ion specifically interact to form a Cu(II)–ATP complex of Cu2+·ATP. Using other divalent metal ions such as Zn2+, Co2+ and Ni2+ as the metal cofactors resulted in low conversions and ee values (Table 1, entries 5-7). The introduction of extra achiral ligands significantly inhibited the reaction catalysed by Cu2+·ATP and 3a was obtained with ee values of 29–46% (Table 1, entries 8–10). The distinct circular dichroism (CD) spectra of ATP in the presence of different copper(II) complexes (Supplementary Fig. S1) suggest that the presence of additional ligands might block the interaction between 1a and the Cu2+ ion, leading to poor reactivity and low enantioselectivity. Different copper(II) salts and molar ratios of ATP/Cu2+ were screened (Supplementary Tables S1 and S2) and an ATP/Cu(OTf)2 ratio of 5:1 resulted in optimal catalysis with 98% conversion and 72% ee in favour of the endo isomer (Table 1, entry 11). The addition of 10 mM Mg2+ ions resulted in a significantly reduced ee value compared with that obtained from the Cu2+·ATP catalysed D–A reaction (Table 1, entry 12), whereas Na+ and K+ ions caused no significant difference (Supplementary Table 3). Because Mg2+ ions are known to bind to ATP47,48,49,50, this result indicates that the presence of Mg2+ might compete with the binding of Cu2+ to ATP. Concerning the reaction temperature, the ee value decreased at an elevated temperature (Table 1, entry 11 vs. entry 13).

Table 1 Enantioselective Diels–Alder reactions catalysed by metallo–ATP complexesa.

To investigate the substrate specificity of the Cu2+·ATP catalyst, different azachalcones (1b–h) were investigated and modest to good stereoselectivities were obtained (Table 1, entries 14–20). Compared with 1a, azachalcone 1b or 1c, which bears an electron-donating group (4-Me or 4-OMe) on the phenyl moiety, exhibited an enhanced ee in the corresponding reaction (Table 1, entry 11 vs. entries 14 and 15). Changing the methoxy  substitution from the 4′ -position to the 2′ -position on the phenyl moiety of azachalcone (1d) caused a significant decrease in the reactivity and enantioselectivity (Table 1, entry 15 vs. entry 16). For the corresponding D–A reactions using azachalcones with electron-withdrawing moieties (1e and 1f), 1e bearing R = (4-Cl)C6H4 reacted with 2 to give a 93% ee in favour of the exo isomer of 3e (Table 1, entry 17), whereas 1f with R = (4-NO2)C6H4 yielded an 84% ee in favour of the endo isomer of 3f (Table 1, entry 18). In addition, good reactivity and modest enantioselectivity were achieved using azachalcones 1g and 1h with heterocyclic substitutions (Table 1, entries 19 and 20). These results suggest that the steric and electronic effects of the substituents of azachalcones greatly affect the catalytic performance of Cu2+·ATP. In addition, the reaction of substrate 1a (105 mg, 0.5 mmol) was conducted at a large scale to test its potential practicability. At a Cu2+·ATP loading of 5 mol%, the product 3a was obtained with an isolated yield of 80% and an ee value of 65% in favour of the endo isomer. Overall, it was demonstrated that ATP and the Cu2+ ion form a potent and practical entity of Cu2+·ATP for enantioselective D–A reactions.

Kinetics of Cu2+·ATP catalysis

To clarify the catalytic roles of ATP and Cu2+, the apparent second-order rate constants (kapp) of ATP, Cu2+ and Cu2+·ATP were determined by monitoring the UV-Vis absorption of 1a during the corresponding D–A reactions. Compared with the D–A reaction without a catalyst (kapp,uncat), that with ATP had a comparable kapp,ATP (Table 2, entry 1 vs. entry 2), suggesting that ATP is not the catalytic species, as indicated in Table 1. Cu2+ ions are efficient Lewis acid catalysts for the D–A reaction and led to an approximately sevenfold rate acceleration (Table 2, entry 3). When ATP and Cu2+ ions were present as the Cu2+·ATP complex, a 13-fold rate enhancement relative to that of the uncatalysed reaction was observed (Table 2, entry 4). The kinetic parameters suggest that ATP and Cu2+ ions indeed interact to assemble into a Cu(II)–ATP complex, where ATP serves as the chiral scaffold and the Cu2+ ions serve as the catalytically active species.

Table 2 Kinetic study of Cu2+·ATP.

