Metal–Organic Cages: Applications in Organic Reactions

: Supramolecular metal–organic cages, a class of molecular containers formed via coordination-driven self-assembly, have attracted sustained attention for their applications in catalysis, due to their structural aesthetics and unique properties. Their inherent confined cavity is considered to be analogous to the binding pocket of enzymes, and the facile tunability of building blocks offers a diverse platform for enzyme mimics to promote organic reactions. This minireview covers the recent progress of supramolecular metal–organic coordination cages for boosting organic reactions as reaction vessels or catalysts. The developments in the utilizations of the metal–organic cages for accelerating the organic reactions, improving the selectivity of the reactions are summarized. In addition, recent developments and successes in tandem or cascade reactions promoted by supramolecular metal–organic cages are discussed.


Introduction
Catalysis, one of the twelve principles of green chemistry [1], is considered to be a fundamental technology for saving energy and protecting the environment [2].The ultimate aim of catalysis is to improve catalytic efficiency and reduce waste with environmentally friendly reagents [3].Considerable and continued research attention has been attracted by supramolecular catalysis, which can control the chemical reactivity by exploiting the intermolecular interaction [4][5][6].By mimicking the working principles of structurally much more complex enzymes, the relatively simple artificial supramolecular catalysts with discrete and confined cavities can accomplish specific and powerful catalysis, with rate enhancements of 10 5 -fold and higher compared to classical homogeneous catalysis in bulk solvent [7][8][9].A number of different artificial supramolecular host systems have been developed in the last several decades, constructed by covalent bonds [10], the hydrophobic effect [11], hydrogen bonds [12], and metal-ligand interactions [13,14].Among them, metalorganic cages have shown several advantages in their application for boosting organic reactions [15,16].(i) The self-assembly of metal-organic cages is driven by the coordination between the metal ions and ligands, which requires less synthetic effort than covalent versions [17][18][19][20].Consequently, the space, shape, and hydrophobic environment of the cavity can be well controlled by simply modulating the properties of the metal centers and/or the organic ligands [21].This means that specific cavities could be well constructed for particular reactions [22,23].(ii) Catalytic sites or catalysts can be easily incorporated onto the cage framework for catalysis.The catalytic moieties can either be introduced onto the organic ligands or onto the metal centers.Thus, the number and type of the catalytic sites or catalysts integrated in the cavity can be well designed.In this context, a high local concentration of the catalytic active sites can be obtained, which could lead to the improvement of catalytic performance.Moreover, the synergistic interaction between the catalytic active sites and the confined cavity (such as the electron transfer between the active sites and the cavity, the stabilization of the active sites by the cavity, and so on) may also contribute to the catalytic activity or selectivity [24][25][26].(iii) Different and even incompatible catalysts or active sites could be spatially segregated by the cavities of the metal-organic cages [27].This property is extremely desirable for the development of high efficient tandem or cascade reactions, where two or more successive independent catalytic steps were performed in a one pot [28].Consequently, complicated and high value products could be efficiently prepared from simple starting materials with limited reaction steps, energy consumption, and waste [29].
In this mini-review, we focus on describing the organic reactions boosted by metalorganic cages, which are used as reaction vessels and also catalysts.We highlight the structure-property relationship between the cage and the reaction and the principles to boost organic reactions.We first summarize the methods for accelerating the organic reactions, followed by methods to improve the selectivity of the reactions, and finally we target tandem or cascade reactions, which have been extremely rare in supramolecular catalysis.Since there are many wonderful reviews about supramolecular catalysis [30][31][32][33][34], only selected examples of the above three aspects in the past ten years will be included.The synthesis of the metal-organic cages is outside the scope of this review.Readers who are interested in the self-assembly of the supramolecular cages and other related topics are referred to the excellent reviews [35][36][37][38].

Acceleration of Catalytic Activity by Metal-Organic Cages
Metal-organic cages are not only able to modulate the reactivity of the molecules by encapsulating them within their cavities, but are also capable of influencing the reactive states of the catalysts or the active sites in the framework of the cage.These characterizations of the cage can be used to accelerate the reaction rate.The strategies to accelerate organic reactions include (i) fixing the reagents in a favored reactive conformation in the cavity, (ii) significantly promoting the local concentration of the catalysts within the cavity or on the surface of the cavity, and (iii) stabilizing the active sites by the cage.
Improving the hydrolysis reactivity of amides by mechanically twisting them within the cavity of a molecular metal cage, without any chemical processes, was recently achieved by Fujita and co-workers [39].A T d -symmetric platinum molecular metal cage (Pt 6 L 4 (NO 3 ) 12 , MOC-1) containing electron-deficient panel ligands (2,4,6-tris(4-pyridyl)-1,3,5-triazine, L) was used as the host (Figure 1) [40].When an aromatic amide (1) and a large modulator guest phenanthrene were co-encapsulated within the host with a large hydrophobic cavity under the basic solution at 60 • C [41], an outstanding rate enhancement with k cage /k bulk = 14 was observed (k cage and k bulk are rate constants with and without the cage).When two aromatic amides were co-included, rate enhancement with decreased k cage /k bulk value range from about 3 to 6 was also observed.However, when only one aromatic amide was located with its latent trans-planar conformation, negligible enhanced reactivity of the amide was observed.The crystal structures of the inclusion complex, aromatic amide, a large modulator guest, and the molecular metal cage (palladium instead of platinum [42]) reveal that the aromatic amide adopts a twisted cis conformation.A 1:1 cis-trans dimer formation from two aromatic amides was also found in the inclusion complex of two aromatic amides.The authors concluded that the enhanced hydrolysis reactivity of amides may result from the favored cis-twisted conformation over the trans-planar one due to the inclusion of them within the molecular metal cage [43].
An unusual intermolecular [2 + 2] cycloaddition of acyclic alkenes such as chalcones, cinnamates, and benzylideneacetones to produce homo-and hetero-coupled syn-HH cyclobutanes with high diastereoselectivity and yields, which was catalyzed by a molecular metal cage-confined Ru photo-catalyst, was reported by Su and co-workers (Figure 2) [44].The (Pd 6 RuL 3 ) 8 (BF 4 ) 28 cage catalyst (MOC-2) was self-assembled from eight RuL 3 (PF 6 ) 2 (L = 2-(pyridin-3-yl)-1H-imidazo [4,5-f] [1,10]-phena-nthroline) visible-light photosensitizer and six Pd(BF 4 ) 2 , resulting in a truncated octahedron photoactive nanospace for dynamic guests inclusion and exchange [45].Mixing the cage catalyst with 6 eq of acyclic alkenes (2 and 3, for example chalcone and ethyl 4-bromocinnamate) in a DMSO-d 6 -D 2 O solution (1:3, v/v) resulted in a remarkable upfield shift of proton signals of substrates and splitting of host signals in the 1 HNMR spectra, revealing the efficiently capture of the substrates in the cavity of the cage.Visible-light irradiation (450 nm, N 2 atmosphere, 3 h, r.t.) leads to the formation of the cyclobutane products, as demonstrated by the disappearance of the resonances of substrate protons and the restoration of those of the cage catalyst.The cycloaddition reaction can even proceed very well in a 93% yield with at a very low photocatalyst loading of 0.08 mol%.To prove the catalytic efficiency of the cage catalyst, cycloaddition reactions between chalcone and methyl cinnamate as well as cycloaddition of methyl cinnamate were performed with the cage catalyst and the free photosensitizer RuL 3 (PF 6 ) 2 .A ~12-fold and a surprising 1.044 × 10 4 -fold rate enhancement was observed, respectively, in these two reactions.The box-like cavity of the cage as revealed by the crystal structure was believed to selectively arrange two substrates in one portal, and facilitated the following dimerization and dynamic processes of guest capture and delivery [46].Moreover, the cage catalyst also shows nice water-solubility due to the multiple nitrogen atoms in the ligands of RuL 3 (PF 6 ) 2 .The hydrophobic effect of the cage was considered to accelerate the transfer of the insoluble substrates and photoproducts out of the aqueous reaction media, thus helping the product to release from the host [47].
