
JOURNAL INFORMATION
==============================
NLM Title Abbreviation: Angew Chem Int Ed Engl
Journal ID: Angew Chem Int Ed Engl
Journal ID: ANIE

Angewandte Chemie (International Ed. in English)

ISSN: 1433-7851
EISSN: 1521-3773

ARTICLE INFORMATION
==============================
PMCID: PMC12416454
PMID: 40785269
DOI: 10.1002/anie.202513902
Article ID: ANIE202513902
Article version: 1
Subjects: Research Article, Research Article, Coordination Cages

Flexibility‐Aided Orientational Self‐Sorting and Transformations of Bioactive Homochiral Cuboctahedron Pd12L16

Chattopadhyay Subhasis 1 2 3
Durník Robin 2 4 5
Kiesilä Anniina Dr. 6
Kalenius Elina Dr. 6
Linnanto Juha M. Dr. 7
Babica Pavel Assoc. Prof. 5
Kuta Jan Dr. 5
Marek Radek Prof. Dr. 1 3 8
Jurček Ondřej Dr. https://orcid.org/0000-0002-9809-656X 1 2 3 jurceko@pharm.muni.cz

1 Department of Chemistry Faculty of Science Masaryk University Kamenice 5 Brno CZ‐62500 Czechia
2 Department of Natural Drugs Faculty of Pharmacy Masaryk University Palackého 1946/1 Brno CZ‐61200 Czechia
3 CEITEC–Central European Institute of Technology Masaryk University Kamenice 5 Brno CZ‐62500 Czechia
4 Department of Biochemistry Faculty of Science Masaryk University Kamenice 5 Brno CZ‐62500 Czechia
5 RECETOX Faculty of Science Masaryk University Kotlarska 2 Brno CZ‐61137 Czechia
6 Department of Chemistry University of Jyvaskyla P. O. Box 35 Jyväskylä FI‐40014 Finland
7 Institute of Physics University of Tartu W. Ostwald Street 1 Tartu 50411 Estonia
8 National Center for Biomolecular Research Faculty of Science Masaryk University Kamenice 5 Brno CZ‐62500 Czechia
* E‐mail: jurceko@pharm.muni.cz

Electronic publication date: 2025 Aug 10
Print publication date: 2025 Sep 8
Volume: 64
Issue: 37
Electronic Location ID: e202513902
Revised 2025 Jul 23; Received 2025 Jun 25; Accepted 2025 Jul 23
Copyright: © 2025 The Author(s). Angewandte Chemie International Edition published by Wiley‐VCH GmbH
License: This is an open access article under the terms of the http://creativecommons.org/licenses/by/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
License URL: https://creativecommons.org/licenses/by/4.0/

Keywords: Biological activity, Chirality, Self‐assembly, Structural transformation, Supramolecular chemistry

Funding: Czech Science Foundation 10.13039/501100001824 24–10760S Grant Agency of Masaryk University MUNI/A/1575/2023 MUNI/C/0123/2023 Core Facility NMR of CIISB, Instruct‐CZ Center, supported by MEYS CR LM2023042 European Regional Development Fund Project “UP CIISB” CZ.02.1.01/0.0/0.0/18_046/0015974 RECETOX Research Infrastructure LM2023069 European Union’s Horizon 2020 research and innovation program 857560 (CETOCOEN Excellence)
Figure count: 6
Table count: 0
Page count: 11

source-schema-version-number: 2.0
cover-date: September 8, 2025
details-of-publishers-convertor: Converter:WILEY_ML3GV2_TO_JATSPMC version:6.6.2 mode:remove_FC converted:08.09.2025

This work is dedicated to Professor Michaela Vorlíčková, a leading expert in CD spectroscopy of nucleic acids, on the occasion of her 80th birthday

S. Chattopadhyay , R. Durník , A. Kiesilä , E. Kalenius , J. M. Linnanto , P. Babica , J. Kuta , R. Marek , O. Jurček , Angew. Chem. Int. Ed.. 2025, 64, e202513902. 10.1002/anie.202513902 PMC12416454 40785269

==============================
Abstract

The rational design and selective self‐assembly of flexible and unsymmetric ligands into large coordination complexes is an eminent challenge in supramolecular coordination chemistry. Here, we present the coordination‐driven self‐assembly of natural ursodeoxycholic‐bile‐acid‐derived unsymmetric tris‐pyridyl ligand (L) resulting in the selective and switchable formation of chiral stellated Pd6 L 8 and Pd12 L 16 cages. The selectivity of the cage originates in the adaptivity and flexibility of the arms of the ligand bearing pyridyl moieties. The interspecific transformations can be controlled by changes in the reaction conditions. The orientational self‐sorting of L into a single constitutional isomer of each cage, i.e., homochiral quadruple and octuple right‐handed helical species, was confirmed by a combination of molecular modelling and circular dichroism. The cages, derived from natural amphiphilic transport molecules, mediate the higher cellular uptake and increase the anticancer activity of bioactive palladium cations as determined in studies using in vitro 3D spheroids of the human hepatic cells HepG2.

Mimicking concave morphologies found in nature, we have introduced flexible, unsymmetric, tridentate ligand (L) which undergoes coordination self‐assembly into highly organized chiral Pd6 L 8 or Pd12 L 16 cages that can undergo interspecific transformations. The cages, being based on natural amphiphilic transport molecules, mediate high cellular uptake and increase the anticancer activity of bioactive palladium cations.

Introduction

Natural chiral hydrophobic cavities are important for many biological functions, e.g., for recognition as parts of transport proteins or for substrate‐specific transformations as parts of enzymes. To understand and mimic these natural systems and their (supra)molecular mechanisms of action, the development of their artificial counterparts (e.g., cages, macrocycles) from chiral molecules is desirable. Easing such efforts, nature readily offers convenient chiral building blocks (terpenoids, amino acids, or carbohydrates) which can be utilized. Intermolecular‐interaction‐mediated self‐assembly together with metal coordination are essential natural processes to construct such higher‐order structures and can easily be adapted in the development of artificial systems.

The discovery of supramolecular coordination cages (SCCs) via coordination‐driven self‐assembly by Saalfrank et al.[ 1 ] led to the rapid development of a large group of metallo‐cycles and metallo‐cages mainly using rigid and symmetric bis/tris‐pyridyl coordinating ligands and tetravalent square‐planar Pd2+.[ 2 ] The edge‐directed self‐assembly of different bis‐pyridyl ligands resulted in the formation of SCCs and macrocycles with the general formula PdnL2n, e.g., Pd2L4, Pd3L6, Pd4L8, Pd6L12, Pd12L24, Pd24L48, Pd30L60, or Pd48L96– the largest SCC described so far.[ 3 ] Whereas tris‐pyridyl ligands led to Pd3nL4n SCCs, with only four types so far, i.e., Pd3L4,[ 4 , 5 , 6 ] Pd9L12 (only one),[ 7 ] Pd18L24 [ 8 ] (one) via edge‐directed self‐assembly, and the most common Pd6L8 [ 9 ] via face‐directed self‐assembly.