Catalytic performance of ATP analogues

To probe the binding sites of Cu2+ ion to ATP essential for catalysis, ATP was replaced with different ATP analogues in the Cu2+·ATP catalysed benchmark D–A reaction. Compared with ATP with three phosphates as the scaffold, ADP and Cu2+ ions catalysed the D–A reaction with a similar conversion but a sharply decreased ee value of 19% (Fig. 2). The further removal of phosphate to either AMP or adenosine resulted in a reduced conversion and racemic product 3a (Fig. 2). To investigate whether ATP is decomposed to ADP or AMP in the Cu2+·ATP catalysed D–A reaction, the reaction medium was analysed by high-performance liquid chromatography (HPLC) and ATP remained nearly unchanged during the reaction at pH 5.5 (Supplementary Fig. S6). These results suggest that the triphosphates are indispensable for Cu2+·ATP and the β- and γ-phosphates of ATP are probably the binding sites for the Cu2+ ions as hypothesised previously51,52,53,54. Furthermore, the ribose of ATP was changed to deoxyribose by using dATP. A significant reduction in both the reactivity and enantioselectivity was observed with Cu2+·dATP (Fig. 2). This result indicates that 2′-OH affects the catalytic performance of Cu2+·ATP and that the altered sugar conformation is less able to facilitate the reaction (Supplementary Fig. S7). In addition, several nucleobase analogues of ATP were tested. Compared with Cu2+·ATP, Cu2+·GTP provided 3a with a comparable 63% ee, probably owing to the presence of all the residues in ATP critical for efficient catalysis as identified above (Fig. 2). However, Cu2+·UTP and Cu2+·CTP generated 3a with significantly decreased ee values (Fig. 2), which might be attributed to changes in the chiral complex structure compared with that of Cu2+·ATP (Supplementary Fig. S8). These results indicate that the purine moiety in ATP analogues is also important for achieving enantioselective catalysis with Cu2+ complexes.

Fig. 2: The effect of ATP analogues for the Cu2+·ATP-catalysed Diels–Alder reaction.

Reaction conditions: 1a (1 mM), 2 (200 mM), ATP analogue (250 μM), Cu(OTf)2 (50 μM), MES buffer (20 mM, pH 5.5), 4 °C, 24 h. All data are the averages of at least two individual experiments (reproducibility: ±5% conversion, ±3% ee).

Physicochemical characterisations of the Cu(II)–ATP complex

To investigate the interaction between ATP and Cu2+ ions, several characterisation techniques were employed. The addition of Cu2+ caused slight changes to the CD spectrum of ATP (Supplementary Fig. S9), indicating that Cu2+ ions hardly change the conformation of ATP. To assess the binding affinity of the Cu(II)–ATP complex, the apparent binding constant (kb) was determined by a UV titration experiment. The kb value for ATP and Cu2+ ions was estimated to be (3.2 ± 0.1) × 105 M−1 based on the curve fitting (Fig. 3a), demonstrating that Cu2+ ions have a high affinity for ATP. Because Cu2+ ions are paramagnetic species, electron paramagnetic resonance (EPR) was employed to verify the interaction between ATP and Cu2+ ions. The EPR spectrum of only Cu2+ showed the typical fine structures of paramagnetic Cu2+ ions (Fig. 3b). When ATP was added to Cu2+ ions to form Cu2+·ATP, the fine structures of the Cu2+ ions changed significantly and shifted (Fig. 3b), indicating that the Cu2+ ions and ATP indeed interact. Furthermore, nuclear magnetic resonance (NMR) was used to detect possible binding sites of Cu2+ ions in ATP. Owing to the paramagnetic property of Cu2+ ions, the addition of Cu2+ ions to ATP causes NMR signal broadening and shifting. When the ATP/Cu2+ ratio was 300:1, the signal of H8 first broadened and shifted downfield relative to that in the 1H NMR of ATP, and further increasing the amount of Cu2+ ions caused the H8 signal to broaden to the baseline and shift downfield (Fig. 3c). The H2 signals exhibited a similar but delayed tendency upon the addition of Cu2+ ions to ATP (Fig. 3c). These results are in agreement with the reported interaction between ATP and Yb3+ ions55, indicating that Cu2+ ions possibly bind to N7 in ATP. The chemical shifts of H1′–H5′ on the ribose ring remained nearly unchanged, even at an ATP/Cu2+ ratio of 50:1 (Fig. 3c), indicating that these protons might be farther away from the paramagnetic centre of Cu2+ ion. Compared with 1H NMR titration, the 31P NMR titration of ATP resulted in significant changes upon the addition of Cu2+ ions. When 0.1% Cu2+ ions were added to ATP, the Pβ and Pγ signals immediately broadened and shifted downfield, and further increasing the ATP/Cu2+ ratio to 100:1 led to the disappearance of the Pβ and Pγ signals and the broadening and downfield shifting of the Pα signal (Fig. 3d). These results indicate that Pβ and Pγ are sensitive to the presence of Cu2+ ions, which probably interact with Pβ-O and Pγ-O in accordance with the calculated structure of Mg2+·ATP, see below. In short, the characterisation data together with the results for the above ATP analogues demonstrate that ATP and Cu2+ ions interact via the β- and γ-phosphates as well as the purine N7. These binding sites for Cu2+ ions have also been suggested in the literatures54,56.

