Abstract
Chromones represent a privileged scaffold in medicinal chemistry and are an omnipresent structural motif in natural products. Chemically encoded non-natural peptidomimetics feature improved stability towards enzymatic degradation, cell permeability and binding affinity, translating into a considerable impact on pharmaceutical industry. Herein, a strategy for the sustainable assembly of chromones via electro-formyl C–H activation is presented. The rational design of the rhodaelectro-catalysis is guided by detailed mechanistic insights and provides versatile access to tyrosine-based fluorogenic peptidomimetics.
Similar content being viewed by others
Introduction
C–H activation has surfaced as a transformative strategy for molecular synthesis, with remarkable applications to materials science and late-stage diversification1,2,3,4,5,6. Metal-catalyzed hydroacylations have attracted major attention due to their high atom-economy, providing efficient access to substituted ketones7,8. In this context, functionalizations of hydroxy-benzaldehydes have proven to be a particularly enabling approach for the assembly of oxygen-containing heterocycles9,10,11, such as β-hydroxyketones, aurones, coumarines, and chromones12,13,14,15,16. Despite of major advances, this approach was largely limited by the need for stoichiometric amounts of chemical oxidants, compromising the inherent sustainable nature of the formyl C–H activation strategy. In recent years, a renaissance of organic electrosynthesis17,18,19,20,21,22,23 has provided a major impetus for efficient C–H activations24,25,26. Thereby, stoichiometric amounts of, often toxic, metal-containing oxidants can be avoided. Yet, while major advances in electro-organic synthesis have been noted27,28,29, challenging electro-functionalizations of oxidation-sensitive aldehydes continue to be scarce. Thus, the lability of aldehydes towards decarbonylation, overoxidation, and nucleophilic attack renders an electrochemical approach by anodic electro-oxidation particularly difficult7,8. The chromone scaffold is among others present in the commercialized drugs flavoxate and nedocromil30,31 as well as in bioactive natural products of relevance to acute myeloid leukemia32 and antiviral activity against SARS-associated corona viruses (Fig. 1a)33,34.
a Chromones as a privileged motive in pharmaceutical and bioactive compounds. b Strategy to access chromones via rhodaelectro catalysis and its use for electro-peptide labeling. c Synthesis of rhodium complex Rh-I and investigations of its redox properties by cyclic voltammetry in CH2Cl2 with nBu4NPF6 (0.2 M). d Reaction development, 0.25–0.50 mmol scale, 4.0–8.0 mL solvent, isolated yields. aWith nBu4NPF6 (0.1 M) for 4 h. Cp* pentamethylcyclopentadienyl, NaOPiv sodium pivalate, Fc ferrocene, GF graphite felt electrode, CCE constant current electrolysis, CPE constant potential electrolysis, tAmylOH 2-methyl-2-butanol.
In this work, we present an electro-formyl C–H activation via rhodaelectro catalysis for the assembly of substituted chromones to provide sustainable access to amino acid chromone hybrids and to label peptides35,36 through metallaelectro catalysis.
Results and discussion
To put our hypothesis into practice, we designed intermediates that feature lower oxidation potentials than the sensitive aldehyde substrates themselves (Fig. 1b). To this end, a stoichiometric transformation of hydroxybenzaldehyde 1, alkyne 2 and [Cp*RhCl2]2 in the presence of base delivered rhodium(I) complex Rh-I (Fig. 1c), which was unambiguously characterized by X-ray diffraction analysis. With the proposed key intermediates in hand, we probed their redox properties towards an oxidation manifold under electrocatalytic conditions. Studies by cyclic voltammetry revealed that the complex Rh-I underwent irreversible oxidation to Rh(III) at Ep = –0.11 V vs. Fc0/+ and therefore exhibited a considerably lower oxidation potential than benzaldehyde 1 (Ep = 1.68 V vs. Fc0/+). With respect to an oxidatively induced reductive elimination from a rhodium(III/IV)-species37,38, calculations by means of DFT at the PW6B95-D3(BJ)/def2-QZVP+SMD(acetonitrile)//PBE0-D3(BJ)/def2-SVP level of theory at 298.15 K revealed an oxidation potential of E1/2 = 0.48 V vs. Fc0/+ of the corresponding seven-membered rhoda(III)-cycle, which is in proximity to the experimentally determined, oxidation potentials of related species at Ep/2 = 0.68 V vs. Fc0/+37,38. Further computational studies of the electro-formyl C–H activation were supportive of a kinetically, favorable oxidatively induced reductive elimination (Supplementary Fig. 21).