More than three decades after the first discovery, most of these metallo‐supramolecular complexes are achiral and symmetric. Several approaches have been employed to construct low‐symmetry, unsymmetric, or chiral coordination complexes using unsymmetric bidentate ligands,[ 10 ] a combination of multiple symmetric ligands (heteroleptic complexes),[ 11 , 12 ] or even single‐type symmetric ligands.[ 13 , 14 , 15 , 16 ] However, unlike natural systems, the presence of stereogenic carbons in the structure of ligands is rare, mostly limited to the peripheral areas of ligands and their resulting complexes,[ 17 ] e.g., using peptides,[ 18 , 19 ] pentasaccharide,[ 20 ] or short alkyl chains.[ 21 ]

To mimic the natural systems more closely and to develop the “next generation” SCCs, where the chiral centres will line their inner cavities requires careful design of ligands from inherently chiral natural compounds. Following this concept, Jurček et al. introduced the metallo‐supramolecular macrocycle Pd3L6 containing 60  stereogenic carbons using the natural bile acid (BA) ursodeoxycholic acid (UDCA) as a core for bis‐pyridyl ligands.[ 22 , 23 ] A comparable study reports Pd2L4 SCCs using cholic‐, deoxycholic‐, or lithocholic‐acid‐based bis‐pyridyl ligands.[ 24 ] Recently, it was shown that ligands derived from chenodeoxycholic acid, an epimer of UDCA, can form even larger complexes, namely, Pd2L4, Pd3L6, Pd4L8, Pd5L10, and Pd6L12, having 120 stereogenic carbons.[ 25 ] Other than this, intriguing coordination complexes have been reported using peptide‐based bis‐pyridyl ligands in the process of folding and assembly,[ 26 ] but leaving tritopic natural‐molecule‐based ligands unstudied.

Moreover, most ligands used to build coordination complexes are rather small and rigid. The directionality of the binding sites together with the rigidity of the ligand predefine their bend angles. These characteristics are crucial to control the self‐assembly processes of symmetric and rigid ligands. Small differences in the bend angle of the ligand lead to significant changes in the final self‐assembly products, e.g., a difference of 3° in the bend angle leads to the Pd24L48→Pd48L96 or Pd12L24→Pd24L48 transformation.[ 27 , 28 , 29 ] However, the use of small, and to some extent flexible, ligands typically results in a mixture of kinetically trapped coordination complexes.[ 30 ] In contrast, the large, flexible, and unsymmetric UDCA‐based bis‐pyridyl ligands, with a deviation of their bend angle up to 30° (90°–120°), showed the selective formation of a single constitutional isomer of Pd3L6.[ 22 , 23 , 25 ] In comparison, the epimeric flexible chenodeoxycholic‐acid‐based ligand spans a bend angle range of 70°–90° (Δ 20°) resulting in a mixture of macrocyclic complexes (Pd2L4, Pd3L6, Pd4L8, Pd5L10, and Pd6L12). It has been proposed that the greater flexibility of the unsymmetric ligand together with certain steric restrictions increase the probability of forming a single species via orientational self‐sorting.[ 25 ] Still, a better understanding of the effect of the structural behaviour of the ligand on the final self‐assembly is required.

Even after a rapid increase in the number of architectonically appealing SCCs, studies of their biomedical aspects and applications are limited.[ 31 ] However, recent contributions demonstrate their drug‐loading ability and promising anticancer therapeutical potential.[ 32 , 33 ] From this perspective, the utilization of natural molecules in the construction of biocompatible SCCs is appealing. Out of the vast library of natural compounds, BAs meet well the requirements for the design of chiral, unsymmetric, flexible, and biocompatible ligands. The BAs are biosynthesized in the human body, where they play a key role in the digestion and transport of lipids and lipid‐soluble nutrients within the enterohepatic circulation using various passive and active transport processes.[ 34 ] BAs are commercially easily available, enantiomerically pure (containing 9–11 chiral centres), and possess a conformationally defined rigid steroid skeleton decorated with hydroxyl groups and a flexible alkyl side chain bearing a carboxylic acid group, that can be easily synthetically transformed into pyridyl coordination sites.[ 22 , 23 , 24 , 25 ] However, preparation of the tris‐pyridyl ligand from a BA, its coordination‐driven self‐assembly with square‐planar tetravalent Pd2+, demonstration of the potential of the ligand for orientational self‐sorting and flexibility‐permitted selective self‐assembly, together with evaluation of its biomedical potential have been missing until now.

Results and Discussion

Natural UDCA was decorated in four synthetic steps with three 4‐aminopyridine groups, attached either through carbamate bonds to C3 and C7 hydroxyls or by an amide bond to the C24 carboxylic acid, resulting in a tridentate ligand (L) (Figure 1, Supporting Information section 2.1). A molecular model of L can be visualized as an elongated triangular panel (towards C24) having two faces, concave (α) and convex (β), containing either pyridyls (Py) or methyls (C18, C19, and C21), respectively (Figure 1).

Figure 1 Coordination‐driven self‐assembly of L into stellated helical octahedral Pd6 L 8 and cuboctahedral Pd12 L 16 SCCs and their transformation reactions: a) using [Pd(ACN)4](BF4)2, b) using Pd(NO3)2. The blue asterisk denotes chiral centres of the steroid skeleton.

The initial complexation of L (10 mM) with [Pd(ACN)4](BF4)2 (metal to ligand ratio M:L 3:4) in [D6]‐DMSO (70 °C, 1 h), marked as reaction mixture 1 (RM1) (Figure 1a), led to a quantitative (i.e., the spectrum lacks signals corresponding to the non‐coordinated free ligand) formation of coordination species as confirmed by 1H NMR spectroscopy (Figure 2a, green, Figure S13). The 1H NMR signals showed a high‐frequency coordination shift. The significant broadening of the 1H signals and the presence of multiple aromatic signals give rise to a few possibilities considering the unsymmetry and flexibility of L, the formation of: 1) a mixture of coordination complexes with varying molecular formula (size), 2) multiple architectures of a complex having the same molecular formula, 3) multiple constitutional isomers of a single architecture (varying C3‐C7‐C24‐pyridyl‐Pd2+ connectivity–rotation of triangular panel), 4) multiple conformational isomers of a coordination complex, or 5) a combination of the abovementioned possibilities.

Figure 2 NMR characterisation of Pd6 L 8 and Pd12L16. a) 1H NMR spectra of L, mixture of Pd6 L 8 and Pd12 L 16 (RM1), Pd6 L 8 (RM2 3:2), and Pd12 L 16 (RM2) in [D6]‐DMSO at 298.2 K and 700 MHz. 1H DOSY NMR spectra of b) Pd12 L 16 (RM2) and c) Pd6 L 8 (RM2 3:2) ([D6]‐DMSO, 303.2 K and 700 MHz).

In the next step, the reaction mixture was analysed by electrospray ionization (ESI‐MS) and ion mobility mass spectrometry (IM‐MS) (Figure S14) revealing charge state distributions for ions [Pd6 L 8(BF4)n](12‐n)+ (n = 1–8) and [Pd12 L 16(BF4)n](24‐n)+ (n = 6–16) (Table S1, being in the size range of small proteins, ca 7 and 14 kDa, respectively). The drift tube collision cross sections in nitrogen (DTCCSN2) of [Pd12 L 16(BF4)14]10+ and [Pd6 L 8(BF4)2]10+ were 1976 and 1420 Å2, corresponding roughly to diameters of 5.0 and 4.3 nm, respectively. According to the IM‐MS, the formation of the larger cage Pd12 L 16 was more abundant under these conditions (∼75% of Pd12 L 16 compared to Pd6 L 8). Furthermore, the number of species and their diffusion coefficients (D) were determined by 1H DOSY NMR (Figure S15).