Fig. 3: The interaction between ATP and Cu2+ ions.

a The apparent binding constant (kb) of ATP and Cu2+ ions derived from the plot of [ATP]/Δεa,259 nm vs. [ATP] at various concentrations (5–100 μM) for Cu(OTf)2 (5 μM). b EPR spectra of ATP, Cu2+ and Cu2+·ATP. Experimental conditions: ATP (50 mM), Cu(OTf)2 (10 mM), glycerol (20 v/v%), MES (20 mM, pH 5.5), 100 K. c 1H NMR and d 31P NMR spectroscopic titrations of ATP (120 mM) by varying the concentration of CuCl2 (0.12–24 mM) in D2O.

Cu2+·ATP complex calculation and proposed reaction mechanism

To further substantiate the hypothesised fine structure of the catalytically active Cu2+·ATP complex and explore a potential reaction mechanism, density functional theory (DFT) calculations were performed. Based on several proposed Cu2+·ATP models in the literature56, a stable structure of Cu2+·ATP was obtained by a gas-phase calculation in which the Cu2+ ion is bound to the N7 atom of adenine and the β- and γ-phosphate oxygen atoms and accompanying a  hydrogen bond Pγ-O···H-N6 (Fig. 4a). The relative electronic energy (ΔE) of the optimised Cu2+·ATP structure was 1.1 kcal mol−1 lower than that of a previously described model obtained by a molecular orbital method57 and 22.7 kcal mol−1 lower than that of the Cu2+·ATP model without a hydrogen bond (Supplementary Fig. S22). With this Cu2+·ATP model in hand, the relative electronic energies of the precursor complexes of 1a-Cu2+·ATP and 2 were calculated. Using the major product 3a (endo) as an example, the ΔE value of the precursor of 1a-Cu2+·ATP and 2 that yielded 3a (endo) via the attack of the Si face was 8.1 kcal mol−1 lower than that of the precursor for the Re face attack (Fig. 4b). The Si face attack was favoured owing to the hydrogen bonding between cyclopentadiene 2 and the phosphate oxygen atoms (Fig. 4b). This result indicates that the Cu2+·ATP-catalysed D–A reaction gives 3a (Si-endo) with the preferred configuration of 1R, 2S, 3S, 4S (Supplementary Fig. S24), which is accordance with the experimental results (Supplementary Figs. S20, S21). The ΔE values of the precursors and products of 3a (exo) were also calculated. The precursor of 3a (Si-exo) was more stable than that of 3a (Re-exo), further suggesting that cyclopentadiene 2 favoured the attack of 1a-Cu2+·ATP from the Si face (Supplementary Fig. S23). However, the ΔE value of 3a (Re-exo) was 1.3 kcal mol−1 lower than that of 3a (Si-exo), in accordance with the experimental results (Supplementary Figs. S21,S24). Based on the experimental and theoretical results, a plausible reaction mechanism was proposed (Fig. 4c). The addition of 1a to the stable Cu2+·ATP catalyst gave rise to the intermediate 1a-Cu2+·ATP, which was in a pentacoordination state with newly formed Cu···N(1a) and Cu···O(1a) interactions. Because of the hydrogen bonding between ATP and 2, 2 preferentially attacked the Si face of 1a in the intermediate of 1a-Cu2+·ATP, leading to a relatively stable transition state consisting of Cu2+·ATP, 1a and 2. The transition state of 1a-Cu2+·ATP-2 automatically converted to the intermediate of 3a-Cu2+·ATP, which was accompanied by the breaking of the Cu···O(1a) bond, and the major D–A product 3a (Si-endo) was obtained after release from Cu2+·ATP.

Fig. 4: DFT calculations and a plausible reaction mechanism.

a Coordination structures of Cu2+·ATP. For clarity, one trifluoromethanesulfonate anion coordinating to the copper(II) ion is omitted in Cu2+·ATP. b Relative electronic energies (ΔE) of the precursors of 3a (endo) with 2 attacking the intermediate 1a-Cu2+·ATP from the Si and Re faces. c Plausible mechanism of the Cu2+·ATP catalysed Diels–Alder reaction of 1a with 2 yielding 3a (endo) showing the favoured Si face attack of the intermediate of 1a and Cu2+·ATP.