An unusual intermolecular [2 + 2] cycloaddition of acyclic alkenes such as chalcones, cinnamates, and benzylideneacetones to produce homo-and hetero-coupled syn-HH cyclobutanes with high diastereoselectivity and yields, which was catalyzed by a molec-

PEER REVIEW 4
of the insoluble substrates and photoproducts out of the aqueous reaction media, thus helping the product to release from the host [47].Increasing the local catalyst concentration to the mole level in a metal-organic cage, which is unfavorable for general homogenous transition-metal catalysts in solution, can significantly improve the activity and selectivity of catalyst [48].A metal-organic cage (Pd12L24(OTf)24, MOC-3) with tunable high concentration (0.05 M to 1.1 M) of gold(I) catalyst inside (Figure 3) was prepared by Joost N. H. Reek and co-workers through a self-assembly strategy using ditopic nitrogen ligands (L) with Ph3PAuCl unit [49].The hydroalkoxylation of the g-allenol (4) catalyzed by the metal-organic cage with a high local concentration of gold-chloride catalyst of 1.07 M selectively leads to the five-membered ring product with a high yield of 90% and a turnover numbers (TON) number of 1.86.The metal-organic cages with decreasing catalyst concentrations generally give lower yields of the product.In contrast, the free gold(I) complexes, including the AuCl, Ph3PAuCl, and Ph3PAuCl containing ditopic nitrogen ligands, are all inactive for the reaction under the same reaction conditions.Based on the UV/Vis spectra of the solution of the metal-organic cage, which show a broad shoulder for the d 10 -d 10 aurophilic interactions of the metal complex [50], the author proposed that the unexpected activity and selectivity of the metal-organic cage may be a result of the new formed active gold Increasing the local catalyst concentration to the mole level in a metal-organic cage, which is unfavorable for general homogenous transition-metal catalysts in solution, can significantly improve the activity and selectivity of catalyst [48].A metal-organic cage (Pd 12 L 24 (OTf) 24 , MOC-3) with tunable high concentration (0.05 M to 1.1 M) of gold(I) catalyst inside (Figure 3) was prepared by Joost N. H. Reek and co-workers through a self-assembly strategy using ditopic nitrogen ligands (L) with Ph 3 PAuCl unit [49].The hydroalkoxylation of the g-allenol (4) catalyzed by the metal-organic cage with a high local concentration of gold-chloride catalyst of 1.07 M selectively leads to the five-membered ring product with a high yield of 90% and a turnover numbers (TON) number of 1.86.The metal-organic cages with decreasing catalyst concentrations generally give lower yields of the product.In contrast, the free gold(I) complexes, including the AuCl, Ph 3 PAuCl, and Ph 3 PAuCl containing ditopic nitrogen ligands, are all inactive for the reaction under the same reaction conditions.Based on the UV/Vis spectra of the solution of the metal-organic cage, which show a broad shoulder for the d 10 -d 10 aurophilic interactions of the metal complex [50], the author proposed that the unexpected activity and selectivity of the metalorganic cage may be a result of the new formed active gold species (active cationic gold species due to chloride dissociation or a multinuclear (P-Au-Cl)n complex) inside the cage.It should be mentioned that the cage collapsed when a lot of ring product was formed.
They also developed a non-covalent anchoring strategy to improve the catalyst concentration within the molecular metal cage (Figure 4) [51].Guanidinium-binding motifs (L), which can form complexes with sulfonates and carboxylates via hydrogen bonding [52], were introduced into organic ligands of the cage (Pd12L24(OTf)48, MOC-4).The resulting molecular metal cage with 24 endohedral guanidinium-binding sites can strongly bind sulfonate-containing gold catalyst (TPPMSAuCl, TPPMS = triphenylphosphinomonosulfonate), leading to a high gold catalyst concentration in the cavity of the cage.Meanwhile, the guanidinium sites can also weakly bind carboxylate-functionalized substrates, thus locating them in the proximity of the gold catalysts.As a consequence, in the Au(I)-catalyzed cyclization reaction of acetylenic acid (5) to give enol lactone under basic conditions, the gold catalysts inside the molecular metal cage show a remarkable 40-fold rate enhancement compared to that of the free TPPMSAuCl.The authors believe that the cooperative binding, in which sulfonate-bearing catalyst bounding stronger than the carboxylates, and the pre-organization effect of the cage were the keys to the increased activity of the encapsulated sulfonate-containing Au(I) catalysts.They also developed a non-covalent anchoring strategy to improve the catalyst concentration within the molecular metal cage (Figure 4) [51].Guanidinium-binding motifs (L), which can form complexes with sulfonates and carboxylates via hydrogen bonding [52], were introduced into organic ligands of the cage (Pd 12 L 24 (OTf) 48 , MOC-4).The resulting molecular metal cage with 24 endohedral guanidinium-binding sites can strongly bind sulfonate-containing gold catalyst (TPPMSAuCl, TPPMS = triphenylphosphinomonosulfonate), leading to a high gold catalyst concentration in the cavity of the cage.Meanwhile, the guanidinium sites can also weakly bind carboxylate-functionalized substrates, thus locating them in the proximity of the gold catalysts.As a consequence, in the Au(I)-catalyzed cyclization reaction of acetylenic acid (5) to give enol lactone under basic conditions, the gold catalysts inside the molecular metal cage show a remarkable 40-fold rate enhancement compared to that of the free TPPMSAuCl.The authors believe that the cooperative binding, in which sulfonate-bearing catalyst bounding stronger than the carboxylates, and the preorganization effect of the cage were the keys to the increased activity of the encapsulated sulfonate-containing Au(I) catalysts.