The isolation and electroanalytical characterization of the key intermediate set the stage for studies on the electrocatalysis, initially with a constant potential of 1.0 V vs. Fc0/+, employing [Cp*RhCl2]2 as the catalyst and NaOPiv as the base to ensure an oxidatively induced reductive elimination. Hence, the desired chromone was obtained in 57% yield (Fig. 1d, entry 2), which is in line with our initial hypothesis for the rhodaelectro-catalyzed formyl activation. Further optimization demonstrated that the reaction furnished chromone 5 likewise under user-friendly galvanostatic conditions, with tAmOH/water (3:1) as the reaction medium avoiding additional electrolytes (Fig. 1d, entry 1). Control experiments revealed the crucial role of the rhodium precatalyst (entry 3). Importantly, the reaction was also viable with commercial equipment (Fig. 1d, entry 6). To test the role of electricity in the rhodaelectro-catalyzed formyl C–H activation, we performed an in-operando monitoring of the catalysis at different currents by in situ 1H-NMR spectroscopy (Fig. 2a).
a Influence of the applied current on the reaction rate. b Competition experiment. c KIE studies.
Indeed, the reaction rates are strongly dependent on the applied currents, indicating a turnover limiting electron transfer being operative. Additionally, an on/off experiment was conducted, clearly reflecting the key role of electricity for efficient catalyst turnover. To probe the catalysts mode of action, an intermolecular competition experiment between differently substituted salicylic aldehydes 6/7, revealed an inherent higher reactivity of the electron-rich substrate (Fig. 2b), being suggestive of a base-assisted internal electrophilic substitution-type (BIES) manifold39. A minor kinetic isotope effect was observed, again being in line with a rate limiting reoxidation (Fig. 2c).
To benchmark the presented electro-catalyzed formyl C–H activation, we compared its performance with challenging substrates 10, such as electron-deficient diphenylacetylenes and alkynes with aliphatic substituents. Thus, the efficacy towards the formation of products 12–15 was found to be uniquely effective under the electrocatalytic conditions, as compared to Cu(OAc)2 as the oxidant, highlighting the superior performance of the rhodaelectro catalysis (Fig. 3a). Additionally, the scalability of the rhodaelectro-catalyzed transformation was highlighted with a multigram scale synthesis with reduced catalyst loading (Fig. 3b).
a Comparison between rhodaelectro catalysis and Cu(OAc)2·H2O as the chemical oxidant. b Rhodaelectro-catalyzed annulation on multigram scale.
With the optimized conditions in hand, we explored the scope of salicylic aldehydes 16 (Fig. 4a). Overall, differently substituted aldehydes efficiently furnished the desired products 18–36. Especially redox sensitive groups, such as bromo-, iodo-, and thioether substituents were fully tolerated and delivered the corresponding products 27–29. The electron-deficient substrate with an ester substituent under water-free conditions, as well as salicylic aldehyde with an amine functionality were selectively converted to the corresponding chromones 30 and 31, respectively. Even bulky disubstituted aldehydes delivered the desired products 34 and 35 in very good yields. To our delight, also the estrone derivative was converted to the product 37.
a Aldehyde scope for various substituents. b Scope of viable alkynes. c Calculated energy profiles for the regioselective determining step for the Me/CF3 substituted alkyne at the PW6B95-D3(BJ)/def2-QZVP+SMD(methanol)//PBE0-D3(BJ)/def2-SVP level of theory and spatial localization of noncovalent interactions in the transition states. In the latter, red indicate repulsive interactions, with blue and green being attributed to strong and weak attractive interactions. aWith nBu4NPF6 (0.1 Μ) instead of H2O. b7 h. cRatio determined by 19F-NMR. dWith 5.0 mol % [Cp*RhCl2]2.