In summary, we discovered that an enantioselective catalyst composed of the single nucleoside triphosphate ATP in complex with Cu2+ ions is able to catalyse a D–A reaction with significant rate acceleration and high enantioselectivity. The purine structure and the phosphates at the β- and γ- positions are vital factors contributing to the enantioselective activity of the Cu2+·ATP catalyst. Based on control experiments, physicochemical characterisations, and DFT calculations, a fine coordination structure of Cu2+·ATP in which the Cu2+ ion binds to the N7 atom, β-phosphate oxygen atom, and γ-phosphate oxygen atom and an intramolecular hydrogen bond between the 6-amino and γ-phosphate oxygen moieties was proposed. Of the reported metallo-biohybrid catalysts for enantioselective D–A reactions (Supplementary Table S6), the Cu2+·ATP catalyst reported here is competitive, especially taking into account the much simpler chiral scaffold compared to DNA, RNA and proteins. Importantly, compared to ADP and AMP, ATP proved to be a superior ligand for Cu2+ binding and formation of a potent Cu(II)–ATP complex. The proposed fine coordination structure of the complex is able to explain the origin of chiral induction and should facilitate the rational design of further simple but efficient nucleotide-based catalysts.

In addition to its potential use in synthesis approaches, this work suggests that nucleotides could have played a role in the chemical selection of complex mixtures in prebiotic reactions in early evolutionary scenarios. The observation that several of the currently ubiquitous cofactors involved in redox and C–C bond formation reactions (e.g. nicotinamide-, flavine-, pantotheine- and cobalamine-based cofactors) contain adenine nucleosides and nucleotides could indicate a more pronounced role of ATP as ligand in prebiotic reactions. Although the D–A reaction described in this work is not proposed to be highly important in early chemical evolution, aldol reactions are thought to have played a crucial role in the establishment of early metabolic pathways58, even providing early access to nucleotides from very simple starting materials59. However, how mirror symmetry breaking in aldol and other reactions occurred is still an open question60,61. Since aldol reactions are also catalysed by divalent metal ions62,63,64,65, it is possible that nucleotide polyphosphates have played a role in the chemical selection of important metabolites in the abiotic stages of the emergence of life. Experiments investigating the potential of nucleotides to initiate chiral induction in aldol reactions are currently underway in our lab.


Typical procedure for Cu2+·ATP catalysed D–A reactions

A stock solution of ATP in water (final conc. 250 μM) and a freshly prepared aqueous solution of Cu(OTf)2 (final conc. 50 μM) were added to an MES buffer solution (20 mM, pH 5.5) in a 10 mL vial to a total volume of 1000 μL. After stirring for 30 min at 4 °C, a thoroughly mixed solution of azachalcone 1a (10 μL of a 0.1 M stock solution in CH3CN, 1 μmol) and freshly distilled cyclopentadiene 2 (16 μL, 200 μmol) were immediately added. After the mixture was stirred for 24 h at 4 °C, the aqueous media were extracted by diethyl ether (3 × 2 mL) and flushed through a short gel column (a 5 cm length of glass dropper was filled with the silica gel to a height of ca. 2 cm with some cotton at the bottom). The combined organic layers were removed under reduced pressure and the residue was directly analysed by chiral HPLC using a Daicel Chiralpak ODH column (250 × 4.6 mm) with hexane and isopropanol as the eluents. The conversion of 1a was calculated using the following Eq. (1):

$${\mathrm{Conversion}}\;{\mathrm{of}}\;{\mathbf{1a}}\left( \% \right) = A_{{\mathbf{3a}}}/(A_{{\mathbf{3a}}} + A_{{\mathbf{1a}}}/f),$$

where A1a and A3a are the HPLC areas of 1a and 3a, respectively. The relative correction factor f was 0.595.

Kinetic assays for Cu2+·ATP

All kinetic measurements were performed by monitoring the disappearance of the absorption of 1a at 326 nm, followed by the reference66, using UV–Vis spectroscopy at 4 °C. ATP (final conc. 250 μM) in an MES buffer (20 mM, pH 5.5) was added to a 2 mL quartz cuvette containing a small magnet and stirred for 10 min, and then an aqueous solution of Cu(OTf)2 (final conc. 50 μM) was added. After stirring for another 20 min, a fresh solution of 1a (4, 6 or 10 μL of 0.1 M stock solution in CH3CN) was added. Followed by an immediate addition of 2 (final conc. 5 mM), the measurement was started and the cuvette was sealed tightly. The initial rate (Vinit) was determined from the slope of the line fitted to the decrease in the absorption of 1a versus time, and the following Eq. (2) was used to calculate Vinit:

$$V_{{\mathrm{init}}} = {\mathrm{d}}\left[ {A_{{\mathbf{1a}}}} \right]/{\mathrm{d}}t\cdot \left( {d\cdot \left( {\varepsilon _{{\mathbf{1a}}} - \varepsilon _{{\mathbf{3a}}}} \right)} \right)^{ - 1},$$

where d[A1a]/dt is the slope of the absorption of 1a vs. time during the initial 15% of the reaction, and d is the path length of the cuvette. The parameters ε1a and ε3a are the molar extinction coefficients of 1a and 3a, respectively (Supplementary Figs. S13,S14). All parameters were measured at least three times.