The same supramolecular encapsulation strategy was also applied to prepare high local concentrations of electrochemical ruthenium catalyst (up to 0.54 M) [53].In water oxidation, the pre-organized Ru catalyst (Ru(bda)(PySO 3 TBA) 2 ) inside the cage (MOC-4) shows a high TOF value of 125 s −1 , which is two orders of magnitude (130 times) higher than that of the homogeneous ruthenium catalyst.The clear increase of k cat , which was determined by combining the Randles-Sevcik equation with kinetics plots, along with the increased catalyst concentration, indicated the reaction proceeded via a I2M mechanism, where the ruthenium-oxo intermediate was formed via a rate determining proton coupled electron transfer step.In the kinetic isotope effect (KIE, H 2 O/D 2 O) studies with different local catalyst concentrations (0-0.54M), relatively small KIE value (<1.5) was observed at low catalyst concentrations (0-0.27M), but the KIE increases up to 2.2 with higher catalyst concentration (0.27-0.54 M).These results further confirmed that the dinuclear radical coupling with typical diffusion limitations was facilitated in higher catalyst concentration and the rate-determining step (RDS) step was changed from the radical coupling step to the preceding proton coupled oxidation step by altering the local catalyst concentration [54].
Chemistry 2022, 4, FOR PEER REVIEW Figure 4. Cyclization reaction of acetylenic acid catalyzed by MOC-4 binding wi [51].Reaction additions: 5 (10.0 mM), Au + (0.5 mM), MOC-4 (0.125 mM), NEt3 (0 r.t.Reprinted/adapted with permission from Ref. [51].Copyright 2016 Springer N The same supramolecular encapsulation strategy was also applied t local concentrations of electrochemical ruthenium catalyst (up to 0.54 M oxidation, the pre-organized Ru catalyst (Ru(bda)(PySO3TBA)2) inside the shows a high TOF value of 125 s −1 , which is two orders of magnitude (13 than that of the homogeneous ruthenium catalyst.The clear increase of An impressively efficient catalysis of the Kemp elimination, the ring-opening reaction of benzisoxazole to give 2-cyanophenolate in the presence of a base, in the cavity of a metal coordination cage, was recently reported by Ward, Williams, and Hunter [7].The cubic Co 8 L 12 (BF 4 ) 16 (MOC-5) coordination cage catalyst (Figure 5), which contains eight Co 2+ ions at the vertices and sixteen hydroxyl groups on the external surface, features by its water solubility, hydrophobic interior cavity, and positive charge on the cage [55].These characteristics of the cage promoted the binding of benzisoxazole (6) in the cavity, the accumulation of hydroxide ions around the cage surface, and the release of the hydrophilic product when the reaction was carried out in a basic aqueous media [56].As a result, the cage catalyst showed pH independence of catalytic activity in the Kemp elimination of benzisoxazole, leading to a significant rate enhancement (2 × 10 5 -fold) for the cage-catalyzed reaction compared with the un-catalyzed one at pD of 8.5.The concentrated partially desolvated hydroxide ions around the cage surface were considered to be the critical factor for the high catalytic activity of the cage.This was confirmed by competition experiments where the observed rate of reaction was reduced to that of the background rate by adding a large excess of chloride ions (47 mM) into the reaction solution.In their continued study of this catalytic system, they discovered a new autocatalytic pathway catalyzed by the accumulated basic product 2-cyanophenolate at the cage surface [57].When chloride or fluoride inhibited the reaction to some extent, the product 2-cyanophenolate with a bigger affinity toward the cationic cage could displace them and acted as a base for the next reaction cycle.This autocatalytic reaction inspired them to develop an anion-controlled reaction for a cavity-bound guest.For instance, the Kemp elimination could optionally be accelerated by adding phenolates with different basicity.
Chemistry 2022, 4, FOR PEER REVIEW a metal coordination cage, was recently reported by Ward, Williams, and Hunter [7 The cubic Co8L12(BF4)16 (MOC-5) coordination cage catalyst (Figure 5), which contain eight Co 2+ ions at the vertices and sixteen hydroxyl groups on the external surface, fe tures by its water solubility, hydrophobic interior cavity, and positive charge on the cag [55].These characteristics of the cage promoted the binding of benzisoxazole (6) in th cavity, the accumulation of hydroxide ions around the cage surface, and the release the hydrophilic product when the reaction was carried out in a basic aqueous med [56].As a result, the cage catalyst showed pH independence of catalytic activity in th Kemp elimination of benzisoxazole, leading to a significant rate enhancement (2 10 5 -fold) for the cage-catalyzed reaction compared with the un-catalyzed one at pD 8.5.The concentrated partially desolvated hydroxide ions around the cage surface we considered to be the critical factor for the high catalytic activity of the cage.This wa confirmed by competition experiments where the observed rate of reaction was reduce to that of the background rate by adding a large excess of chloride ions (47 mM) into th reaction solution.In their continued study of this catalytic system, they discovered new autocatalytic pathway catalyzed by the accumulated basic produ 2-cyanophenolate at the cage surface [57].When chloride or fluoride inhibited the rea tion to some extent, the product 2-cyanophenolate with a bigger affinity toward the ca ionic cage could displace them and acted as a base for the next reaction cycle.This aut catalytic reaction inspired them to develop an anion-controlled reaction for a cav ty-bound guest.For instance, the Kemp elimination could optionally be accelerated b adding phenolates with different basicity.) MOC-7 with similar structure (same cavity in shape and size) but different charge properties (MOC-6 has five times more net charge than MOC-7) were synthesized as supports of the Co catalysts.The cage was mixed with CoCl 2 in N,N-dimethylformamide (DMF) at room temperature for 30 min.The encapsulated Co 2+ ions within the cage were then reduced by NaBH 4 .Disappearance of the adsorption band of Co 2+ from 600 to 700 nm accompanied by an immediate color change from green to dark brown were observed upon the addition of NaBH 4 .These findings clearly indicate the formation of Co nanoparticles (Co Ps@MOC-6 and Co NPs@MOC-7), which were characterized by Transmission Electron Microscopy (TEM), Energy Dispersive X-ray Spectroscopic (EDS), and X-ray Photoelectron Spectroscopic (XPS).The accumulated Co 2+ can be reduced into small Co nanoparticles (2.5 nm) without aggregation by NaBH 4 due to the stronger interaction between MOC-6 and Co ions.In contrast, large Co nanoparticles (100 nm) were obtained in the case of MOC-7 with fewer net charges.As a result, better catalytic performances in the hydrolysis of ammonium borane were observed for the catalyst Co Ps@MOC-6 (TOF, 90.1 min −1 ) than Co NPs@MOC-7 (TOF, 22.5 min −1 ).They also used the same strategy to synthesize an encapsulated small Ru nanoparticle catalyst (2.5 nm), which showed high activity in the methanolysis of ammonia borane (TOF, 304.4 min −1 ) [59].
In most of the reported metal-organic cages involving catalysis, the reactions were generally considered to take place inside the cavity.However, Jonathan R. Nitschke and co-workers reported an unusual mode of catalytic action where the reaction occurs on the surface of the cage [60].In their case, the cage was used to mediate electron transfer rather than to encapsulate the substrate (Figure 7).To achieve this goal, a redoxswitchable Zn II 4 L 6 metal-organic cage (MOC-8) decorated with NDI naphthalenediimide (NDI) moieties at its edges was synthesized.C60, a co-catalyst that can stabilize radicals, was then included inside the cage to enhance the catalytic activity [61].The resulting catalyst system (10 mol% cage, 20 mol% C60) showed moderate activity in the oxidative coupling reaction of tetraaryl borates (37%−81% yield).EPR experiments of the mixed cage, C60 and a tetraaryl borate(potassium tetrakis(4-chlorophenyl)borate, 8) showed sharp signals at 2.0016 and 2.0013, which was attributed to C60* radical species.Based on this, the authors concluded that C60 may serve as a stabilizing agent by taking up some radical spin density from the cage, which was reduced by the borates.In most of the reported metal-organic cages involving catalysis, the reactions generally considered to take place inside the cavity.However, Jonathan R. Nitschke co-workers reported an unusual mode of catalytic action where the reaction occur the surface of the cage [60].In their case, the cage was used to mediate electron tra rather than to encapsulate the substrate (Figure 7).To achieve this goal, a dox-switchable Zn II 4L6 metal-organic cage (MOC-8) decorated with NDI naph lenediimide (NDI) moieties at its edges was synthesized.C60, a co-catalyst that can bilize radicals, was then included inside the cage to enhance the catalytic activity The resulting catalyst system (10 mol% cage, 20 mol% C60) showed moderate activi the oxidative coupling reaction of tetraaryl borates (37%−81% yield).EPR experimen the mixed cage, C60 and a tetraaryl borate(potassium tetrakis(4-chlorophenyl)bora showed sharp signals at 2.0016 and 2.0013, which was attributed to C60* radical spe Based on this, the authors concluded that C60 may serve as a stabilizing agent by ta up some radical spin density from the cage, which was reduced by the borates.

Improvement of Selectivity by Metal-Organic Cages
Improving the selectivity toward the desired product is one of the most important goals for the design of a catalyst.Common synthetic catalysts usually require complex ligands and specific reaction conditions to render selectivity.In contrast, metal-organic

Improvement of Selectivity by Metal-Organic Cages
Improving the selectivity toward the desired product is one of the most important goals for the design of a catalyst.Common synthetic catalysts usually require complex ligands and specific reaction conditions to render selectivity.In contrast, metal-organic cage catalysts readily self-assemble from simple components, offering tailored confined spaces to enable remarkable product selectivity.Strategies to improve the selectivity of a supramolecular cage catalyst include (i) taking advantage of the constriction imposed by the size and shape of the cavity to limit the type of the products, (ii) using the cage as a stoichiometric supramolecular "protecting group" to selectively hide functional groups at multiple sites of the substrate, and (iii) encapsulating a catalyst inside the cage to provide a secondary environment for direct chemo and regioselectivity.
Instead of size discrimination between substrates [62], partial confinement of a single molecule can selectively protect functional groups at multiple sites of the molecule, thus leading to size-selective reactions.Very recently, Raymond, Bergman, and Toste reported a selective catalytic hydrogenation of a polyene group over other unsaturation bonds in the same substrate by encapsulating a rhodium complex hydrogenation catalyst [Rh][BH4] ([Rh] = [(DMPE)Rh(COD)], DMPE = 1,2-bis(dimethylphosphino)ethane) into a tetrahedron metal-organic cage (Na 12 Ga 4 L 6 , MOC-9, Figure 8) [63,64].Polyenol containing three points of unsaturation (10) was selectively hydrogenated into a monohydrogenated product with a yield of 74% in the presence of the supramolecular catalyst, while was converted into the fully hydrogenated alcohol with the free rhodium complex.It is proposed that only a portion of the substrate (outermost unsaturation bonds) can enter into the cavity of the cage and then be catalyzed by the encapsulated rhodium complex.This partial confinement strategy provides an unusual and beautiful supramolecular catalyst-directed regioselectivity as opposed to common directing-group selectivity.In contrast to small molecule synthesis, metal-organic cage catalysts breadth of reactivity ranging from small molecules to complex biomol proteins remain rare.By emulating the catalytic action of the pyridine-ba ketoreductase enzymes (KREDs) [65], Raymond, Bergman, and Toste rep organic cage host-mediated pyridine-borane reduction with a wide subst including enones, ketones, aldehydes, oximes, hydrazones, and imines In contrast to small molecule synthesis, metal-organic cage catalysts having a wide breadth of reactivity ranging from small molecules to complex biomolecules such as proteins remain rare.By emulating the catalytic action of the pyridine-based cofactor of ketoreductase enzymes (KREDs) [65], Raymond, Bergman, and Toste reported a metal-organic cage host-mediated pyridine-borane reduction with a wide substrate (11) scope including enones, ketones, aldehydes, oximes, hydrazones, and imines (Figure 9) [66].Pyridineborane (12), a mild reducing agent, was encapsulated into a highly anionic [K 2 Ga 4 L 6 ] (MOC-10) metal-organic cage [67].The host systems showed selectivity towards ketones.In the reduction of α, β-unsaturated ketones with reducible C=C and C=O bonds, only the C=O bond was selectively reduced under the metal-organic cage-mediated reduction conditions.In contrast, when the reaction was run in the presence of a strongly binding inhibitor of the catalysts (NEt 4+ ), almost only the corresponding saturated ketone products were consistently observed as the products.It was proposed by the authors that the co-encapsulation of the enone substrate with pyridine−borane and the stabilization of the protonated intermediate of the carbonyl group may accelerate the 1,2-reduction pathway over the 1,4-pathway [68].Partial encapsulation was then employed for the host-catalyzed reduction to more complex substrates.Hydrophilic substrates like Lysine, amino acid peptide, and human insulin, which contain both anionic carboxylate groups and amine groups, were expected to be partially encapsulated by the catalyst due to the unfavorable coulombic interactions between the carboxylate group and the host.In turn, selectively reductive alkylation of the ε-terminus of the substrates with fluorobenzaldehydes was achieved in the presence of the supramolecular catalysts.