The presence of two species was confirmed, and hydrodynamic diameters were calculated from the diffusion coefficients using the Stokes–Einstein equation, 4.8 nm (Pd6 L 8) and 5.8 nm (Pd12 L 16). This provided a moderate fit with the sizes observed using the IM‐MS data (Table S1) and confirmed the formation of Pd6 L 8 and Pd12 L 16 SCCs. Additionally, MS analysis confirmed that Pd12 L 16 is a monomer rather than a dimer of Pd6 L 8 based on the specific intensity abundance, concentration independence, collision cross section, and the presence of odd charge states of the ions (Supporting Information Section 2.2.1).

The selective formation of the Pd12 L 16 cage was achieved by dissolving L in [D6]‐DMSO (10 mM) and using a different metal salt Pd(NO3)2 at M:L 3:4 (70 °C, 1 h) (RM2; Figures 1, 2,b, Supporting Information Section 2.2.3). The selective formation of the Pd6 L 8 cage was achieved under two different reaction conditions, where L (10 mM) reacts with: 1) Pd(NO3)2 at M:L 3:2 ratio in [D6]‐DMSO (70 °C, 1 h) (RM2 3:2; Figures 1, 2,c, Supporting Information Section 2.2.3); and 2) Pd(NO3)2 at M:L 3:4 in the solvent mixture [D3]‐ACN:[D6]‐DMSO 5:95 (70 °C, 1 h) (RM3; Supporting Information Section 2.2.5). The solvent system was adjusted according to the low solubility of L in neat [D3]‐ACN. The 1H NMR spectra of both cages in [D6]‐DMSO were distinguished by a sharp singlet signal at about 8.7 ppm indicating the presence of Pd6 L 8 (Figure 2a). The 1H DOSY NMR experiments then reliably confirmed the selective formation of Pd12 L 16 or Pd6 L 8 in both solvent systems (Table S3). The values of the diffusion coefficient (D) for products in RM2, RM2 3:2, and RM3 were 4.98−11 m2 s−1 (4.9 nm, Figure 2b), 5.98 × 10−11 m2 s−1 (4 nm, Figure 2c), and 6.24 × 10−11 m2 s−1 (the presence of [D3]‐ACN slightly increases D, Figure S33), respectively.

In the next step, we investigated a possible interspecific transformation of the coordination species followed by 1H NMR and 1H DOSY NMR techniques. Post‐synthetic addition of 5% (v/v) [D3]‐ACN to the [D6]‐DMSO solution of Pd12 L 16 (RM2) (heated at 70 °C, 1 h) led to complete conversion to Pd6 L 8 (Supporting Information Section 2.2.6.1). In another experiment, adding two equivalents of L to the [D6]‐DMSO solution of Pd6 L 8 (RM2 3:2), changing the M:L ratio from 3:2 to 3:4 (heated at 70 °C, 1 h), led to complete conversion to Pd12 L 16 (Supporting Information Section 2.2.6.2). Conversely, three equivalents of Pd(NO3)2 were added to the Pd12 L 16 solution (RM2), changing the ratio M:L from 3:4 to 3:2 (heated at 70 °C, 1 h). Spectroscopic analyses showed partial conversion of Pd12 L 16 to Pd6 L 8 (Supporting Information Section 2.2.6.3), even after prolonged heating. A comparison of the diffusion coefficients for all reaction products is shown in Table S3.

Considering these observations and the related literature,[ 35 , 36 ] we suggest a hypothesis for the reaction mechanism of self‐assembly processes. The coordination effect of the coordinating species participating in the reaction mixture decreases in the sequence: pyridyl >> DMSO > NO3 ‐ > acetonitrile >> BF4 −.[ 37 ] At the same time, their coordination ability should also be considered through their concentration factors, thus the effect of [D6]‐DMSO, present in all experiments in very high excess, can be considered as comparable. Therefore, we hypothesize that the L‐pyridyls↔NO3 −↔acetonitrile exchange affects the self‐assembly processes the most significantly (ignoring the effect of the weak coordination of BF4 −). A reaction mixture may contain a Pd6 L 8 self‐assembly as a minor product in a mixture with Pd12 L 16 if [Pd(ACN)4](BF4)2 (3:4 M:L, Pd:ACN ratio 1:4) was used or as neat single product when using Pd(NO3)2 with 5% (v/v) [D3]‐ACN in [D6]‐DMSO as the solvent system (Pd:ACN ratio 1:128). In the case of [Pd(ACN)4](BF4)2, Pd6 L 8 is present as a minor product in a mixture with Pd12 L 16 (as determined by comparison of the heights of the drift peaks by IM‐MS, Table S1), whereas in a large excess of ACN, Pd6 L 8 is the only species. Also, prolonged heating of the [Pd(ACN)4](BF4)2 reaction leads to an increase in the concentration of Pd6 L 8. In addition, the reaction of L (10 mM) with [Pd(ACN)4](BF4)2 at M:L 3:2 in [D6]‐DMSO (70 °C, 1 h) was carried out, where the Pd6 L 8 species was observed as the major product in a mixture with Pd12 L 16 (Table S2, Figure 1a). In general, the higher the concentration of ACN the higher the ratio of Pd6 L 8 to Pd12 L 16. Finally, we also performed an experiment where the product was prepared via RM1 and the reaction mixture was subsequently evaporated to dryness under vacuum to remove/reduce the ACN content. The solid residue was re‐dissolved in [D6]‐DMSO at 70 °C (RM1–ACN). 1H NMR spectroscopy showed only a trace amount of ACN remaining (the ratio between Pd:ACN was reduced from 1:4 to 1:0.2, Figure S16). The 1H‐DOSY NMR spectrum showed complete conversion of the Pd12 L 16 and Pd6 L 8 mixture into Pd12 L 16 only (Figure S17), which further confirms that it is the elevated amount of ACN that directs the reaction towards Pd6 L 8. In this experiment, we could also compare the effect of a palladium counter‐anion, having the reaction mixture containing either BF4 − or NO3 − anion in [D6]‐DMSO in [M]:[L] = 3:4 ratio. Both reaction mixtures provided the Pd12 L 16 complex, suggesting that the anion effect on the final self‐assembly is similar or negligible, unlike the eminent effect of ACN.

A similar situation can also be observed in the case of varying Pd(NO3)2 concentration, i.e., a higher concentration of Pd(NO3)2 in RM2 3:2 leads to selective formation of Pd6 L 8 in comparison to RM2, or the addition of an excessive amount of Pd(NO3)2 to Pd12 L 16 (prepared under conditions of RM2), shifting the molar ratio M:L from 3:4 to 3:2 (resulting in three equivalents of Pd(NO3)2 present as a free salt in the reaction mixture), leads to a partial transformation to Pd6 L 8. In contrast, the addition of L to RM2 3:2, shifting the M:L from 3:2 to 3:4 (decreasing the concentration of free Pd(NO3)2), promotes the formation of Pd12 L 16.