The apparent second-order rate constant (kapp) was determined according to the procedure described in the literature66. The following Eq. (3) was used to calculate kapp:

$$k_{{\mathrm{app}}} = {\mathrm{d}}\left[ {A_{{\mathbf{1a}}}} \right]/{\mathrm{d}}t \cdot \left( {d \cdot \left( {\varepsilon _{{\mathbf{1a}}} - \varepsilon _{{\mathbf{3a}}}} \right) \cdot \left[ {{\mathbf{1a}}} \right]_0 \cdot \left[ {\mathbf{2}} \right]_0} \right)^{ - 1} = V_{{\mathrm{init}}}/\left( {\left[ {{\mathbf{1a}}} \right]_0 \cdot \left[ {\mathbf{2}} \right]_0} \right).$$

where [1a]0 and [2]0 are the initial concentrations of 1a and 2, respectively.

DFT calculations

All calculations of the reactions were performed in the gas-phase with Gaussian 1667. The molecular geometries of the precursors, transition states, and products were optimised at the B3LYP-D3/LANL2DZ ~3–21G level; the 3–21G basis set was used for C, H, O, N, and P, whereas LANL2DZ was used for Cu. Then, single-point energy corrections were obtained using M06-2X-D3/LANL2DZ ~6–311 G(d,p); the 6–311G(d,p) basis set was used for C, H, O, N and P, whereas LANL2DZ was used for Cu. The kinetic barriers of the non-catalytic reaction and the catalytic reaction with the most stable precursor were evaluated by calculating the single-point energies at the M06-2X-D3/LANL2DZ ~6–311G(d,p) level. All the structures were verified to be local minima by frequency calculations, whereas all the transition state species had only one imaginary frequency.

Data availability

The data supporting the findings of this work are available within the article and its Supplementary Information files. All other relevant data of this study are available from the corresponding author upon reasonable request. Source data are provided with this paper.


  1. 1.

    Reetz, M. T. Directed evolution of artificial metalloenzymes: a universal means to tune the selectivity of transition metal catalysts? Acc. Chem. Res. 52, 336–344 (2019).

    CAS  PubMed  Article  Google Scholar 

  2. 2.

    Mirts, E. N., Bhagi-Damodaran, A. & Lu, Y. Understanding and modulating metalloenzymes with unnatural amino acids, non-native metal ions, and non-native metallocofactors. Acc. Chem. Res. 52, 935–944 (2019).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  3. 3.

    Liang, A. D., Serrano-Plana, J., Peterson, R. L. & Ward, T. R. Artificial metalloenzymes based on the biotin-streptavidin technology: enzymatic cascades and directed evolution. Acc. Chem. Res. 52, 585–595 (2019).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  4. 4.

    Lewis, J. C. Beyond the second coordination sphere: engineering dirhodium artificial metalloenzymes to enable protein control of transition metal catalysis. Acc. Chem. Res. 52, 576–584 (2019).

    CAS  PubMed  Article  Google Scholar 

  5. 5.

    Zeymer, C. & Hilvert, D. Directed evolution of protein catalysts. Annu. Rev. Biochem. 87, 131–157 (2018).

    CAS  PubMed  Article  Google Scholar 

  6. 6.

    Schwizer, F. et al. Artificial metalloenzymes: reaction scope and optimization strategies. Chem. Rev. 118, 142–231 (2018).

    CAS  PubMed  Article  Google Scholar 

  7. 7.

    Dolan, M. A. et al. Catalytic nanoassemblies formed by short peptides promote highly enantioselective transfer hydrogenation. ACS Nano 13, 9292–9297 (2019).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  8. 8.

    Vicens, L. & Costas, M. Biologically inspired oxidation catalysis using metallopeptides. Dalton Trans. 47, 1755–1763 (2018).

    CAS  PubMed  Article  Google Scholar 

  9. 9.

    Zheng, L., Marcozzi, A., Gerasimov, J. Y. & Herrmann, A. Conformationally constrained cyclic peptides: powerful scaffolds for asymmetric catalysis. Angew. Chem. Int. Ed. 53, 7599–7603 (2014).

    CAS  Article  Google Scholar 

  10. 10.

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

    CAS  Article  Google Scholar 

  11. 11.

    Ball, Z. T. Designing enzyme-like catalysts: a rhodium(II) metallopeptide case study. Acc. Chem. Res. 46, 560–570 (2013).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  12. 12.

    Sambasivan, R. & Ball, Z. T. Metallopeptides for asymmetric dirhodium catalysis. J. Am. Chem. Soc. 132, 9289–9291 (2010).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  13. 13.

    Coquiere, D., Bos, J., Beld, J. & Roelfes, G. Enantioselective artificial metalloenzymes based on a bovine pancreatic polypeptide scaffold. Angew. Chem. Int. Ed. 48, 5159–5162 (2009).

    CAS  Article  Google Scholar 

  14. 14.

    Cozzi, P. G., Gualandi, A., Potenti, S., Calogero, F. & Rodeghiero, G. Asymmetric reactions enabled by cooperative enantioselective amino- and lewis acid catalysis. Top. Curr. Chem. 378, 1 (2020).