Pyridine-borane (12), a mild reducing agent, was encapsulated in [K2Ga4L6] (MOC-10) metal-organic cage [67].The host systems sh wards ketones.In the reduction of α, β-unsaturated ketones with C=O bonds, only the C=O bond was selectively reduced under cage-mediated reduction conditions.In contrast, when the reaction w ence of a strongly binding inhibitor of the catalysts (NEt 4+ ), almost o ing saturated ketone products were consistently observed as the pr posed by the authors that the co-encapsulation of the enone su dine−borane and the stabilization of the protonated intermediate of may accelerate the 1,2-reduction pathway over the 1,4-pathway [68 tion was then employed for the host-catalyzed reduction to more Hydrophilic substrates like Lysine, amino acid peptide, and human tain both anionic carboxylate groups and amine groups, were expe encapsulated by the catalyst due to the unfavorable coulombic inter carboxylate group and the host.In turn, selectively reductive ε-terminus of the substrates with fluorobenzaldehydes was achieved the supramolecular catalysts.Cui, Liu, and co-workers reported a chiral cage featuring chiral dihydroxyl groups by installing a chiral molecular catalyst into a supramolecular metal-organic cage [69].The chiral cage catalyst exhibited improved enantioselectivity in the asymmetric conjugate addition of styrylboronic acids to α,β-enones and the asymmetric addition diethylzinc to aldehydes compared to the free chiral molecular catalyst.Different from the free chiral molecular one, the cage catalyst showed substituent size-dependent activity and a much higher binding capacity towards the substrate in the asymmetric conjugate addition reaction between 4-hydroxychalcone and styrylboronic acids.The authors, therefore, proposed that the enhancement of stereoselectivity of the cage catalyst was due to the steric hindrance and confinement effect of the supramolecular porous cage.Moreover, the same group further developed a cage extension strategy to construct heterogeneous chiral porous catalysts through intracage hydrogen-bonds and hydrophobic interactions (Figure 10) [70].A spinol-based metal-organic cage (MOC-11) with phosphoric and carboxylic groups capable of forming strong hydrogen bonds to another was first prepared using enantiopure 1,1 -spirobiindane-7,7 -phosphoric acid and Ni 4 -p-tertbutylsulfonylcalix[4]arene clusters as the ligands.The hierarchical assembly of the cage leaded to the formation of a porous hydrogen bonded organic framework (HOF) [71].The resulting chiral HOF acted as a recyclable Brønsted acid catalyst in the asymmetric [3+2] coupling of indoles (13) with quinone monoimine (14) and Friedel-Crafts alkylations of indole with aryl aldimines.Up to a 92% isolated yield and a 99.9% ee for the benzofuroindoline derivatives and a 92% isolated yield and a 91% ee for 3-indolylmethanamine derivatives can be obtained with extremely low loading of the HOF catalyst (0.04 mol%).These results surpassed those of the homogeneous counterparts including the acidic ligand or the chiral free metal-organic cage.Moreover, the solid porous catalyst can maintain its efficiency and enantioselectivity after 10 runs.
tion reaction between 4-hydroxychalcone and styrylboronic acids.The authors, therefore, proposed that the enhancement of stereoselectivity of the cage catalyst was due to the steric hindrance and confinement effect of the supramolecular porous cage.Moreover, the same group further developed a cage extension strategy to construct heterogeneous chiral porous catalysts through intracage hydrogen-bonds and hydrophobic interactions (Figure 10) [70].A spinol-based metal-organic cage (MOC-11) with phosphoric and carboxylic groups capable of forming strong hydrogen bonds to another was first prepared using enantiopure 1,1′-spirobiindane-7,7′-phosphoric acid and Ni4-p-tertbutylsulfonylcalix[4]arene clusters as the ligands.The hierarchical assembly of the cage leaded to the formation of a porous hydrogen bonded organic framework (HOF) [71].The resulting chiral HOF acted as a recyclable Brønsted acid catalyst in the asymmetric [3+2] coupling of indoles (13) with quinone monoimine (14) and Friedel-Crafts alkylations of indole with aryl aldimines.Up to a 92% isolated yield and a 99.9% ee for the benzofuroindoline derivatives and a 92% isolated yield and a 91% ee for 3-indolylmethanamine derivatives can be obtained with extremely low loading of the HOF catalyst (0.04 mol%).These results surpassed those of the homogeneous counterparts including the acidic ligand or the chiral free metal-organic cage.Moreover, the solid porous catalyst can maintain its efficiency and enantioselectivity after 10 runs.Very recently, Fujita and coworkers reported an unusual demethylenation reaction of cyclopropane with high chemo-selectivity towards the alkene product boosted by a Pd 6 L 4 metal-organic cage (MOC-12) [72], which showed improvement in photo-oxidation reactions of inert alkanes and photohydrations of alkynes [73,74].Cyclopropane compounds such as drospirenone, trans-2-phenyl-1cyclopropanecarboxylic acid, and chrysanthemic acid underwent demethylenation reaction to give corresponding alkene products and formaldehyde or acetone inside the photoactive coordination cage under UV light irradiation (Figure 11).In contrast, the reference cage without the triazine-cored ligand was inactive in the demethylenation reaction of drospirenone.After the photoreaction, nitrite anion (NO 2 − ) was detected by the Griess method.A photoinduced electron transfer from the guest to the cage was proposed for the demethylenation reaction.The electron-deficient triazine-based ligand acted as an electron acceptor and accepted an electron from the bound guest under UV irradiation, generating a radical anion centered on the cage and a radical cation centered on the guest.Nucleophilic attack of a nitrate anion of the radical cation and successive fragmentation leaded to the demethylenated product and a fragment nitrite radical.The catalytic cycle was completed by the one electron back to the nitrate radical from the cage.The same cage also showed site-selective electrophilic addition of linear diterpenoids with nitrate anions [75].
and accepted an electron from the bound guest under UV irradiation, generating a radi cal anion centered on the cage and a radical cation centered on the guest.Nucleophilic attack of a nitrate anion of the radical cation and successive fragmentation leaded to the demethylenated product and a fragment nitrite radical.The catalytic cycle was com pleted by the one electron back to the nitrate radical from the cage.The same cage also showed site-selective electrophilic addition of linear diterpenoids with nitrate anions [75].Oxidative coupling of 2-naphthol in bulk solution usually leads to 1,1-bis(2-naphthol), while the 1,4-coupling product was only detected as a minor by product.In contrast to these results in bulk solution, Su and co-workers recently report ed an unusual photoinduced 1,4-coupling of naphthols and derivatives rather than the normal 1,1-coupling to form asymmetric biaryl compounds with moderate regio-and enantioselectivity in the above-mentioned MOC-2 [76].The robust enantiopure meta coordination cage containing photoredox-active and stereogenic Ru II metalloligands no only provided a confined chiral coordination space, but also photoactive active sites thus serving as dual-functional photoredox-and stereochemically active nanoreactors Under light irradiation, the oxidative coupling of 3-bromo-2-naphthol (16) in the pres ence of 10 mol% of the chiral metal coordination cage exclusively give the 1,4-coupling biaryl product with a yield of 32% and an ee value of 34% (Figure 12).On the contrary, a yield of only 8% and an ee value of 10% of the product were obtained by 80 mol% of the Oxidative coupling of 2-naphthol in bulk solution usually leads to 1,1-bis(2-naphthol), while the 1,4-coupling product was only detected as a minor byproduct.In contrast to these results in bulk solution, Su and co-workers recently reported an unusual photoinduced 1,4-coupling of naphthols and derivatives rather than the normal 1,1-coupling to form asymmetric biaryl compounds with moderate regio-and enantioselectivity in the above-mentioned MOC-2 [76].The robust enantiopure metal coordination cage containing photoredox-active and stereogenic Ru II metalloligands not only provided a confined chiral coordination space, but also photoactive active sites, thus serving as dual-functional photoredox-and stereochemically active nanoreactors.Under light irradiation, the oxidative coupling of 3-bromo-2-naphthol (16) in the presence of 10 mol% of the chiral metal coordination cage exclusively give the 1,4-coupling biaryl product with a yield of 32% and an ee value of 34% (Figure 12).On the contrary, a yield of only 8% and an ee value of 10% of the product were obtained by 80 mol% of the chiral RuL 3 metalloligand complexes.In the control experiments using a racemic metal coordination cage and FeCl 3 or CuCl 2 as the coupling agent, both the 1,4-coupling and the normal 1,1-coupling reaction were prevented.The authors proposed that the confined chiral space of the metal coordination cage, which preferentially encapsulated the substrate between the two parallel walls of the homochiral configuration of the Ru centers, was attributed to the enhanced regio-and enantioselectivity of the naphthol coupling [77].