Subsequently, we turned our attention to the viable alkyne 10 scope in greater detail (Fig. 4b). Electron-rich alkynes were efficiently converted to the corresponding chromones 40 and 41. meta- and sterically encumbered ortho-arylalkynes were also tolerated by the rhodaelectro catalysis regime (42–45).
Subsequently, a wealth of unsymmetrically substituted alkynes were efficiently converted to the desired products 46–57 (Fig. 4b). Thereby, terminal as well as keto- and amido-substituted alkynes underwent the rhodaelectro catalysis with high efficacy (52–56). As to the regioselectivity of unsymmetrical alkynes, the alkyne substrates furnished chromones 49–55 with the aryl motif proximal to the oxygen-heteroatom. Under otherwise identical reaction conditions, alkynes with nitrile, ester or free acid groups provided less satisfactory results. It is noteworthy that hydroxyheptyne (10w) exclusively yielded the regioisomer 57. In order to gain insights into the origin of the regioselectivity for the synthesis of chromone 57, DFT calculations were carried out for the migratory insertion step at the PW6B95-D3(BJ)/def2-QZVP+SMD(methanol)//PBE0-D3(BJ)/def2-SVP level of theory (Supplementary Fig. 22). The transition state leading to regioisomer 57 with the hydroxyl distal to the carbonyl group of the substrate is favored by 2.4 kcal mol−1, which can be attributed to favorable hydrogen bonding interactions with the hydroxyl group in the transition state structure (TS(1-2), Supplementary Fig. 22). Likewise, the regioselectivity of the electrocatalytic C–H activation with 4-methyl-4’-trifluoromethyltolane could be rationalized by DFT computation (Fig. 4c). The calculated regioselectivity was in good agreement with the experimental observations. Here, noncovalent interactions were found to be of minor relevance.
Chemically encoded peptidomimetics reduce enzymatic degradation and feature superior binding affinities, cell permeability, and pharmacokinetics40,41. However, the functionalization of structurally complex peptides mostly rely on terminal peptides, azide-based click chemistry or on the innate reactivity of cysteine42,43,44. Given the practical importance of late-stage peptide diversifications, we became attracted to tyrosine modifications through rhodaelectro catalysis to access tyrosine-derived fluorescent amino acids. Indeed, tyrosines were chemo-selectively annulated with tolane and naphthalene derived alkynes furnishing the desired products 60–62. Next, we probed dipeptides to explore rhodaelectro-catalytic site-specific labeling (Fig. 5a). To our delight, a broad variety of dipeptides was efficiently converted to the corresponding products 63–68. Notably, even oxidation-sensitive serine (65) and methionine (66) containing peptides were transformed to the desired products. Furthermore, potentially coordinating dipeptides with unprotected tryptophan, or tyrosine regioselectively provided the desired products 67 and 68.
a Scope of rhodaelectro-catalyzed peptide labeling. b Photoelectrochemical annulation towards π-extended peptide labels. c Absorption and emission properties of 62 (10 mg/L) and 70 (0.5 mg/L) in CHCl3. d Late-stage functionalization of pentapeptide 71. aWith tAmylOH/H2O (3:1). b7 h. cWith 5.0 mol % of [Cp*RhCl2]2. dWith MeOH instead of EtOH.
We explored an improvement of the photoelectronic properties of the thus-obtained labels by a late-stage annulation. The aryl moieties were selectively transformed into π-extended peptide labels in a photoelectrochemical process to obtain the desired products (Fig. 5b). Labels 69 and 70 demonstrated improved photoelectronic properties in comparison to the corresponding diaryl precursors 60 and 62 (Fig. 5c and Supplementary Table 2). Since compound 70 exhibited intense fluorescence at 458 nm, it bears considerable potential as a fluorogenic probe45,46.
Finally, we probed our strategy for the functionalization of structurally complex oligopeptide 71. Indeed, peptide 72 was obtained in excellent yield, highlighting the unique power of the rhodaelectro-catalytic labeling strategy (Fig. 5d).