    CAS  Article  Google Scholar 

  15. 15.

    Afewerki, S. & Cordova, A. Combinations of aminocatalysts and metal catalysts: a powerful cooperative approach in selective organic synthesis. Chem. Rev. 116, 13512–13570 (2016).

    CAS  PubMed  Article  Google Scholar 

  16. 16.

    Karmakar, A., Maji, T., Wittmann, S. & Reiser, O. L-proline/CoCl2-catalyzed highly diastereo- and enantioselective direct aldol reactions. Chem. Eur. J. 17, 11024–11029 (2011).

    CAS  PubMed  Article  Google Scholar 

  17. 17.

    Paradowska, J., Stodulski, M. & Mlynarski, J. Catalysts based on amino acids for asymmetric reactions in water. Angew. Chem. Int. Ed. 48, 4288–4297 (2009).

    CAS  Article  Google Scholar 

  18. 18.

    Aplander, K., Ding, R., Lindstroem, U. M., Wennerberg, J. & Schultz, S. alpha-amino acid induced rate acceleration in aqueous biphasic Lewis acid catalyzed Michael addition reactions. Angew. Chem. Int. Ed. 46, 4543–4546 (2007).

    CAS  Article  Google Scholar 

  19. 19.

    Darbre, T. & Machuqueiro, M. Zn-proline catalyzed direct aldol reaction in aqueous media. Chem. Commun. 1090–1091 (2003).

  20. 20.

    Gyarmati, J. et al. Asymmetric induction by amino acid ligands in chromium(II)-assisted reduction of ketones. J. Organomet. Chem. 586, 106–109 (1999).

    CAS  Article  Google Scholar 

  21. 21.

    Otto, S., Boccaletti, G. & Engberts, J. A chiral Lewis-acid-catalyzed Diels–Alder reaction. Water-enhanced enantioselectivity. J. Am. Chem. Soc. 120, 4238–4239 (1998).

    CAS  Article  Google Scholar 

  22. 22.

    Roelfes, G. & Feringa, B. L. DNA-based asymmetric catalysis. Angew. Chem. Int. Ed. 44, 3230–3232 (2005).

    CAS  Article  Google Scholar 

  23. 23.

    Yum, J. H. et al. Modular DNA-based hybrid catalysts as a toolbox for enantioselective hydration of alpha,beta-unsaturated ketones. Org. Biomol. Chem. 17, 2548–2553 (2019).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  24. 24.

    Mansot, J. et al. A rational quest for selectivity through precise ligand-positioning in tandem DNA-catalysed Friedel-Crafts alkylation/asymmetric protonation. Chem. Sci. 10, 2875–2881 (2019).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  25. 25.

    Dey, S. & Jaeschke, A. Tuning the stereoselectivity of a DNA-catalyzed michael addition through covalent modification. Angew. Chem. Int. Ed. 54, 11279–11282 (2015).

    CAS  Article  Google Scholar 

  26. 26.

    Park, S. et al. Development of DNA-based hybrid catalysts through direct ligand incorporation: toward understanding of DNA-based asymmetric catalysis. ACS Catal. 4, 4070–4073 (2014).

    CAS  Article  Google Scholar 

  27. 27.

    Fournier, P., Fiammengo, R. & Jaeschke, A. Allylic amination by a DNA-diene-iridium(I) hybrid catalyst. Angew. Chem. Int. Ed. 48, 4426–4429 (2009).

    CAS  Article  Google Scholar 

  28. 28.

    Oltra, N. S. & Roelfes, G. Modular assembly of novel DNA-based catalysts. Chem. Commun. 6039–6041 (2008).

  29. 29.

    Dey, S., Ruehl, C. L. & Jaeschke, A. Catalysis of Michael additions by covalently modified G-quadruplex DNA. Chem. Eur. J. 23, 12162–12170 (2017).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  30. 30.

    Cheng, M. et al. Enantioselective sulfoxidation reaction catalyzed by a G-quadruplex DNA metalloenzyme. Chem. Commun. 52, 9644–9647 (2016).

    CAS  Article  Google Scholar 

  31. 31.

    Li, Y. et al. Terpyridine-Cu(II) targeting human telomeric DNA to produce highly stereospecific G-quadruplex DNA metalloenzyme. Chem. Sci. 6, 5578–5585 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  32. 32.

    Wang, C., Jia, G., Li, Y., Zhang, S. & Li, C. Na+/K+ switch of enantioselectivity in G-quadruplex DNA-based catalysis. Chem. Commun. 49, 11161–11163 (2013).

    CAS  Article  Google Scholar 

  33. 33.

    Wilking, M. & Hennecke, U. The influence of G-quadruplex structure on DNA-based asymmetric catalysis using the G-quadruplex-bound cationic porphyrin TMPyP4 center dot Cu. Org. Biomol. Chem. 11, 6940–6945 (2013).