A second coordination-sphere tuning strategy to control the selectivity of the metal coordination cage catalysts was developed by Reek and co-workers [78].One of the examples was the regio-and enantioselective hydroformylation of styrenes (17) catalyzed by the encapsulated chiral Rh catalyst in a metal coordination cage (Figure 13, MOC-13) [79].The self-assembly of the macrocyclic dipalladium complexes and tetracarboxylate zinc porphyrins [80], and further inclusion of a chiral phosphoramidite led to the formation of the metal coordination cage with a chiral ligand inside its cavity.In the presence of this special supramolecular ligand, Rh(acac)(CO) 2 provided high chemo-and stereoselectivities towards R-2-phenylpropanal, with an ee value of 74% and a TON of 797.In contrast, phosphoramidite ligand ((S)-α) (1% ee, TON of 342) alone or the Rh-complex of α(Zn-TPP) 2 (9% ee, TON of 363) exhibited extremely low selectivity and activity.Molecular modeling studies indicated that the substrate could approach the Rh center with multiple orientations for the Rh-complex of phosphoramidite ligand, while most of the coordination modes of styrene to the Rh center were prevented in the cage catalyst.This was an elegant example of using space control of the chirality around the catalyst center to achieve enantioselectivity that cannot be attained in bulk.
Chemistry 2022, 4, FOR PEER REVIEW 14 chiral RuL3 metalloligand complexes.In the control experiments using a racemic metal coordination cage and FeCl3 or CuCl2 as the coupling agent, both the 1,4-coupling and the normal 1,1-coupling reaction were prevented.The authors proposed that the confined chiral space of the metal coordination cage, which preferentially encapsulated the substrate between the two parallel walls of the homochiral configuration of the Ru centers, was attributed to the enhanced regio-and enantioselectivity of the naphthol coupling [77].A second coordination-sphere tuning strategy to control the selectivity of the metal coordination cage catalysts was developed by Reek and co-workers [78].One of the examples was the regio-and enantioselective hydroformylation of styrenes ( 17) catalyzed by the encapsulated chiral Rh catalyst in a metal coordination cage (Figure 13, MOC-13) [79].The self-assembly of the macrocyclic dipalladium complexes and tetracarboxylate zinc porphyrins [80], and further inclusion of a chiral phosphoramidite led to the formation of the metal coordination cage with a chiral ligand inside its cavity.In the presence of this special supramolecular ligand, Rh(acac)(CO)2 provided high chemo-and stereoselectivities towards R-2-phenylpropanal, with an ee value of 74% and a TON of 797.In contrast, phosphoramidite ligand ((S)-α) (1% ee, TON of 342) alone or the Rh-complex of α(Zn-TPP)2 (9% ee, TON of 363) exhibited extremely low selectivity and activity.Molecular modeling studies indicated that the substrate could approach the Rh center with multiple orientations for the Rh-complex of phosphoramidite ligand, while most of the coordination modes of styrene to the Rh center were prevented in the cage catalyst.This was an elegant example of using space control of the chirality around the catalyst center to achieve enantioselectivity that cannot be attained in bulk.An enantiopure metal coordination cage was synthesized by Raymond, Bergman, and Toste using a terephthalamide-based ligand with chiral amide groups (Figure 14) [81].Since this supramolecular cage (K4Ga4L6, MOC-14) was prepared without the use of any templates or cationic species, it was applied as a chiral catalyst for the transfor- An enantiopure metal coordination cage was synthesized by Raymond, Bergman, and Toste using a terephthalamide-based ligand with chiral amide groups (Figure 14) [81].Since this supramolecular cage (K 4 Ga 4 L 6 , MOC-14) was prepared without the use of any templates or cationic species, it was applied as a chiral catalyst for the transformation of neutral compounds, for example, the Prins cyclizations of monoterpene derivatives.The chiral metal coordination cage showed high conversion in the Prins cyclizations of citronellal (18) and its homologues (yield up to 94%), and gave an enantiomer cycle product with an ee value up to 65%.The effects of the host spacer on product selectivity were investigated by using an analogous pyrene-core metal coordination cage with a large cavity [64].Product enantioselectivity was significantly decreased (33% ee) with the pyrenecore host.These results clearly demonstrated that the host-mediated enantioinduction decreases with host cavity size.[81].Reaction additions: 18 (0.04 mmol), MOC-14 (2.5 mol%), CD 3 OD/D 2 O (0.5 mL), pD = 8, r.t.Reprinted/adapted with permission from Ref. [81].Copyright 2013 American Chemical Society.

Tandem or Cascade Reactions Catalyzed by Metal-Organic Cages
Tandem or cascade reactions, which combine two or more incompatible successive catalytic steps in one pot, are highly attractive in terms of improving catalytic efficiencies, and reducing energy consumption and waste, as well as rendering more complex materials from a simple set of starting materials.However, such tandem or cascade reactions have been less developed due to the necessary compatibility of multiple catalysts or catalytic active sites, reagents, and intermediates generated during the reaction.The unique structure of metal-organic cages makes them desirable platforms for tandem or cascade reactions by (i) perfectly spatially segregating incompatible catalysts inside or outside their cavities, or (ii) integrating different catalytically active sites in the framework of a single cage, or (iii) constructing suitable reaction space to avoid interference between the different catalytic cycles.However, the construction of such distinctive metal-organic cages has proven to be challenging, and only small steps have been taken in this field.