To further distinguish the effect of nitrate or palladium, we have carried out additions of TBANO3 to RM1 (Supporting Information Section 2.2.2.2). 1.5 eq. (M:L:NO3 ‐ 3:4:6) and 3 eq. (M:L:NO3 ‐ 3:4:12) of TBANO3 with respect to L were subsequently added and the reaction mixture was heated at 70 °C for 1 h upon every addition (Figure S18a,b). 1H DOSY NMR analysis showed that the solutions contained mixtures of SCCs (even after continued heating for 24 h) (Figure S18c). These experiments indicate that the addition of nitrate does not influence strongly the equilibrium between the SCCs and it is rather the palladium concentration or the ACN content that control the interspatial transformation and direct the equilibrium towards the Pd6 L 8 product.

The presence of two different species in the reaction mixture suggests the formation of thermodynamic and kinetically trapped species. A previous study by Hiraoka et al. showed that the use of ACN as solvent accelerates the ligand exchange rate (solvent‐assisted exchange self‐assembly process) which prevents the formation of primitive intermediates leading to kinetic products, thus it is promoting the formation of a thermodynamic product.[ 36 ] The participation of ACN in the self‐assembly process results in preference for the Pd6 L 8 complex which can thus be seen as a thermodynamic product, whereas its absence in DMSO shifts the equilibrium towards the kinetic product Pd12 L 16.

If we consider that ACN allows greater coordination bond reversibility and increases the dynamic character of the Pd(II)‐ligand coordination bond, a mechanism of reaction selectivity could be deduced. In the process of self‐assembly, ligands form variously sized oligomers with metals before the final transformation into a cage (cyclization).[ 38 ] In the presence of an increased concentration of ACN, the more dynamic ligand exchange can be responsible for the formation of a greater number of smaller oligomers, ultimately leading to a smaller Pd6 L 8 complex. This can also be true with an elevated concentration of Pd(NO3)2 where the increased concentration of palladium cations plays a key role in the formation of the oligomer (the higher the Pd2+ concentration, the smaller the oligomer and the smaller the final complex). In contrast, in the absence of ACN, the oligomers grow larger as the ligand exchange is slower, ultimately leading to large kinetically trapped oligomers which eventually transform into Pd12 L 16.

Nuclear shielding in diamagnetic systems is highly local in nature, therefore the measured 1H NMR shifts for ligands in Pd6 L 8 and Pd12 L 16 are very similar, also being composed of similarly built subunits (Figure 3a). However, the larger Pd12 L 16 cage has a slightly longer correlation time (shorter relaxation time) and therefore somewhat broader NMR lines (Figure 2a). Additionally, we assume that the line width and apparent presence of multiple sets for the proton NMR signals of pyridyls and the ‐NH‐ groups indicate the representation of multiple stable conformational isomers of L in the complexes. In this regard, a variable temperature 1H NMR study of RM2 was conducted over the temperature range 298.2–398.2 K in increments of 10 K (Figure S27, Supporting Information Section 2.2.4). The study shows sharpening of the proton signals with increasing temperature, but also the eventual appearance of a second set of signals to the detriment of the original set corresponding to Pd12 L 16. This can represent either formation of conformational or constitutional isomers, or the formation of a new coordination product (these could not be confirmed by MS). At the end of the experiment, the sample was cooled back down to 298.2 K and the 1H‐ and 1H DOSY NMR spectra that were recorded showed the presence of a single coordination species Pd12 L 16 but represented by somehow sharper and better distinguished 1H NMR signals (Figures S27 and S28). For comparison, the reaction RM2 was carried out (M:L 3:4, 10 mM L, Pd(NO3)2, [d6]‐DMSO) at 100 °C, heated for 1 h, and 1H‐ and 1H DOSY NMR spectra were recorded showing the formation of Pd12 L 16 similar to the previous case of incremental heating to 125 °C (Figure S29).

Figure 3 Computational models and cartoon representations. a) PdC24 L 4 building subunit, b) Pd6 L 8, c) Pd12 L 16, and d) nomenclatures used for the triangular panel.

To our understanding,[ 39 ] the appearance of a complex second set of multiplets with a significant chemical shift refers to the formation of a new coordination self‐assembly, most likely a new constitutional isomer (caused by rotation of the ligand(s)). This is more probable than the formation of new conformational isomers, as we assume that the spectral changes would be more subtle in such a case. Moreover, the structural changes are represented in a smaller fraction of the species and are temporary as they are connected to the elevated temperature and the original form is restored upon cooling. In contrast, the signal sharpening which remains even upon cooling could be related to conformational changes, i.e., in direct synthesis at 70 °C for 1 h, some ligands are trapped in local minima, whereas heating to higher temperatures helps them reach global minima conformations corresponding better to complex spatial requirements (where they also remain upon cooling).

As we previously foretold, comparing the self‐assemblies of epimeric bis‐pyridyl ligands derived from UDCA and chenodeoxycholic acid[ 25 ] the structural selectivity for the self‐assembly lies in the ligand's geometry and the flexibility provided by the bending of the ligand molecule but also by a conformational change through rotation on the C3‐O‐CO‐NH‐pyridine axis.

Concerning the present ligand L, we believe that even larger conformational changes in the near vicinity of pyridyls, i.e., on the C3‐O‐CO‐NH‐pyridine, C7‐O‐CO‐NH‐pyridine, and C24‐CO‐NH‐pyridine axes, are manifested in solution. The conformational flexibility was further demonstrated by experiments using the mono‐pyridyl ligands LM3 and LM24 , where the binding moiety is at either position C3 or C24, respectively. For more details, please see Supporting Information Section 4.

Finally, having both neat species in hand, many attempts were made to grow a single crystal, but none of the crystals obtained showed X‐ray diffraction. Therefore, a combination of circular dichroism (CD) and molecular modelling was employed as an alternative for the structural analysis at the molecular level. Computational models of Pd6 L 8 and Pd12 L 16 were prepared and their energy was optimized using the DFT approach (Figure 3). An important steric restraint, resulting from the presence of two faces of L, is that the convex face should always point outside of the complex. Also, the elongated shape of L towards C24 further restricts its orientation in the final self‐assembly. This implies that the structure of L prohibits the formation of many theoretically possible constitutional isomers of SCCs. Following this, the structural subunit PdC24 L 4 was constructed, where four ligands are connected to Pd2+ by their C24‐pyridyls (marked as PdC24) with the β‐faces pointing outwards (Figure 3a, the model subunit was designed in correlation with a CD spectroscopic study which is described in detail later). Two of these subunits were interconnected through four Pd2+ ions with the remaining pyridyls at C3 and C7 (marked as PdC3‐C7), leading to Pd6 L 8. The resulting optimized structure of Pd6 L 8 represents an axially elongated, symmetric, and stellated octahedron (Figure 3b). To further support such structural organization, we performed semiempirical calculations on the Pd6 L 8 complex using the PM6 method (palladium charge balanced by two nitrates) to study the effect of the rotation of L on the structure (constitutional isomerism). Initially, we found that the rotation of L is geometrically preferred in the counterclockwise direction, whereas rotation in the clockwise direction leads to large distances between the pyridyls and the tris‐pyridyl‐coordinated Pd cation that significantly change the overall shape of the complex and increase its total energy by a few tens of kcal per mole. In contrast, the complex with the ligand counterclockwise‐rotated retains its original shape overall but leads to an energy increase of about 7 kcal mol−1 in comparison to the original symmetric complex suggested. If all the ligands in the complex were similarly rotated, the total energy would increase by about 40 kcal mol−1 (approximately 5 kcal mol−1 per ligand rotation). Finally, we compared the total energies of the ligands taken out of both optimized Pd6 L 8 complexes (original and rotated). The differences in total energy between these two ligand conformations result similarly in about 4–6 kcal mol−1, suggesting that the energy difference between the complexes originates mostly in the conformational change of the ligand, while the energy of the coordination interactions between Pd(II) and the pyridyls of the ligand seem to be quite similar for both complexes.

The structure of Pd12 L 16 was similarly constructed using four PdC24 L 4 subunits. Two PdC24 L 4 subunits were placed sideways to each other and connected by two PdC3‐C7. Two of these dimeric units were then joined together with the remaining PdC3‐C7 (resembling the structure of a tennis ball) and fulfilling the square planar coordination sphere by pyridyls forming the cuboctahedral Pd12 L 16 species. The geometry‐optimized structure of Pd12 L 16 illustrates a low‐symmetry stellated cuboctahedron, where each triangular face contains one L, two opposite square faces have two Ls each, and each remaining square face has one L in an alternating up and down (zig–zag) fashion (Figure 3c). This supramolecular cuboctahedron, cultivating the face‐directive self‐assembly of a tritopic ligand, is unprecedented.

The structural behaviour of L in Pd6 L 8 and Pd12 L 16 can be easily described and followed through the Pd–Pd distances. The average PdC24—PdC3‐C7 and PdC3‐C7—PdC3‐C7 distances for Pd6 L 8 are 18.4 and 15.6 Å, and those of Pd12 L 16 are 19.5 and 19.9 Å, respectively. The average angles of the triangular faces for Pd6 L 8 are 50° (for ∠PdC3—PdC24—PdC7) and 65° (for ∠PdC7—PdC3—PdC24 and ∠PdC3—PdC7—PdC24), whereas those of Pd12 L 16 contain 60° in both cases (Figure 3d, Tables S4–S7). Thus, the octahedron is made of isosceles triangles, while the cuboctahedron is composed of nearly equilateral triangular panels. The Pd3‐7—Pd3‐7 distances correlate to the NPy(C3)—NPy(C7) distance corresponding to a change of ∠NPy(C3)—C5—NPy(C7) bend angle of L. The ∠NPy(C3)—C5—NPy(C7) bend angle range is 90°–93° for Pd6 L 8 and 105°–127° for Pd12 L 16 (Figure S40). At the same time, a small increase in the PdC3‐C7—PdC24 distance for Pd12 L 16 demonstrates the decrease in the convexity of L. In summary, although the ligand is an unsymmetric building block in its nature, inside the complexes it can effectively adapt because of its flexibility and it is contained as a regular symmetric panel according to the steric and geometric requirements of the given self‐assembly (resembling the multicomponent self‐assembly process of viral capsids). In particular, L adapts to the structure of a stellated octahedron or cuboctahedron in directions NPy(C3)—NPy(C7), NPy(C3)—NPy(C24), NPy(C7)—NPy(C24) differing by 3.1, 1.0, and 0.8 Å, with respect to each other, with higher standard deviations observed for Pd12 L 16 (Figure 3d, Table S8). In comparison, the DFT‐optimized model of L (ground state in polar solvent, B3LYP/6–31G*) has a medial NPy(C3)—NPy(C7) distance to both species, an NPy(C7)—NPy(C24) distance similar to that of Pd6 L 8, and is 0.8 Å longer than in Pd12 L 16, but it is significantly stretched in the direction NPy(C3)—NPy(C24) by 5.0 and 4.0 Å as compared to Pd6 L 8 and Pd12 L 16, respectively (Table S8).

The selective formation of different complexes using a single usually small and relatively rigid ligand were previously achieved by: 1) a templating effect of the solvent[ 40 ] or counter anion,[ 41 , 42 ] 2) using different metal nodes (i.e., Pt2+ and Pd2+),[ 43 ] 3) using different binding sites of the same ligand,[ 44 ] or 4) using photoactive ligands via changing their conformation.[ 45 ] Fine balancing and prediction of the structural flexibility in the design of a ligand is a challenging task (often resulting in a mixture of kinetically trapped species with limited flexibility)[ 30 , 43 ] and the construction of larger species with flexible ligands rather relies on the method of trial‐and‐error. Therefore, the selective formation of coordination complexes utilizing the inherent flexibility of the ligand represents a novel approach, which we refer to as flexibility‐aided self‐assembly, moreover, considering the unsymmetry of the ligand, we propose to term this phenomenon flexibility‐aided orientational self‐sorting.

Finally, the CD spectra of L, Pd6 L 8, and Pd12 L 16 were recorded to investigate the chirality of the SCCs (Figure 4a). L gives a single negative CD band at 247 nm in the spectral region studied, whereas Pd6 L 8 and Pd12 L 16 display a strong positive band at 275 nm and strong negative band at 260 nm, resulting in a (+ −) CD couplet. The coordination sphere of PdC3‐C7 in models of Pd6 L 8 and Pd12 L 16 shows a clockwise C3‐C7‐C3‐C7 connectivity (3,7‐PdC3‐C7‐3,7) which induces right‐handed helicity (Δ), as observed on following two diagonally interconnected ligands through the C24‐C3‐Pd‐C3‐C24 backbone (Figures 4b and S41). This helicity is present in both cages, leading overall to a right‐handed quadruple (ΔΔΔΔ) for Pd6 L 8 or octuple helices (ΔΔΔΔΔΔΔΔ) for Pd12 L 16. In general, the structural organization of natural polymers into a right‐handed double helix is quite abundant, e.g., the most common form of DNA, canonical B‐DNA, is represented by a similar (+ −) CD couplet.[ 46 ] The CD spectra of the SCCs were further compared with the previously reported bis‐pyridyl UDCA‐based ligand Ld and its macrocyclic complex Pd3(Ld )6 [ 22 ] (Figure 4c) where, interestingly, almost the opposite (− +) CD couplet was observed (Figure 4a). It was demonstrated that a single crown‐like constitutional isomer of Pd3(Ld )6 was formed where the Pd2+ centres have a different clockwise connectivity C3‐C3‐C7‐C7 (3,3‐Pd‐7,7) (Figure 4c). If, similarly, the curvature of the C24‐C3‐Pd‐C3‐C24 backbone is followed, a hairpin structure is observed. This makes Pd3(Ld )6 an ensemble of three hairpin subunits accounting for the opposite CD couplet. Interestingly, a similar inverse CD couplet can also be observed in nature representing the duplex‐to‐hairpin transformation of B‐DNA.[ 46 ]

Figure 4 Structural analysis of supramolecular coordination complexes using CD spectroscopy. a) CD spectra of ligands and their coordination complexes in methanol at 25 °C. Interpretation of helical structures of b) Pd6 L 8 or Pd12 L 16, and c) Pd3(Ld )6, following the C24‐C3‐Pd‐C3‐C24 backbone.

To further support the structural assignment, we carried out a computational study at the time‐dependent DFT level to calculate the CD spectrum of Pd12 L 16. Even though the large size of the Pd12 L 16 complex and the computational demands connected with that led us to make certain approximations in the DFT calculations, the resulting spectrum using a fragment‐based approach is in very good agreement with the experimental one (Figure 4a, Supporting Information Section 3.1). The calculations also describe well the spectrum of Pd3(Ld )6 (Figure 4a).

Therefore, a different structural organization of the same steroidal backbone in a supramolecular complex can lead to a different CD spectrum. Finding such a relationship between the CD signal and the structure represents a very useful tool in elucidating the connectivity of the ligands, i.e., 3,7‐PdC3‐C7‐3,7 and the 24,24‐PdC24‐24,24 corresponding to it, confirming the structural isomerism in Pd6 L 8 and Pd12 L 16. On that account, revealing the origin of the CD signals in these SCCs and comparing them with known supramolecular species (Supporting Information Section 3.1) and natural products, the CD method provides a potent alternative for the structural determination of chiral supramolecular species comparable to the often challenging single‐crystal X‐ray diffraction.

Moreover, CD spectroscopy may serve as a reliable and sensitive method to study the aqueous stability of Pd6 L 8 and Pd12 L 16 at the low concentrations relevant for biological studies (Figure S42). Consequently, considering the inherent transport ability of BAs within the enterohepatic circulation and the biological activity of Pd2+, we used spheroids of the human hepatoblastoma cell line HepG2 as a 3D in vitro model of the target liver tissue. The hepatospheroids were exposed to samples of Pd(NO3)2, L, Pd6 L 8, and Pd12 L 16. While L has shown negligible toxicity after an 8‐d exposure, the spheroid viability in response to an inorganic salt or SCCs decreased with increasing concentrations of Pd2+ ions. A comparison of equimolar Pd2+ concentrations (24 µM) of Pd(NO3)2, Pd6 L 8, and Pd12 L 16 shows spheroid viability reduced to 90%, 78%, and 65% of the control, respectively, as determined by the adenosine triphosphate (ATP) assay (Figure 5a) or by following on the spheroid size (Figure S57a, Table S9). At the end of the experiments, the spheroids were isolated, carefully and thoroughly washed, and submitted for studies by inductively coupled plasma mass spectrometry (ICP‐MS) to determine the palladium content. The palladium content in the spheroids displays a significant negative correlation with their viability (Figure 5b) and size (Figure S57b). The spheroids treated with 24 µM Pd(NO3)2 contained a low palladium content of 0.15 ng per spheroid, whereas 4 µM Pd6 L 8 and 2 µM Pd12 L 16 delivered 1.5 and 4.26 ng of bioactive palladium per spheroid, respectively (Figure 5b, Table S8). This corresponds to 0.023% cellular absorption efficiency at 24 µM Pd(NO3)2 (in total 638.5 ng of Pd(II) was applied), while the Pd6 L 8 and the Pd12 L 16 showed 0.2% and 0.7% uptake efficiency, respectively, at the equivalent palladium dose. The intraspheroid to extracellular concentration (IC:EC) ratios calculated across different species exhibited even more pronounced differences due to the reduced spheroid size in response to SCCs. At the high palladium dose, the IC:EC ratio was 248 for Pd12 L 16 and 51 for Pd6 L 8, representing a 56‐fold and 12‐fold increase, respectively, compared to the IC:EC ratio of 4.4 observed for Pd(NO3)2 (Table S9, Figure S57c). These results show that with larger and more charged SCCs, the affinity for uptake by the hepatospheroids, and consequently their toxic effects, increased (Figures 5b and S57b).

Figure 5 Toxicological studies of the SCCs. a) Concentration‐response of HepG2 spheroid viability (ATP content) after 8 days of exposure to Pd(NO3)2, L, Pd6 L 8, and Pd12 L 16. The asterisk (*) indicates a statistically significant (P < 0.05) difference from the solvent control. b) Relation of spheroid viability to palladium content measured in spheroids. ρ represents Spearman's rank correlation coefficient with a P value.

Conclusion

The majority of the artificial metallo‐supramolecular complexes, unlike natural systems, are made of achiral, symmetric, and rigid ligands. In contrast, nature is ruled by the handedness of molecules, whereas the amino acids build and regulate living organisms through their left‐handedness, evolution assigned carbohydrates with right‐handed features. Control over the handedness of supramolecular coordination complexes (SCCs), by simply changing the connectivity of the ligand to a metal node, e.g., by introducing additional denticity to the ligand, is a very interesting feature which can lead to stimuli‐responsive interconversion of such cavity containing structures.

In this study, we provide a concept to design chiral, unsymmetric, flexible ligands employing natural bile acids in the construction of the “next generation” supramolecular coordination cages (SCCs). Implementing this concept we have introduced the first natural‐molecule‐based unsymmetric tris‐pyridyl ligand (L) derived from ursodeoxycholic bile acid. Because of the inherent flexibility, L can structurally adjust by expanding or contracting in coordination self‐assemblies with square‐planar Pd(II) into a Pd6 L 8 octahedron or a giant Pd12 L 16 cuboctahedron (5 nm in diameter). We refer to this controllable self‐assembly process as “flexibility‐aided orientational self‐sorting”. At the same time, Pd12 L 16 represents the very first face‐directed self‐assembly of a tridentate ligand into the cuboctahedron, introducing a new group of M12L16 species.

So far, product mixtures resulting from rigid and symmetric ligands usually contain smaller species as the kinetically trapped ones,[ 30 , 43 ] whereas, in our case, it was found that using the unsymmetric and flexible ligand, the larger Pd12 L 16 species represents the metastable form. Similarly in nature, while some large kinetic self‐assemblies might be less stable because of increased complexity and the potential for structural defects, others can be exceptionally stable because of robust intermolecular interactions and intricate design. Our example seems to be reaching the natural complexity.

The course of the reaction and the resulting product can be controlled by the choice of reaction conditions or later via interspecific transformations between Pd6 L 8 and Pd12 L 16. The process of structural switching has a special significance in view of the system adaptation to the environment representing a step forward towards artificial adaptive materials derived from natural products.

As for the structural elucidation, single‐crystal X‐ray diffraction is of paramount importance for SCCs. However, getting a quality single crystal is a challenging task in many cases, especially for flexible SCCs like ours. Accordingly, we employed various analytical techniques, with importance laid on CD spectroscopy in combination with molecular modelling, which confirm the orientational self‐sorting and structural isomerism of the SCCs into right‐handed quadruple (ΔΔΔΔ) or octuple (ΔΔΔΔΔΔΔΔ) helices. We envisage that CD spectroscopy, being a well‐established, sensitive, fast, and reliable tool for large chiral systems (e.g., DNA, proteins), can also be used to predict the structural organization of chiral metallo‐supramolecular complexes.

Because of their tailored composition, these water‐soluble SCCs show the enhanced cellular absorption and increased toxicity effect of palladium(II) cations when studied with spheroids of hepatoblastoma HepG2 cells. This anticancer effect of Pd2+ amplifies with increasing charge and size of the SCC. ICP‐MS analysis of hepatospheroids reveals 12‐fold (Pd6 L 8) or 56‐fold (Pd12 L 16) higher uptake of toxic Pd2+ than that of inorganic Pd(NO3)2. This interesting toxicological behaviour of bile‐acid‐based SCCs highlights the importance of developing natural molecule‐based coordination ligands and their supramolecular complexes for modern therapeutics.

The steroidal SCCs developed mimic natural hydrophobic cavities containing 80 (Pd6 L 8) or a record 160 chiral centres (Pd12 L 16) in their structures. Therefore, these SCCs, along with possible biomedical applications, are equally attractive for molecular recognition, enantioselective catalysis, and material science. We believe that the future of metallo‐supramolecular chemistry lies in the development and control of SCCs containing chiral nanocavities. Advancement in the field requires crucial understanding of their structural design and of their additional features such as substrate selective recognition, concerted interplay of reactive functionalities, or reversible dynamic structural changes for efficient substrate‐product turnover. The flexibility‐aided orientational self‐sorting we have introduced herein represents a novel approach leading to such supramolecular self‐assemblies at the interface of synthetic and biological fields. It will certainly be an interesting endeavour to continue in the future.

Supporting Information

The authors have cited additional references within the Supporting Information.[ 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 ]

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting Information

Acknowledgements

The authors thank Dr. Miroslava Bittová for measuring some of the MS spectra. This work has received support from the Czech Science Foundation (Grant No. 24–10760S to R.M.) and the Grant Agency of Masaryk University (MUNI/A/1575/2023 and MUNI/C/0123/2023). The authors acknowledge the Core Facility NMR of CIISB, Instruct‐CZ Center, supported by MEYS CR (LM2023042) and the European Regional Development Fund Project “UP CIISB” (No. CZ.02.1.01/0.0/0.0/18_046/0015974). This work has been supported by the RECETOX Research Infrastructure (LM2023069) financed by MEYS CR. This work was also supported by the European Union’s Horizon 2020 research and innovation program under grant agreement No 857560 (CETOCOEN Excellence). This publication reflects only the author's view, and the European Commission is not responsible for any use that may be made of the information it contains.

Open access publishing facilitated by Masarykova univerzita, as part of the Wiley ‐ CzechELib agreement.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.