    CAS  PubMed  Article  Google Scholar 

  34. 34.

    Wang, C. et al. Enantioselective Friedel-Crafts reactions in water catalyzed by a human telomeric G-quadruplex DNA metalloenzyme. Chem. Commun. 48, 6232–6234 (2012).

    CAS  Article  Google Scholar 

  35. 35.

    Wang, C. et al. Enantioselective Diels–Alder reactions with G-quadruplex DNA-based catalysts. Angew. Chem. Int. Ed. 51, 9352–9355 (2012).

    CAS  Article  Google Scholar 

  36. 36.

    Roe, S., Ritson, D. J., Garner, T., Searle, M. & Moses, J. E. Tuneable DNA-based asymmetric catalysis using a G-quadruplex supramolecular assembly. Chem. Commun. 46, 4309–4311 (2010).

    CAS  Article  Google Scholar 

  37. 37.

    Xu, X. et al. Enantioselective Diels–Alder reactions using a G-triplex DNA-based catalyst. Catal. Commun. 74, 16–18 (2016).

    CAS  Article  Google Scholar 

  38. 38.

    Pyle, A. M. Ribozymes: a distinct class of metalloenzymes. Science 261, 709–714 (1993).

    ADS  CAS  PubMed  Article  Google Scholar 

  39. 39.

    Seelig, B. & Jaschke, A. A small catalytic RNA motif with Diels–Alderase activity. Chem. Biol. 6, 167–176 (1999).

    CAS  PubMed  Article  Google Scholar 

  40. 40.

    Seelig, B., Keiper, S., Stuhlmann, F. & Jaeschke, A. Enantioselective ribozyme catalysis of a bimolecular cycloaddition reaction. Angew. Chem. Int. Ed. 39, 4576–4579 (2000).

    CAS  Article  Google Scholar 

  41. 41.

    Stuhlmann, F. & Jaschke, A. Characterization of an RNA active site: Interactions between a Diels–Alderase ribozyme and its substrates and products. J. Am. Chem. Soc. 124, 3238–3244 (2002).

    CAS  PubMed  Article  Google Scholar 

  42. 42.

    Serganov, A. et al. Structural basis for Diels–Alder ribozyme-catalyzed carbon-carbon bond formation. Nat. Struct. Mol. Biol. 12, 218–224 (2005).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  43. 43.

    Duchemin, N. et al. Expanding biohybrid-mediated asymmetric catalysis into the realm of RNA. Chem. Commun. 52, 8604–8607 (2016).

    CAS  Article  Google Scholar 

  44. 44.

    Marek, J. J. & Hennecke, U. Why DNA is a more effective scaffold than RNA in nucleic acid-based asymmetric catalysis—supramolecular control of cooperative effects. Chem. Eur. J. 23, 6009–6013 (2017).

    CAS  PubMed  Article  Google Scholar 

  45. 45.

    Wang, C. et al. Highly efficient cyclic dinucleotide based artificial metalloribozymes for enantioselective Friedel-Crafts reactions in water. Angew. Chem. Int. Ed. 59, 3444–3449 (2020).

    CAS  Article  Google Scholar 

  46. 46.

    Ropartz, L., et al. Phosphine containing oligonucleotides for the development of metallodeoxyribozymes. Chem. Commun. 1556–1558 (2007).

  47. 47.

    Wilson, J. E. & Chin, A. Chelation of divalent cations by ATP, studied by titration calorimetry. Anal. Biochem. 193, 16–19 (1991).

    CAS  PubMed  Article  Google Scholar 

  48. 48.

    Abraha, A., de Freitas, D. E., Margarida, M., Castro, C. A. & Geraldes, C. F. Competition between Li+ and Mg2+ for ATP and ADP in aqueous solution: a multinuclear NMR study. J. Inorg. Biochem. 42, 191–198 (1991).

    CAS  PubMed  Article  Google Scholar 

  49. 49.

    Heyde, M. E. & Rimai, L. A Raman spectroscopic study of the interaction of Ca2+ and Mg2+ with the triphosphate moiety of adenosine triphosphate in aqueous solution. Biochemistry 10, 1121–1128 (1971).

    CAS  PubMed  Article  Google Scholar 

  50. 50.

    Burton, K. Formation constants for the complexes of adenosine di- or tri-phosphate with magnesium or calcium ions. Biochem. J. 71, 388–395 (1959).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  51. 51.

    Hoffmann, S. K. et al. Copper(II) ions interactions in the systems with triamines and ATP. Potentiometric and spectroscopic studies. J. Inorg. Biochem. 177, 89–100 (2017).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  52. 52.

    Tallineau, C. et al. Evidence for the involvement of (Cu-ATP)2- in the inhibition of human erythrocyte (Ca2+ + Mg2+)-ATPase by copper. Biochim. Biophys. Acta 775, 51–56 (1984).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  53. 53.