In 2013, Raymond, Bergman, and Toste illustrated an intriguing example of hydrolysiscyclization tandem reactions catalyzed by a spatially separated natural enzyme (esterases, or lipases) and gold(I) complexes (Me 3 P + Au − ) encapsulated in MOC-10 (Figure 15a) [82].The supramolecular encapsulation of organometallic complexes prevented the detrimental interactions between the complexes and the enzyme in the solution, therefore preserving the activity of both.Thus, the sequential transformation was performed smoothly starting from the enzymatical hydrolysis of allenic acetate ( 19) by esterase to form alcohol, which was then cyclized into substituted tetrahydrofuran by (Me 3 P + Au − ) encapsulated in MOC-10 with excellent selectivity (100%) and yield (100%).Only 62% yield of the product was formed when the free Au catalyst was used instead of the bound one, and no such tetrahydrofuran product was obtained in the absence of the esterase or the encapsulated gold catalysts.This strategy was also employed in the encapsulated ruthenium ((Me 3 P)CpRu(NCMe) 2 ) − catalyzed isomerization of 1-propenol (20) to give propanal and the sequential reduction of propanal to propanol via ADH (Figure 15b).Very recently, Fujita and co-workers demonstrated an interesting one-pot tandem oxidation-Diels−Alder reaction using a site-isolation strategy (Figure 16) [83].The incompatible oxidation catalyst TEMPO [84] and the Diels−Alder catalyst chiral amine [85] were physically separated within the cavity of two self-assembled metal-organic cages M12L24 (MOC-15, MOC-16), thus affording allylic oxidation of alcohol ( 21) followed by a Very recently, Fujita and co-workers demonstrated an interesting one-pot tandem oxidation-Diels−Alder reaction using a site-isolation strategy (Figure 16) [83].The incompatible oxidation catalyst TEMPO [84] and the Diels−Alder catalyst chiral amine [85] were physically separated within the cavity of two self-assembled metal-organic cages M 12 L 24 (MOC-15, MOC-16), thus affording allylic oxidation of alcohol (21) followed by a stereoselective Diels−Alder cyclization.The cyclization product with four adjacent stereogenic centers was obtained in a moderate yield of 56% and a high ee value of 93% only in the presence of both two metal-organic cages, whereas in the absence of any of the metal-organic cages only the oxidation product formed.The M 12 L 24 metal-organic cages, which have an inner space sterically insulated by a rigid self-assembled shell with numerous apertures allowing for small molecular permeability [86], was the key to this site-isolation strategy.Isolating active sites within the metal-organic cages and the confined porous materials to protect the metal-organic cages is not only a novel way to improve the stability of metal-organic assemblies, but also a powerful strategy for constructing heterogeneous catalysts for one-pot sequential reactions.Very recently, we demonstrated the first confinement self-assembly of TEMPO-containing metal-organic cages (MOC-17, Figure 17) within amino-functionalized mesoporous carbon (FDU-ED) [87,88].The resulting hybrid materials served as a recyclable heterogeneous catalyst (MOC-17@FDU-ED) for the one-pot alcohol oxidation (22) and Knoevenagel condensation sequential reactions.Isolating active sites within the metal-organic cages and the confined porous materials to protect the metal-organic cages is not only a novel way to improve the stability of metalorganic assemblies, but also a powerful strategy for constructing heterogeneous catalysts for one-pot sequential reactions.Very recently, we demonstrated the first confinement self-assembly of TEMPO-containing metal-organic cages (MOC-17, Figure 17) within amino-functionalized mesoporous carbon (FDU-ED) [87,88].The resulting hybrid materials served as a recyclable heterogeneous catalyst (MOC-17@FDU-ED) for the one-pot alcohol oxidation (22) and Knoevenagel condensation sequential reactions.

OR PEER REVIEW 18
The bifunctional catalyst MOC-17@FDU-ED exhibited enhanced catalytic activities (TOF, 261.18 min −1 ) compared to those of the simple mixing metal-organic cage and mesoporous matrix catalysts (TOF, 54.54 min −1 ).Moreover, MOC-17@FDU-ED maintained its catalytic performances after 5 runs.However, due to the leaching problem, the activities of the mixed catalysts (ligand L+FDU-ED and cage MOC-17+FDU-ED) dramatically decreased from the second run.The improved activities and recyclability of MOC-17@FDU-ED could be attributed to the high dispersion catalytically active sites inside the composites MOC-17@FDU-ED.
The bifunctional catalyst MOC-17@FDU-ED exhibited enhanced catalytic activities (TOF, 261.18 min −1 ) compared to those of the simple mixing metal-organic cage and mesoporous matrix catalysts (TOF, 54.54 min −1 ).Moreover, MOC-17@FDU-ED maintained its catalytic performances after 5 runs.However, due to the leaching problem, the activities of the mixed catalysts (ligand L+FDU-ED and cage MOC-17+FDU-ED) dramatically decreased from the second run.The improved activities and recyclability of MOC-17@FDU-ED could be attributed to the high dispersion catalytically active sites inside the composites MOC-17@FDU-ED.
A single coordination cage containing multiple catalytically active sites can also promote one-pot sequential reactions [89].A novel tetrahedral coordination cage (MOC-18, Figure 18) bearing three Mn(salen) and three Cr(salen) active species was constructed by Cui and coworkers using six dicarboxylate ligands derived-from enantiopure Mn(salen) and Cr(salen) as linear linkers and four Cp 3 Zr 3 clusters as three-connected vertices.The resulting supramolecular catalyst shows high activity (62-82%, yield) and selectivity (91-99.9%ee) for the sequential asymmetric 2,2-dimethyl-2H-chromene (DMCH, 23) epoxidation/epoxide ring-opening reactions with TMSN 3 or anilines.Reaction kinetics studies indicated that the cage catalyst remained active at a very low loading (0.005-0.01 mol%); however, the mixture of Mn(salen) and Cr(salen) catalysts were totally inactive [90].It was proposed that the improved activity and enantioselectivity of the cage catalyst over the free catalysts was due to the stabilization of catalytically active metallosalen units and concentration of reactants within the cavity.It is also worth noting that this homogeneous cage catalyst can be easily recovered as precipitate by adding diethyl ether into the solution after the reaction.Its structure and the catalytic activity were maintained after at least five runs with slight deterioration in enantioselectivity.Relay multicatalysis is capable of transforming simple molecules into more complex and high-value products [91].The key to this system is to avoid the possible interference between the different catalytic cycles as well as the tolerance between different substrates and intermediates.Nitschke and co-workers demonstrated a novel interference-free "assembly line" (Figure 19) capable of converting small molecules furan (24) into a chiral hydroxylactone, in which an in situ assembled metal−organic cage plays a vital role [92].The precursors (i.e., 2-formylpyridine, iron(II) ions, and 4,4′-diaminobiphenyl-2,2′-disulfonic acid) of the metal−organic cage (MOC-19) were totally compatible with the substrates and other catalysts (nitromethane, dioxygen, methylene blue, and L-proline) in the present systems.The in situ assembled metal−organic cage selectively converted the high-energy endoperoxide intermediate generated from the hetero-Diels−Alder cycloaddition of oxygen with furan into fumaraldehydic acid, which is then catalyzed by the L-proline into the final hydroxyfuranone product.Control experiments in the absence of the metal−organic cage led to nonselective oxidation products and trace of the chiral hydroxylactone product, indicating the key role of the metal−organic cage.Relay multicatalysis is capable of transforming simple molecules into more complex and high-value products [91].The key to this system is to avoid the possible interference between the different catalytic cycles as well as the tolerance between different substrates and intermediates.Nitschke and co-workers demonstrated a novel interference-free "assembly line" (Figure 19) capable of converting small molecules furan (24) into a chiral hydroxylactone, in which an in situ assembled metal−organic cage plays a vital role [92].The precursors (i.e., 2-formylpyridine, iron(II) ions, and 4,4 -diaminobiphenyl-2,2 -disulfonic acid) of the metal−organic cage (MOC-19) were totally compatible with the substrates and other catalysts (nitromethane, dioxygen, methylene blue, and L-proline) in the present systems.The in situ assembled metal−organic cage selectively converted the high-energy endoperoxide intermediate generated from the hetero-Diels−Alder cycloaddition of oxygen with furan into fumaraldehydic acid, which is then catalyzed by the L-proline into the final hydroxyfuranone product.Control experiments in the absence of the metal−organic cage led to nonselective oxidation products and trace of the chiral hydroxylactone product, indicating the key role of the metal−organic cage.