In summary, we have reported on a rhodaelectro-catalyzed transformation of hydroxy-benzaldehydes by electrochemical formyl C–H activation, featuring scalability, high functional group tolerance, and improved efficiency in comparison to chemical oxidants. The strategy proved applicable to the functionalization of tyrosine derivatives, enabling difficult to perform, site-selective electro-labeling of amino acids and peptides by formyl C–H activation. A mediated photoelectrochemical oxidation allowed for an enhancement of fluorescence properties of the thus-obtained amino acids. The rhodaelectro catalysis was inspired by in-depth mechanistic insights, through the isolation and electroanalytical characterization of key intermediates, providing strong support for an oxidation-induced reductive elimination.
Methods
Rhodaelectro-catalyzed formyl C–H activation
The electrocatalysis was carried out in an undivided cell, with a graphite felt (GF) anode (25 × 10 × 6.0 mm) and a platinum cathode (25 × 10 × 0.125 mm). The 2-hydroxylbenzaldehyde (0.75 mmol), the alkyne (0.25 mmol), NaOPiv (62 mg, 0.50 mmol), [Cp*RhCl2]2 (3.9 mg, 2.5 mol %), and tAmylOH/H2O (4.0 mL, 3:1) were placed in a 10 mL cell. Electrocatalysis was performed at 100 °C with a constant current of 4 mA maintained for 3.5–7.0 h. Then, the DC-power supply was stopped and the reaction mixture was diluted with EtOAc (2.0 mL). The electrodes were washed with EtOAc (Pt: 1 × 5.0 mL; C: 3 × 10.0 mL). The solvents were combined with the reaction mixture, silica gel was added and the solvents were removed in vacuo. Subsequent column chromatography on silica gel afforded the corresponding products.
Data availability
The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information files. The X-ray crystallographic coordinates for structures reported in this study have been deposited at the Cambridge Crystallographic Data Centre (CCDC), under deposition numbers 2046225, 2046228-2046229, 2046502-2046508. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. All other requests for materials and information should be addressed to the corresponding authors. Source data are provided with this paper.
References
Bergman, R. G. Organometallic chemistry: C–H activation. Nature 446, 391–393 (2007).
Yeung, C. S. & Dong, V. M. Catalytic dehydrogenative cross-coupling: forming carbon−carbon bonds by oxidizing two carbon−hydrogen bonds. Chem. Rev. 111, 1215–1292 (2011).
Wencel-Delord, J. & Glorius, F. C–H bond activation enables the rapid construction and late-stage diversification of functional molecules. Nat. Chem. 5, 369–375 (2013).
Larsen, M. A. & Hartwig, J. F. Iridium-catalyzed C–H borylation of heteroarenes: ccope, regioselectivity, application to late-stage functionalization, and mechanism. J. Am. Chem. Soc. 136, 4287–4299 (2014).
Saint-Denis, T. G., Zhu, R.-Y., Chen, G., Wu, Q.-F. & Yu, J.-Q. Enantioselective C(sp3)‒H bond activation by chiral transition metal catalysts. Science 359, eaao4798 (2018).
He, C., Whitehurst, W. G. & Gaunt, M. J. Palladium-catalyzed C(sp3)–H bond functionalization of aliphatic amines. Chem 5, 1031–1058 (2019).
Shen, Z., Dornan, P. K., Khan, H. A., Woo, T. K. & Dong, V. M. Mechanistic insights into the rhodium-catalyzed intramolecular ketone hydroacylation. J. Am. Chem. Soc. 131, 1077–1091 (2009).
Willis, M. C. Transition metal catalyzed alkene and alkyne hydroacylation. Chem. Rev. 110, 725–748 (2010).
Murphy, S. K., Coulter, M. M. & Dong, V. M. β-Hydroxy ketones prepared by regioselective hydroacylation. Chem. Sci. 3, 355–358 (2012).
Shi, Z., Schröder, N. & Glorius, F. Rhodium(III)-catalyzed dehydrogenative heck reaction of salicylaldehydes. Angew. Chem. Int. Ed. 51, 8092–8096 (2012).
Zeng, H. & Li, C.-J. A complete switch of the directional selectivity in the annulation of 2-hydroxybenzaldehydes with alkynes. Angew. Chem. Int. Ed. 53, 13862–13865 (2014).
Shimizu, M., Tsurugi, H., Satoh, T. & Miura, M. Rhodium-catalyzed oxidative coupling between salicylaldehydes and internal alkynes with C–H bond cleavage to produce 2,3-disubstituted chromones. Chem. Asian J. 3, 881–886 (2008).
Sun, P. et al. Controllable Rh(III)-catalyzed annulation between salicylaldehydes and diazo Compounds: divergent synthesis of chromones and benzofurans. Org. Lett. 18, 6464–6467 (2016).
Baruah, S., Kaishap, P. P. & Gogoi, S. Ru(II)-catalyzed C–H activation and annulation of salicylaldehydes with monosubstituted and disubstituted alkynes. Chem. Commun. 52, 13004–13007 (2016).
Yamane, S. et al. Iridium-catalyzed aerobic coupling of salicylaldehydes with alkynes: a remarkable switch of oxacyclic product. Chem. Eur. J. 24, 7852–7855 (2018).
Cai, L., Zhu, X., Chen, J., Lin, A. & Yao, H. Rh(III)-catalyzed C–H activation/annulation of salicylaldehydes with sulfoxonium ylides for the synthesis of chromones. Org. Chem. Front. 6, 3688–3692 (2019).
Sauer, G. S. & Lin, S. An electrocatalytic approach to the radical difunctionalization of alkenes. ACS Catal. 8, 5175–5187 (2018).
Yan, M., Kawamata, Y. & Baran, P. S. Synthetic organic electrochemical methods since 2000: on the verge of a renaissance. Chem. Rev. 117, 13230–13319 (2017).
Wiebe, A. et al. Electrifying organic synthesis. Angew. Chem. Int. Ed. 57, 5594–5619 (2018).
Francke, R. & Little, R. D. Redox catalysis in organic electrosynthesis: basic principles and recent developments. Chem. Soc. Rev. 43, 2492–2521 (2014).
Moeller, K. D. Using Physical organic chemistry to shape the course of electrochemical reactions. Chem. Rev. 118, 4817–4833 (2018).
Fuchigami, T. & Inagi, S. Recent advances in electrochemical systems for selective fluorination of organic compounds. Acc. Chem. Res. 53, 322–334 (2020).
Yamamoto, K., Kuriyama, M. & Onomura, O. Anodic oxidation for the stereoselective synthesis of heterocycles. Acc. Chem. Res. 53, 105–120 (2020).
Sauermann, N., Meyer, T. H., Qiu, Y. & Ackermann, L. Electrocatalytic C–H activation. ACS Catal. 8, 7086–7103 (2018).
Ma, C., Fang, P. & Mei, T.-S. Recent advances in C–H functionalization using electrochemical transition metal catalysis. ACS Catal. 8, 7179–7189 (2018).
Yang, Q.-L. et al. Electrochemistry-enabled Ir-catalyzed vinylic C–H functionalization. J. Am. Chem. Soc. 141, 18970–18976 (2019).
Gandeepan, P., Finger, L. H., Meyer, T. H. & Ackermann, L. 3d Metallaelectrocatalysis for resource economical syntheses. Chem. Soc. Rev. 49, 4254–4272 (2020).
Liu, J., Lu, L., Wood, D. & Lin, S. New redox strategies in organic synthesis by means of electrochemistry and photochemistry. ACS Cent. Sci. 6, 1317–1340 (2020).
Yang, L. et al. Electrochemically enabled C3 formylation and -acylation of Indoles with aldehydes. Org. Lett. 21, 7702–7707 (2019).
Gaspar, A., Matos, M. J., Garrido, J., Uriarte, E. & Borges, F. Chromone: a valid scaffold in medicinal chemistry. Chem. Rev. 114, 4960–4992 (2014).
Keri, R. S., Budagumpi, S., Pai, R. K. & Balakrishna, R. G. Chromones as a privileged scaffold in drug discovery: a review. Eur. J. Med. Chem. 78, 340–374 (2014).