1 R. W. Saalfrank , A. Stark , K. Peters , H. G. von Schnering , Angew. Chem. Int. Ed. Engl. 1988, 27, 851–853.
2 R. Chakrabarty , P. S. Mukherjee , P. J. Stang , Chem. Rev. 2011, 111, 6810–6918.21863792 10.1021/cr200077m PMC3212633
3 S. Saha , I. Regeni , G. H. Clever , Coord. Chem. Rev. 2018, 374, 1–14.
4 T. K. Ronson , C. Carruthers , J. Fisher , T. Brotin , L. P. Harding , P. J. Rizkallah , M. J. Hardie , Inorg. Chem. 2010, 49, 675–685.20000713 10.1021/ic901972h
5 S. Bandi , A. K. Pal , G. S. Hanan , D. K. Chand , Chem. – Eur. J. 2014, 20, 13122–13126.25138657 10.1002/chem.201403808
6 H. Dasary , M. Sarkar , D. K. Chand , Chem. Commun. 2022, 58, 8480–8483.10.1039/d2cc02837a 35792679
7 D. Preston , A. R. Inglis , J. D. Crowley , P. E. Kruger , Chem. Asian J. 2019, 14, 3404–3408.31483940 10.1002/asia.201901238
8 Q.‐F. Sun , S. Sato , M. Fujita , Nat. Chem. 2012, 4, 330–333.22437720 10.1038/nchem.1285
9 N. B. Debata , D. Tripathy , H. S. Sahoo , Coord. Chem. Rev. 2019, 387, 273–298.
10 D. Tripathy , N. B. Debata , K. C. Naik , H. S. Sahoo , Coord. Chem. Rev. 2022, 456, 214396.
11 S. Pullen , J. Tessarolo , G. H. Clever , Chem. Sci. 2021, 12, 7269–7293.34163819 10.1039/d1sc01226f PMC8171321
12 K. Wu , E. Benchimol , A. Baksi , G. H. Clever , Nat. Chem. 2024, 16, 584–591.38243023 10.1038/s41557-023-01415-7
13 W. Meng , T. K. Ronson , J. R. Nitschke , Proc. Natl. Acad. Sci 2013, 110, 10531–10535.23754424 10.1073/pnas.1302683110 PMC3696797
14 P. Bonakdarzadeh , F. Pan , E. Kalenius , O. Jurček , K. Rissanen , Angew. Chem. Int. Ed. 2015, 54, 14890–14893.10.1002/anie.201507295 26457686
15 P. Howlader , E. Zangrando , P. S. Mukherjee , J. Am. Chem. Soc. 2020, 142, 9070–9078.32315163 10.1021/jacs.0c03551
16 P. Howlader , S. Mondal , S. Ahmed , P. S. Mukherjee , J. Am. Chem. Soc. 2020, 142, 20968–20972.33284597 10.1021/jacs.0c11011
17 L.‐J. Chen , H.‐B. Yang , M. Shionoya , Chem. Soc. Rev. 2017, 46, 2555–2576.28452389 10.1039/c7cs00173h
18 T. Kikuchi , S. Sato , D. Fujita , M. Fujita , Chem. Sci. 2014, 5, 3257.
19 B. E. Barber , E. M. G. Jamieson , L. E. M. White , C. T. McTernan , Chem 2024, 10, 2792–2806.
20 S. Sato , Y. Yoshimasa , D. Fujita , M. Yagi‐Utsumi , T. Yamaguchi , K. Kato , M. Fujita , Angew. Chem. 2015, 127, 8555–8559.10.1002/anie.201501981 26015171
21 Y. Domoto , K. Yamamoto , S. Horie , Z. Yu , M. Fujita , Chem. Sci. 2022, 13, 4372–4376.35509455 10.1039/d2sc00111j PMC9006968
22 O. Jurček , P. Bonakdarzadeh , E. Kalenius , J. M. Linnanto , M. Groessl , R. Knochenmuss , J. A. Ihalainen , K. Rissanen , Angew. Chem. Int. Ed. 2015, 54, 15462–15467.10.1002/anie.201506539 26530790
23 O. Jurček , Nonappa, E. K. , P. Jurček , J. M. Linnanto , R. Puttreddy , H. Valkenier , N. Houbenov , M. Babiak , M. Peterek , A. P. Davis , R. Marek , K. Rissanen , Cell Rep. Phys. Sci. 2021, 2, 100303.
24 S. K. Sen , R. Natarajan , Inorg. Chem. 2019, 58, 7180–7188.31117622 10.1021/acs.inorgchem.8b03610
25 O. Jurček , S. Chattopadhyay , E. Kalenius , J. M. Linnanto , A. Kiesilä , P. Jurček , P. Radiměřský , R. Marek , Angew. Chem. Int. Ed. 2024, 63, e202409134.10.1002/anie.202409134 38845398
26 T. Sawada , M. Fujita , Chem 2020, 6, 1861–1876.
27 D. Fujita , Y. Ueda , S. Sato , N. Mizuno , T. Kumasaka , M. Fujita , Nature 2016, 540, 563–566.30905932 10.1038/nature20771
28 J. Bunzen , J. Iwasa , P. Bonakdarzadeh , E. Numata , K. Rissanen , S. Sato , M. Fujita , Angew. Chem. Int. Ed. 2012, 51, 3161–3163.10.1002/anie.201108731 22328185
29 H. Yokoyama , Y. Ueda , D. Fujita , S. Sato , M. Fujita , Chem. – Asian J. 2015, 10, 2292–2295.26097112 10.1002/asia.201500519
30 D. Fujita , Y. Ueda , S. Sato , H. Yokoyama , N. Mizuno , T. Kumasaka , M. Fujita , Chem 2016, 1, 91–101.
31 A. Casini , B. Woods , M. Wenzel , Inorg. Chem. 2017, 56, 14715–14729.29172467 10.1021/acs.inorgchem.7b02599
32 T. R. Cook , V. Vajpayee , M. H. Lee , P. J. Stang , K.‐W. Chi , Acc. Chem. Res. 2013, 46, 2464–2474.23786636 10.1021/ar400010v PMC3833955
33 H. Sepehrpour , W. Fu , Y. Sun , P. J. Stang , J. Am. Chem. Soc. 2019, 141, 14005–14020.31419112 10.1021/jacs.9b06222 PMC6744948
34 R. Durník , L. Šindlerová , P. Babica , O. Jurček , Molecules 2022, 27, 2961.35566302 10.3390/molecules27092961 PMC9103499
35 Y. Yan , J. Huang , B. Z. Tang , Chem. Commun. 2016, 52, 11870–11884.10.1039/c6cc03620a 27494003
36 S. Kai , Y. Sakuma , T. Mashiko , T. Kojima , M. Tachikawa , S. Hiraoka , Inorg. Chem. 2017, 56, 12652–12663.28945075 10.1021/acs.inorgchem.7b02152
37 R. Díaz‐Torres , S. Alvarez , Dalton Trans. 2011, 40, 10742.21927754 10.1039/c1dt11000d
38 S. Komine , S. Takahashi , T. Kojima , H. Sato , S. Hiraoka , J. Am. Chem. Soc. 2019, 141, 3178–3186.30689952 10.1021/jacs.8b12890
39 It was previously observed (please see reference 22) with a similar bile‐acid‐based ligand, that ligand's rotation in either dimeric Pd2L2Cl4 (changing partially from coordination directionality C3‐PdCl2‐C7 to C3‐PdCl2‐C3 or C7‐PdCl2‐C7) or trimeric Pd3L3Cl6 (undergoing similar ligand orientational transformation) can be similarly seen as two sets of 1H NMR signals.
40 K. Suzuki , M. Kawano , M. Fujita , Angew. Chem. 2007, 119, 2877–2880.
41 H. Lee , J. Han , D. Kim , O.‐S. Jung , Dalton Trans. 2021, 50, 14849–14854.34604882 10.1039/d1dt00859e
42 T. Zhang , L.‐P. Zhou , X.‐Q. Guo , L.‐X. Cai , Q.‐F. Sun , Nat. Commun. 2017, 8, 15898.28621312 10.1038/ncomms15898 PMC5481752
43 D. Fujita , A. Takahashi , S. Sato , M. Fujita , J. Am. Chem. Soc. 2011, 133, 13317–13319.21809891 10.1021/ja2059236
44 P. Howlader , P. Bhandari , D. Chakraborty , J. K. Clegg , P. S. Mukherjee , Inorg. Chem. 2020, 59, 15454–15459.32997933 10.1021/acs.inorgchem.0c02452
45 M. Han , Y. Luo , B. Damaschke , L. Gómez , X. Ribas , A. Jose , P. Peretzki , M. Seibt , G. H. Clever , Angew. Chem. Int. Ed. 2016, 55, 445–449.10.1002/anie.201508307 26609916
46 J. Kypr , I. Kejnovská , K. Bednářová , M. Vorlíčková , in Compr. Chiroptical Spectrosc. John Wiley & Sons, Ltd, Hoboken, New Jersey, USA 2012, 2, pp. 575–586, 10.1002/9781118120392.ch17.
47 S. M. Stow , T. J. Causon , X. Zheng , R. T. Kurulugama , T. Mairinger , J. C. May , E. E. Rennie , E. S. Baker , R. D. Smith , J. A. McLean , S. Hann , J. C. Fjeldsted , Anal. Chem. 2017, 89, 9048–9055.28763190 10.1021/acs.analchem.7b01729 PMC5744684
48 V. Gabelica , A. A. Shvartsburg , C. Afonso , P. Barran , J. L. P. Benesch , C. Bleiholder , M. T. Bowers , A. Bilbao , M. F. Bush , J. L. Campbell , I. D. G. Campuzano , T. Causon , B. H. Clowers , C. S. Creaser , E. De Pauw , J. Far , F. Fernandez‐Lima , J. C. Fjeldsted , K. Giles , M. Groessl , C. J. Hogan , S. Hann , H. I. Kim , R. T. Kurulugama , J. C. May , J. A. McLean , K. Pagel , K. Richardson , M. E. Ridgeway , F. Rosu , et al., Mass Spectrom. Rev. 2019, 38, 291–320.30707468 10.1002/mas.21585 PMC6618043
49 M. J. Frisch , G. W. Trucks , H. B. Schlegel , G. E. Scuseria , M. A. Robb , J. R. Cheeseman , G. Scalmani , V. Barone , B. Mennucci , G. A. Petersson , H. Nakatsuji , M. Caricato , X. Li , H. P. Hratchian , A. F. Izmaylov , J. Bloino , G. Zheng , J. L. Sonnenberg , M. Hada , M. Ehara , K. Toyota , R. Fukuda , J. Hasegawa , M. Ishida , T. Nakajima , Y. Honda , O. Kitao , H. Nakai , T. Vreven , J. A. Montgomery Jr. , et al., Gaussian 09, Revision D.01., 2013, http://www.rsc.org/suppdata/c5/sc/c5sc02423d/c5sc02423d1.pdf.
50 J. Linnanto , J. Korppi‐Tommola , J. Phys. Chem. A 2004, 108, 5872–5882.
51 M. Schneider , M. F. Grossi , D. Gadara , Z. Spáčil , P. Babica , L. Bláha , J. Hazard. Mater. 2022, 424, 127447.34688008 10.1016/j.jhazmat.2021.127447
52 C. Gütz , R. Hovorka , C. Klein , Q.‐Q. Jiang , C. Bannwarth , M. Engeser , C. Schmuck , W. Assenmacher , W. Mader , F. Topić , K. Rissanen , S. Grimme , A. Lützen , Angew. Chem. Int. Ed. 2014, 53, 1693–1698.10.1002/anie.201308651 24453210
53 T. R. Schulte , J. J. Holstein , G. H. Clever , Angew. Chem. Int. Ed. 2019, 58, 5562–5566.10.1002/anie.201812926 PMC6563462 30761694
54 K. Wu , J. Tessarolo , A. Baksi , G. H. Clever , Angew. Chem. Int. Ed. 2022, 61, e202205725.10.1002/anie.202205725 PMC9544203 35616285