R PEER REVIEW 20
Besides the sequential multiple-steps transformation, an interesting tandem acetal hydrolysis/cage-to-cage interconversion was recently reported by Hooley and co-workers (Figure 20) [93].Self-assembly of a carefully designed carboxylic acid-containing ligand into a Fe 4 L 6 iminopyridine cage leaded to a tetrahedral metal−organic cage with internalized acid groups (MOC-20), which worked as an effective tandem catalyst for the hydrolysis of aromatic acetals (25) and cage-to-cage interconversion of self-assembled M 2 L 3 helicates (MOC-Br, MOC-Py) [94].The cage catalyst showed up to a 1000-fold rate enhancement of acetal solvolysis compared with the free carboxylic acid ligand.The rate enhancement was demonstrated by NMR analysis.There was a strong host-guest affinity between the cage catalyst and the aromatic acetals, while no binding was observed between the free acid ligand and the aromatic acetals.Moreover, the acid cage catalyst was tolerant of the M 2 L 3 helicates, which are sensitive to acid, as well as carboxylate or chloride ions.Therefore, the deprotection of the 2-formylpyridine acetal and the followed helicate aldehyde exchange can proceed smoothly in the presence of the acid cage catalyst, while it cannot be performed using a stronger acid such as CF 3 CO 2 H.Besides the sequential multiple-steps transformation, an interesting tandem acetal hydrolysis/cage-to-cage interconversion was recently reported by Hooley and co-workers (Figure 20) [93].Self-assembly of a carefully designed carboxylic acid-containing ligand into a Fe4L6 iminopyridine cage leaded to a tetrahedral metal−organic cage with internalized acid groups (MOC-20), which worked as an effective tandem catalyst for the hydrolysis of aromatic acetals (25) and cage-to-cage interconversion of self-assembled M2L3 helicates (MOC-Br, MOC-Py) [94].The cage catalyst showed up to a 1000-fold rate enhancement of acetal solvolysis compared with the free carboxylic acid ligand.The rate enhancement was demonstrated by NMR analysis.There was a strong host-guest affinity between the cage catalyst and the aromatic acetals, while no binding was observed between the free acid ligand and the aromatic acetals.Moreover, Very recently, the same catalyst was applied to the multi-step Pictet-Spengler cyclization of tryptophols and various aldehyde derivatives (Figure 21) [95].The flexible endohedrally acid-functionalized metal−organic cage showed good activities in the Pictet-Spengler cyclization of tryptophol (26) and serials of benzaldehyde dimethyl acetal (27) with a good yield of cycle product (80-87%), while the unfunctionalized cage and the carboxylic acid ligand were both not capable of catalyzing the reaction.UV/vis absorbance titration experiments indicated that the acid functionalized metal−organic cage MOC-20 with a large cavity can bind the products, intermediates, and both substrates with suitable size at the same time.Further in situ monitoring of the reaction by 1 H NMR spectra provided evidence for the formation of an acetal intermediate and the final cycle product.The binding and activation of two or more different substrates in a single active site of the same cage were proposed for the high activity of the acid functionalized supramolecular catalyst.Very recently, the same catalyst was applied to the multi-step Pictet-Spengler cyclization of tryptophols and various aldehyde derivatives (Figure 21) [95].The flexible endohedrally acid-functionalized metal−organic cage showed good activities in the Pictet-Spengler cyclization of tryptophol (26) and serials of benzaldehyde dimethyl acetal (27) with a good yield of cycle product (80-87%), while the unfunctionalized cage and the carboxylic acid ligand were both not capable of catalyzing the reaction.UV/vis absorbance titration experiments indicated that the acid functionalized metal−organic cage MOC-20 with a large cavity can bind the products, intermediates, and both substrates with suitable size at the same time.Further in situ monitoring of the reaction by 1 H NMR spectra provided evidence for the formation of an acetal intermediate and the final cycle product.The binding and activation of two or more different substrates in a single active site of the same cage were proposed for the high activity of the acid functionalized supramolecular catalyst.

Conclusions
This paper provides and overview of several recent examples i breadth and attractive characteristics of metal-organic cages in boosting tions.Different metal-organic cages that offer tailored confined spaces hav to be able to accelerate reactions and enhance or alter product selectiv promote tandem or cascade reactions.The unique cavity structures of

Conclusions
This paper provides and overview of several recent examples illustrating the breadth and attractive characteristics of metal-organic cages in boosting organic reactions.Different metal-organic cages that offer tailored confined spaces have been shown to be able to accelerate reactions and enhance or alter product selectivity, as well as promote tandem or cascade reactions.The unique cavity structures of metal-organic cages and the various functional groups provide many advantages for catalysis.Regarding rate accelerations, metal-organic cages work by fixing the reagents in a favored reactive conformation, or promoting the local concentration of the catalysts within the cavity, or stabilizing the active sites by the cage.For enhancement or alteration of selectivity, metal-organic cages offer confinement effects to limit the type of the products, or selectively hide functional groups at multiple sites of the substrate, or encapsulating a catalyst inside the cage to provide a secondary environment for directing chemo and regioselectivity.Chiral metalorganic cages have also been developed as catalysts in asymmetric synthesis for achieving enantioselectivity.Metal-organic cages have also emerged as a powerful platform for engineering multifunctional catalysts for tandem or cascade reactions by perfectly spatially segregating incompatible catalysts inside or outside their cavities, or integrating different catalytically active sites in the framework of a single cage, or constructing suitable reaction space to avoid interference between the different catalytic cycles.It should be noted that the use of MOCs to achieve synergistic catalysis and promote tandem reactions is relatively rare compared with other reactions and should be explored more in the future.
Although impressive progress has been made in the application of metal-organic cages for catalysis, tuning reactivity within well-defined metal-organic cages is still in its infancy, and some challenges and limitations still need to be surmounted.For example, the range of reaction types is narrow, limiting their use in practical organic synthesis.Moreover, accurate and selective recognition and anchoring of specific molecules remains challenging.To address this, there is much to explore for the further development of novel classes of metal-organic cages with targeted functions.Furthermore, the balance between stability and activity should be emphasized and investigated, due to the relatively low stabilities of supramolecular metal-organic cages systems during reaction processes.There is no doubt that metal-organic cages have an extremely bright future in catalysis applications, and we hope that this minireview offer researchers general guidance and insight into the rational design of sophisticated metal-organic cages containers to manipulate chemical reactivity within confined nanospaces.