Zocchi, L., Wu, S. C., Wu, J., Hayama, K. L. & Benavente, C. A. The cyclin-dependent kinase inhibitor flavopiridol (alvocidib) inhibits metastasis of human osteosarcoma cells. Oncotarget 9, 23505–23518 (2018).
Kim, M. K. et al. 2,6-Bis-arylmethyloxy-5-hydroxychromones with antiviral activity against both hepatitis C virus (HCV) and SARS-associated coronavirus (SCV). Eur. J. Med. Chem. 46, 5698–5704 (2011).
Kim, M. K., Yoon, H., Barnard, D. L. & Chong, Y. Design, synthesis and antiviral activity of 2-(3-Amino-4-piperazinylphenyl)chromone derivatives. Chem. Pharm. Bull. (Tokyo) 61, 486–488 (2013).
Krueger, A. T. & Imperiali, B. Fluorescent amino acids: modular building blocks for the assembly of new tools for chemical biology. Chembiochem 14, 788–799 (2013).
Harkiss, A. H. & Sutherland, A. Recent advances in the synthesis and application of fluorescent α-amino acids. Org. Biomol. Chem. 14, 8911–8921 (2016).
Kong, W.-J. et al. Flow rhodaelectro-catalyzed alkyne annulations by versatile C–H activation: mechanistic support for rhodium(III/IV). J. Am. Chem. Soc. 141, 17198–17206 (2019).
Wu, Z.-J. et al. Scalable rhodium(III)-catalyzed aryl C−H phosphorylation enabled by anodic oxidation induced reductive elimination. Angew. Chem. Int. Ed. 58, 16770–16774 (2019).
Rogge, T., Oliveira, J. C. A., Kuniyil, R., Hu, L. & Ackermann, L. Reactivity-controlling factors in carboxylate-assisted C–H activation under 4d and 3d transition metal catalysis. ACS Catal. 10, 10551–10558 (2020).
Henninot, A., Collins, J. C. & Nuss, J. M. The current state of peptide drug discovery: back to the future? J. Med. Chem. 61, 1382–1414 (2018).
Blaskovich, M. A. T. Unusual amino acids in medicinal chemistry. J. Med. Chem. 59, 10807–10836 (2016).
Noisier, A. F. M. & Brimble, M. A. C–H Functionalization in the synthesis of amino acids and peptides. Chem. Rev. 114, 8775–8806 (2014).
Ohata, J., Martin, S. C. & Ball, Z. T. Metal-mediated functionalization of natural peptides and proteins: panning for bioconjugation gold. Angew. Chem. Int. Ed. 58, 6176–6199 (2019).
Wang, W., Lorion, M. M., Shah, J., Kapdi, A. R. & Ackermann, L. Late-stage peptide diversification by position-selective C−H activation. Angew. Chem. Int. Ed. 57, 14700–14717 (2018).
Gonçalves, M. S. T. Fluorescent labeling of biomolecules with organic probes. Chem. Rev. 109, 190–212 (2009).
Toseland, C. P. Fluorescent labeling and modification of proteins. J. Chem. Biol. 6, 85–95 (2013).
Acknowledgements
Generous support by the DFG (Gottfried-Wilhelm-Leibniz prize) is gratefully acknowledged. We thank Dr. Christopher Golz (University Göttingen) for the X-ray diffraction analysis.
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
M.S. conducted the experimental mechanistic studies, unraveled the rhodaelectro-catalyzed formyl C–H functionalization, and explored the substrate scope, initially assisted by H.Y. A.M.M. isolated and characterized rhodium complexes. J.O. performed the computational studies. L.A. conceived and supervised the project. L.A. and M.S. wrote the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Communications thanks Hai-Chao Xu and the other anonymous reviewer(s) for their contribution to the peer review of this work.
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/.
About this article
Cite this article
Stangier, M., Messinis, A.M., Oliveira, J.C.A. et al. Rhodaelectro-catalyzed access to chromones via formyl C–H activation towards peptide electro-labeling. Nat Commun 12, 4736 (2021). https://doi.org/10.1038/s41467-021-25005-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-021-25005-8
Comments
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.