    Schneider, P. W. & Brintzinger, H. Zum Mechanismus der Metallionen–Katalysierten Hydrolyse von Adenosintriphosphat (ATP). I. Helv. Chim. Acta 47, 1717–1733 (1964).

    CAS  Article  Google Scholar 

  54. 54.

    Cohn, M. & Hughes, T. R. Jr. Nuclear magnetic resonance spectra of adenosine di- and triphosphate. II. Effect of complexing with divalent metal ions. J. Biol. Chem. 237, 176–181 (1962).

    CAS  PubMed  PubMed Central  Google Scholar 

  55. 55.

    Gaidamauskas, E. et al. Metal complexation chemistry used for phosphate and nucleotide determination: an investigation of the Yb3+-pyrocatechol violet sensor. J. Biol. Inorg. Chem. 13, 1291–1299 (2008).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  56. 56.

    Roger Phillips, S. J. Adenosine and the adenine nucleotides. Ionization, metal complex formation, and conformation in solution. Chem. Rev. 66, 501–527 (1966).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  57. 57.

    Fukui, K., Imamura, A. & Nagata, C. A molecular orbital treatment of phosphate bonds of biochemical interest. Bull. Chem. Soc. Jpn. 36, 1450–1453 (1963).

    CAS  Article  Google Scholar 

  58. 58.

    Muchowska, K. B., Varma, S. J. & Moran, J. Synthesis and breakdown of universal metabolic precursors promoted by iron. Nature 569, 104–107 (2019).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  59. 59.

    Benner, S. A., Kim, H.-J. & Biondi, E. Prebiotic chemistry that could not not have happened. Life 9, 84 (2019).

    CAS  PubMed Central  Article  PubMed  Google Scholar 

  60. 60.

    Mauksch, M., Wei, S., Freund, M., Zamfir, A. & Tsogoeva, S. B. Spontaneous mirror symmetry breaking in the aldol reaction and its potential relevance in prebiotic chemistry. Orig. Life Evol. Biosph. 40, 79–91 (2010).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  61. 61.

    Pilar Romero-Fernandez, M., Babiano, R. & Cintas, P. On the asymmetric autocatalysis of aldol reactions: The case of 4-nitrobenzaldehyde and acetone. A critical appraisal with a focus on theory. Chirality 30, 445–456 (2018).

    PubMed  Article  CAS  PubMed Central  Google Scholar 

  62. 62.

    Jacques, B., Coincon, M. & Sygusch, J. Active site remodeling during the catalytic cycle in metal-dependent fructose-1,6-bisphosphate aldolases. J. Biol. Chem. 293, 7737–7753 (2018).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  63. 63.

    Mlynarski, J. & Bas, S. Catalytic asymmetric aldol reactions in aqueous media—a 5 year update. Chem. Soc. Rev. 43, 577–587 (2014).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  64. 64.

    Fessner, W. D. et al. The mechanism of class II, metal-dependent aldolases. Angew. Chem. Int. Ed. 35, 2219–2221 (1996).

    CAS  Article  Google Scholar 

  65. 65.

    Kobes, R. D., Simpson, R. T., Vallee, B. L. & Rutter, W. J. Functional role of metal ions in a class II aldolase. Biochemistry 8, 585–588 (1969).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  66. 66.

    Otto, S., Bertoncin, F. & Engberts, J. Lewis acid catalysis of a Diels–Alder reaction in water. J. Am. Chem. Soc. 118, 7702–7707 (1996).

    CAS  Article  Google Scholar 

  67. 67.

    Frisch, M. et al. Gaussian 16, Revision A. 03. (Gaussian. Inc., Wallingford CT, 2016).

    Google Scholar 

Download references


We are grateful for the financial supports of the National Natural Science Foundation of China (Nos. 21703132, 21773149, 21273142), the Natural Science Foundation of Shaanxi Province of China (2019JQ161) and the Fundamental Research Funds for the Central Universities (GK201802001). We thank Dr. Xinai Guo for technical assistance for the EPR characterisation.

Author information




C.W. and Q.Q. contributed equally to this work. C.W. conceived and directed the project, analysed the data, wrote and revised the manuscript. Q.Q. performed the experiments, analysed the data and prepared the Supplementary Information. W.L. and J.D. carried out the DFT calculations. M.H., S.L., X.D., Y.G., P.W. and W.Z. conducted the spectroscopic characterisations. Y.C. directed the project and revised the manuscript. J.S.H revised and commented the manuscript.

Corresponding author

Correspondence to Changhao Wang.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wang, C., Qi, Q., Li, W. et al. A Cu(II)–ATP complex efficiently catalyses enantioselective Diels–Alder reactions. Nat Commun 11, 4792 (2020). https://doi.org/10.1038/s41467-020-18554-x

Download citation


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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing