Anniversary of the Tröger’s Base Molecule: Synthesis and

MICROREVIEW DOI: 10.1002/ejoc.201201249

The 125th Anniversary of the Tröger’s Base Molecule: Synthesis and Applications of Tröger’s Base Analogues Ögmundur Vidar Rúnarsson,[a] Josep Artacho,[a] and Kenneth Wärnmark*[a] Dedicated to the memory of Julius Tröger on the 150th anniverary of his birth and to Craig S. Wilcox for reopening the field of Tröger’s base chemistry

Keywords: Tröger’s base / Supramolecular chemistry / Synthetic methods / Materials science / Nitrogen heterocycles In 2012 we celebrate the 125th anniversary of the unique molecule Tröger’s base (TB), first synthesized by Julius Tröger in 1887. Being a V-shaped C2-symmetric chiral molecule, it possesses many interesting features. The TB field was reopened in 1985, when Craig S. Wilcox published the first crystallographic study of TB and described the synthesis and potential applications of TB analogues in supramolecular chemistry and in ligand design. This led to increasing interest in the development of synthetic methodology for TB analogues, initially for applications in the field of molecular recognition. In this review we give a short historical overview of TB and its chemical properties. In addition, we cover the fast progress in the development of synthetic methodologies to synthesize TB analogues that has taken place during recent decades. The functionalization of TB at almost any posi-

tion in its skeleton is now possible and we discuss in detail recent developments in the functionalization of TB in the aromatic rings and in the methano bridge. The reopening of the functionalization of the diazocine ring itself is also discussed. In addition, progress in the synthesis of heterocyclic TB analogues and recent developments in the field of fused TB analogues are covered. The improvements in synthetic approaches have resulted in TB analogues with interesting properties that have inspired investigation of TB analogues in new fields of applications, among others as receptors, as molecular torsion balances, as ligands in asymmetric catalysis, as drug candidates, and as new materials for photo- and optical applications. The most recent developments in those fields are also discussed.

Introduction The history of Tröger’s base (TB) began in 1887 when Carl Julius Ludwig Tröger published a paper on condensations between aromatic amines and methylal [CH2(OCH3)2].[1] From the reaction between p-toluidine and methylal in aqueous HCl he isolated an unexpected product that he described as “base C17H18N”. Nearly half a century later, in 1935, the correct chemical structure was determined when Spielman assigned TB as racemic 2,8-dimethyl6H,12H-5,11-methanodibenzo[b,f][1,5]diazocine [(⫾)-1, Figure 1] after carefully studying the reactivity of the substance.[2] In the same year, Wagner reported a proposed mechanism for the condensation by which TB (1) is formed (vide infra).[3] TB consists of a bicyclic aliphatic unit fused with two aromatic rings. The central methanodiazocine unit projects [a] Center of Analysis and Synthesis, Department of Chemistry, Lund University, P. O. Box 124, 22100 Lund, Sweden Fax: +46-462224119 E-mail: [email protected] Homepage: http://www.chem.lu.se/People/Warnmarkgroup/ Eur. J. Org. Chem. 2012, 7015–7041

Figure 1. TB [ (⫾)-1] and the MMFFs-optimized structures of its two enantiomers.

the aromatic rings in nearly perpendicular fashion, making TB a rather rigid V-shaped molecule possessing a hydrophobic cavity (Figure 1). In addition, TB is C2-symmetric and thus a chiral molecule. The methylene bridge precludes pyramidal inversion of the two nitrogen atoms, making them configurationally stable stereogenic centers; TB was one of the first molecules containing such N atoms to be isolated. Because the stereogenic N atoms are bridgehead atoms, only the enantiomers of either the R,R or S,S configuration are possible (Figure 1), with the diastereomeric (R,S)-TB being geometrically unfeasible.

© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

7015

Ö. V. Rúnarsson, J. Artacho, K. Wärnmark

MICROREVIEW In 1944, Prelog and Wieland resolved the two enantiomers of TB by chromatographic separation with (+)-α-lactose hydrate,[4] making TB one of the first molecules to be resolved on an enantiopure stationary phase. In 1986 the structure proposed fifty years earlier by Spielman was unambiguously confirmed by single-crystal X-ray diffraction analysis by Wilcox.[5] The matter of the disputed absolute configuration of TB was settled in 1991, when XRD analysis of a diastereomeric salt of monoprotonated TB containing a chiral counterion of known configuration concluded that (+)-TB had the S,S configuration.[6] Despite the widespread use of liquid chromatography for the resolution of (⫾)-1, efficient enantiomeric separations of functionalized TB analogues were unknown for a long time. However, the use of commercial HPLC stationary phases for this purpose has recently been reported.[7] The separation of enantiomers of TB and its analogues on semipreparative scales and in routine manner is now possible. Prelog was the first to report that TB slowly racemizes in a dilute acidic medium. He postulated that the inversion of configuration occurs through the reversible formation of the methylene-iminium ion 2 (Scheme 1) as a key intermediate,[4] despite the fact than no spectroscopic evidence for such an intermediate had been observed.[8] Nevertheless, it was suggested that the iminium species exists only transiently and in too low a concentration to be detected. This mechanism of racemization is indirectly supported by the fact that ethano-bridged TB analogues, which are unable to form such stabilized iminium intermediates, are configurationally stable to acidic conditions.[9] In concentrated acids, in which both nitrogen atoms are protonated,[8] it is not

possible to form the iminium intermediate, probably because there is no free electron pair available for the opening of the bridge, and so racemization does not occur.

Scheme 1. Mechanism of racemization of TB in acidic media.

A different racemization mechanism was proposed by Trapp and Schurig.[10] They suggested that in the (inert) gas phase at ambient temperature the TB racemization pathway involves a retro-hetero-Diels–Alder ring opening followed by hetero-Diels–Alder ring closure, via intermediate 3 (Figure 2). However, recent work by Schröder and co-workers studying the epimerization of fused bis-TB analogues by ion-mobility MS demonstrated that inversion of the configuration only occurred for the protonated species. The absence of this process in the cationic sodiated species suggested that the sequence via iminium ion 2 is the more probable mechanism.[11] The pKa of the monoprotonated salt of TB in 50 % aqueous alcohol was determined by Wepster to be 3.2.[12] The V-shaped structure of TB forces the free electron pairs of

Ögmundur Vidar Rúnarsson obtained his B.Sc. in 2002 and M.Sc. in pharmaceutical sciences in 2004 from the University of Iceland. He is currently a postdoc fellow in the K.W. group, where he is synthesizing new TB analogues and investigating their drug–receptor interactions. He received his Ph.D. in medicinal chemistry in January 2009 from the University of Iceland, where he worked on synthesis and structure–activity relationship investigations of novel analogues of chitosaccharides (polymers, oligomers, and monomers) against various Gram-negative and Gram-positive bacteria.

Josep Artacho graduated from Universitat de Girona (Spain) in 2004. He joined the K.W. group as an exchange student the following year and then later did his Ph.D. project in the group. The purpose of his PhD work was to fuse a number of Tröger’s base molecules together linearly and in a controlled manner. Special attention was given to the functionalization of the TB core, leading to interesting molecules such as twisted amides and crown ethers. Dr. Artacho obtained his Ph.D. in October 2011. He is currently a postdoc fellow at the University of Copenhagen.

Kenneth Wärnmark obtained his Ph.D. at the Royal Institute of Technology, Stockholm, under the supervision of Prof. Christina Moberg, working with macrocyclic ligands. He then continued with postdoc studies at l’Université de Strasbourg (1994–1996) with Prof. Jean-Marie Lehn, working on ruthenium-N-heterocyclic complexes. He started his independent career at Lund University as assistant professor in 1996 and he was promoted to senior lecturer in 2000. He became associate professor in 2003 and since December 2010 he has been full professor of organic chemistry at Lund University. His research interests, apart from Tröger’s base chemistry, include supramolecular catalysis, self-assembly, molecular tubes, and molecular receptors.

7016

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041

Tröger’s Base’s 125th Anniversary: Synthesis and Applications of Analogues

Figure 2. Proposed intermediate for the retro-hetero-Diels–Alder/ hetero-Diels–Alder mechanism for the racemization of TB.[10]

the bridgehead nitrogen atoms out of conjugation with respect to the aromatic planes, which hinders full conjugation with the aromatic system. This restriction would be expected to result in the pKa of monoprotonated TB having a value between those of the anilinium ion (ca. 5) and the alkylammonium ion (ca. 10). Wepster explained the apparently abnormal pKa of protonated TB by the fact that the benzylamine and the methylenediamine groups present in TB should lower the basicity and estimated this to be by two pKa units.[12–13] We propose that the low pKa is in addition a consequence of stabilization through an anomeric effect. The electron lone pair of one nitrogen atom thus overlaps with the antibonding orbital of the bond between the carbon at the methylene bridge and the second nitrogen (Figure 3), making the N lone pair even less available for proton binding. This effect can be geometrically observed in the bond lengths of unsymmetrical TB derivatives such as protonated TB. Thus, the bond length between the protonated N and the bridging CH2 has hence increased relative to that in TB itself, whereas that between the other N and the bridging CH2 has decreased (Figure 3) as expected for an anomeric effect,[14] according to DFT calculations.[15]

articles dealing with TB chemistry are currently to be found in the literature, and approximately 60 % of these have been published in the last decade. Review articles and book chapters thoroughly covering TB chemistry have been published regularly in the past years.[17] This review focuses on the synthetic aspects of functionalized TB analogues and their broad spectrum of applications in various fields of chemistry.

Synthesis of Tröger’s Base and Analogues The Tröger’s Base Condensation The first TB synthesis involved the condensation between p-toluidine and methylal in aqueous HCl.[1] The various methodologies used for the synthesis of TB analogues are indeed variations of the original conditions. A synthetic equivalent of methylene – namely formaldehyde or a precursor, such as paraformaldehyde or hexamethylenetetramine – is treated with a suitably substituted aniline derivative under acidic conditions, usually aqueous or alcoholic HCl solutions, acetic acid, trifluoroacetic acid (TFA), or methanesulfonic acid (Scheme 2).[16,18] Nonetheless, the search for improved synthetic protocols is still ongoing, as is illustrated by recent reports[19] on the rediscovery of DMSO, first used by Becker in 1993,[20] as a methylene synthetic equivalent. Furthermore, improvements in Lewisacid-catalyzed TB synthesis[21] and in the use either of unusual condensation media such as ionic liquids[22] or diglycolic acid/polyphosphoric acid[23] or of super-acidic conditions have been reported.[24]

Scheme 2. The TB condensation.

Figure 3. Overlap of the lone pair of one nitrogen atom with the antibonding orbital of the bond of the carbon and the other nitrogen atom (left). Selected bond lengths in TB (middle) and in protonated TB (right), supporting the presence of hypoconjugation (see text for details).

Until the 1980s, TB was used mainly as a model substance for the evaluation of new chiral chromatographic techniques, due to its easy separation into its enantiomers (vide supra). In the second half of the 1980s, however, interest in analogues of TB emerged. In a pioneering work in 1985, Wilcox reported the XRD structure of the TB framework with the synthesis of several simple yet unprecedented TB analogues.[16] This triggered further developments in the synthesis of TB analogues, which have found applications as building blocks in various fields of chemistry. Over 530 Eur. J. Org. Chem. 2012, 7015–7041

Condensations based on this general approach have demonstrated high sensitivity both towards the electronic properties of the substituents and towards the substitution patterns of the aniline components. It was long believed that the substituents on the aromatic ring should have electrondonating natures, to avoid the low-yielding, sluggish reactions observed with electron-withdrawing groups.[25] In addition, it was also believed that a substituent in the paraposition might be needed to avoid polymerization. A synthetic breakthrough occurred, however, with the development of a condensation protocol for the synthesis of halogenated TB analogues, introduced by our group in 2001.[26] The use of paraformaldehyde and TFA overcame the longstanding limitation of electron-withdrawing substitutes mentioned above and gave access to TB analogues substituted with halogen atoms in virtually any position in the aromatic rings.[27] This protocol has become the most regularly used method of all the known variations. Accordingly, this methodology does not only allow multi-gram


www.eurjoc.org

7017


MICROREVIEW syntheses of halogenated TB analogues,[28] but also enables the introduction of a wide range of both electron-donating and -withdrawing substituents, such as alkyl chains, MeO and MeS, COOR, CF3, and even the strong electron-acceptor NO2.[7d,29] Importantly, this method has also proved valid for the condensation of aniline itself, giving the didemethylated TB analogue in a remarkable 78 % yield,[30] so para-substitution to avoid polymerization is no longer needed. Wagner was the first to attempt to explain the acid-induced reaction sequences between formaldehyde and p-toluidine by which TB (1) is formed.[3] Both Farrar[31] and Wagner[18a,32] reexamined this work, which led to the proposed mechanism for the formation of the methano[1,5]diazocine skeleton. The formation involves a series of electrophilic aromatic substitutions as key steps and assumes the presence of four important intermediates (4–7, Scheme 3). The first step of the mechanism (for R = Me, Scheme 3) involves an acid-catalyzed condensation between p-toluidine and formaldehyde to form iminium ion 4, which in turn reacts with a second equivalent of the aniline to give intermediate 5. Two sequential methylene additions accompanied by cyclizations yield TB through intermediates 6 and 7. The rate-limiting step of the reaction sequence is the conversion of tetrahydroquinazoline derivative 6 into the reactive intermediate 7 and the subsequent electrophilic aromatic substitution. The presence of electron-withdrawing groups in the aniline component reduces the nucleophilicity of the secondary amine in 6, resulting in a further decrease in the rate of the intramolecular electrophilic substitution, and gives rise to the formation of the dihydroquinazoline A as a side reaction (Scheme 3). Our group suggested that the use of paraformaldehyde in TFA increases the concentration of protonated formaldehyde relative to formalin/HCl in ethanol

and therefore increases the reaction rate for the formation of 7.[26] In addition, TFA allows for the reduction of dihydroquinazoline A to tetrahydroquinazoline 6 after protonation by the relatively strong acid TFA, resulting in B, and subsequent hydride transfer from paraformaldehyde. Sergeyev, on the other hand, argued that iminium ion 4 is involved in the rate-determining formation of 7 (Scheme 3).[7d] He reasoned that the iminium cation 4 is abundant in the early stages of the reaction, resulting in the rapid formation of 7. Later on, however, the concentration of 4 becomes low due to the slow reverse reactions from 5 and 6. The formation of 7 is therefore sluggish and side reactions occur instead. However, he did not discuss why the reactions occur successfully with electron-deficient aniline derivatives in TFA whereas with formalin/HCl they do not. In a recent work, Eberlin and Coelho monitored the condensations between p-toluidine and formaldehyde or hexamethylenetetramine in situ in TFA by electrospray ionization mass and tandem mass spectrometry [ESI-MS/MS].[33] This technique permitted the detection and characterization of – along with iminium ion 4 (Scheme 3) – the oxidized forms of two of the intermediates suggested by Wagner ([6– H] and [10–H], Figure 4). The natures of the cationic species observed were argued to be due to oxidation of 6 and 7 by the ESI source. The “Wagner mechanism” that was supported by the ESI-MS investigation above was also supported by Wan and co-workers when they isolated compounds 11 and 12 (Figure 4) during the synthesis of TB in an ionic liquid (1butylpyridinium tetrafluoroborate) in the presence of 1methyl-3-[2-(sulfoxy)ethyl]-1H-imidazol-3-ium chloride as catalyst.[34] Upon subjection of the isolated intermediates to the reactions conditions and heating to 150 °C, the expected TB (1) was formed.

Scheme 3. The proposed mechanism for the formation of TBs proposed by Wagner and Farrar. 7018

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


heating of the resulting ethyl oxalate derivative at 185 °C in DMSO, with the oxalate moiety functioning as a formaldehyde equivalent (Scheme 5). This approach was developed further in recent years by Li and co-workers (vide supra).[19] TB analogues have also been synthesized by directly inserting the 5,11-methylene bridge into the tetrahydrodibenzo[b,f][1,5]diazocine framework. This is carried out in very good yields with formaldehyde, which yields TB (1, R = R⬘ = H, Scheme 6).[36] Alternatively, aldehydes and ketones can be used instead of formaldehyde, giving access to analogues with substituents on their methylene bridges (Scheme 6, R ⬆ R⬘ = H and R = R⬘ ⬆ H).[37]

Figure 4. Structures of intermediates and byproducts in the TB condensations.

Alternative Syntheses of Tröger’s Base Analogues Elegant one-step condensations of simple aniline derivatives to form TB analogues are not always applicable. Because of the relatively harsh reaction conditions, the range of analogues is to some extent limited. At the same time, direct condensations between aniline derivatives and formaldehyde only give access to symmetric analogues of TB. To circumvent these limitations, other methods have been investigated. Stepwise methods have been employed for the synthesis of unsymmetrical analogues of TB. Webb and Wilcox suggested the rational synthesis of such analogues through the tethering of two differently substituted aniline derivatives through a methylene unit followed by cyclization with formaldehyde (Scheme 4).[30,35] It is noteworthy that this method also provided the option to prepare TB analogues bearing electron-withdrawing groups for the first time, although the examples were limited to such substitution only in one of the aniline rings.

Scheme 6. Synthesis of TB analogues by insertion of the bridging methylene component into the tetrahydrodibenzo[b,f][1,5]diazocine framework.

Maitra and co-workers reported the asymmetric synthesis of a TB derivative in approximately 40 % ee by means of chiral induction with a 7-deoxycholic acid template, although this would nowadays be referred to as a diastereoselective synthesis, thus giving dr = 7:3 (Scheme 7).[38] The alteration of the spacer lengths, linking the aniline component to the chiral template, was found to influence the selectivity to some extent.[39]

Scheme 7. Asymmetric synthesis with a chiral template.

Scheme 4. Stepwise preparation of unsymmetrical TB analogues.

In another stepwise approach, Becker synthesized a highly functionalized TB analogue by treatment of the aniline derivative with ethyl oxalyl chloride,[20] followed by

Functionalization of Tröger’s Base TB analogues can today be formed by different methods allowing for functionalization in almost any position in the TB core (see Figure 5). All the different functionalization patterns shown in Figure 5 are discussed below.

Scheme 5. Stepwise synthesis of a highly functionalized TB analogue. Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7019


MICROREVIEW

Figure 5. TB (1) and functionalization of TB discussed in this review (HCA = heterocyclic amine).

Functionalization of the Aromatic Rings Halogenation and Cross-Coupling Reactions Regrettably, in the prominent history of TB there has been a lack of general methodologies to functionalize the TB core, resulting in restricted access to its analogues. Of the chemistry devoted to the derivatization of TB, most attention has centered on the aromatic rings, due to the easy access to halogenated analogues of TB as previously mentioned.[7d,27–28,29b,30,40] In our 2001 paper introducing the protocol for the synthesis of halogenated analogues of TB,[26] it was demonstrated that such analogues can provide access to many other functional groups through the introduction of ethynyl groups by Corriu–Kumada cross-cou-

pling (Scheme 8, Route a). Indeed, several research groups including ours have applied a series of transition-metal-catalyzed C–C cross-coupling reactions with halogenated TB analogues as substrates. We introduced substituted alkynyl groups by Sonogashira coupling (Scheme 8, Route b).[28c,40b,41] Lützen introduced arylations by Suzuki– Miyaura coupling with arylboronic acids as coupling partners to halogenated TB analogues (Scheme 8, Route c).[41–42] A study published by our group, comparing different palladium-catalyzed cross-coupling methods for the introduction of aryl and heteroaryl groups via metalated TB analogues (Scheme 8, Route d), demonstrated Suzuki coupling to be the best method, giving excellent yields, whereas Stille and Negishi coupling both gave moderate to good yields.[42b] In addition to classical cross-coupling reactions, C–C bond formation has also been achieved by palladium-catalyzed cyanations (Scheme 8, Route e).[42a,43] Carbon-heteroatom bond formation through transitionmetal-catalyzed reactions starting from dihalogenated analogues of TB has also been reported. Ullmann conditions have been employed for C–O bond formation (Scheme 9, Route a).[7c,41] Our group has formed C–N bonds from halogenated TB analogues by halogen/lithium exchange followed by quenching with TSN3 and subsequent reduction of the corresponding azide (vide infra).[44] This can also be carried out by Buchwald–Hartwig protocols such as Cucatalyzed amidation[42a] and Pd-catalyzed amination (Scheme 9, Route b).[42a,43,45] This latter example is thoroughly discussed in the section on functionalization of TBs by amination reactions.

Scheme 8. Selected examples of transition-metal-catalyzed transformations with halogenated analogues of TB. Reaction conditions: a) ethynylmagnesium bromide, Pd(PPh3)4; b) Pd(PhCN)2Cl2, P(tBu)3, CuI, alkyne-R substrate; c) Pd[P(tBu)3]2, CsF, substituted phenylboronic acid; d) for example, nBuLi, THF, –78 °C, B(OCH3)3, then Pd[P(tBu)3]2, CsF, substituted halobenzene; e) Zn(CN)2, Zn, Pd(Ph3)4, dppf.

Scheme 9. Selected examples of transition-metal-catalyzed carbon–heteroatom bond formation with diiodo-TB analogues. Reaction conditions: a) NaOCH3, CuCl, MeOH/DMF; b) 1) Pd2(dba)3, BINAP, NaOtBu, benzophenone imine, toluene. 2) HClaq, THF. 7020

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


Our group has also developed a different approach to synthesize valuable analogues of TB.[46] Symmetrical disubstituted TB analogues were obtained by double halogen/ lithium exchange with 2.4 equiv. nBuLi followed by electrophilic quenching. (Scheme 10, Route A). Alternatively, subjection of the 2,8-dibromo-substituted TB analogue to the same conditions, but now with only 1.1 equiv. of nBuLi, resulted instead in the asymmetrically monosubstituted analogue of TB (Scheme 10, Route B). The selectivity for the single bromine/lithium exchange was attributed to a solvent effect. It was suggested that the dilithiated species was better solvated in THF than in Et2O. This difference in solvation results in a larger difference in energy between the mono- and the dilithiated species in the THF/Et2O solvent mixture than in THF. Through a second bromine/lithium exchange of the singly substituted TB analogue and subsequent electrophilic quenching, asymmetric disubstituted TB analogues were obtained. We have used the desymmetrization protocol as the key step in the synthetic development of fused tris-TB analogues (vide infra). Our developed double halogen/lithium exchange protocol was later used by Diederich and Sergeyev[47] and by Saigo and co-workers as a key step in the regio- and diastereoselective tether-directed remote functionalization of C60 with analogues of TB.[48]

Scheme 10. Halogen/lithium exchange protocol to provide symmetric and asymmetric analogues of TB.

Recently, different studies on the direct halogenation of the aromatic rings of TB analogues have been reported. Didier and Sergeyev synthesized the 8-bromo-2-iodo-substituted TB analogue shown in Scheme 11 from the di-demethylated analogue of TB by sequential iodination and bromination.[49] In addition, Try and co-workers studied the mono- and dichlorination and -bromination of di-demethylated TB and also of differently substituted analogues of TB, resulting in moderate overall yields of halogenated TB analogues.[50]

Scheme 11. Sequential halogenation of di-demethylated TB.

Amination The synthesis of 2-amino-8-bromo-substituted analogues of TB by our desymmetrization protocol discussed above (Scheme 12, Route A) plays a key role in the synthesis of fused analogues of tris-TB.[40b,44] In contrast, by the double halogen/lithium protocol we were able to synthesize the 2,8diamino-substituted TB shown in Scheme 12 (Route B) from a dihalogenated TB analogue.[44] In our 2006 paper we attempted to optimize the conditions further and in particular to increase the scale of the double lithiation approach.[44] It was observed during this investigation that the use of large amounts of nBuLi resulted in more complex mixtures, making the protocol unsuitable for the preparation of diamino-substituted TB analogues on a large scale. An alternative approach using the palladium-catalyzed aminations of aromatic halides developed by Hartwig and Buchwald was then envisaged.[51] This alternative approach gave the best results when the reaction was performed on a 0.5 mmol scale of 13 (Scheme 13) in toluene at 80 °C in the presence of Pd2(dba)3 (0.5 mol-%), BINAP (1.5 mol-%), benzophenone imine (2.4 equiv.), and NaOtBu (2.8 equiv.), followed by acidic hydrolysis to give 15b in an excellent yield of 89 %. Scaling up to 4 mmol of starting material 13 with 0.7 mol-% precatalyst and 2.1 mol% BINAP lowered the yield somewhat to 76 %.

Scheme 12. Synthesis of monoamino-substituted TB analogue 15a and diamino-substituted TB analogue 15b by single and double bromine/lithium exchange protocols. Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7021


MICROREVIEW

Scheme 13. Palladium-catalyzed synthesis of a diamino-substituted TB.

Some years earlier, Didier and Sergeyev had reported the synthesis of 2,8- and 4,10-diamino-substituted TB analogues from the corresponding 2,8-diiodo- and 4,10-dibromo-substituted substrates in 86 and 69 % yields, respectively, also with use of the benzophenone imine/BINAP/NaOtBu/Pd2(dba)3 catalytic system on a 1 mmol scale, but with 25 mol-% precatalyst.[43] Although the yields were comparable to those obtained by us, the amount of precatalyst used was significantly higher. This contrast motivated a more extensive study of the application of the methodology,[45] with the 2,8-dibromo-, 2,8-dibromo-4,10-dimethyl-, 2,8-diiodo-, and 2,8-diiodo-4,10-dimethyl-subsituted compounds as starting materials. These compounds represent different reactivities (bromo and iodo), and solubilities (methyl and non-methyl), which are important issues for controllability of the amination reaction on a larger scale (4 mmol). The investigation revealed that the catalyst was poisoned by the released iodide formed in the reaction. Repetition of the reaction in the presence of NaI supported this, and as expected the yield decreased when NaI was added, explaining the need for higher catalytic loading when iodoaniline derivatives were employed in the TB condensation. Pd-catalyzed monoamination of 2,8-dihalo analogues of TB was attempted, but resulted only in moderate yields of isolated pure monoamino analogues of TB, although satisfactory ratios between the monoamino- and diamino-substituted TB products were observed in the crude reaction mixture. The best yield of monoamino-substituted TB analogue (51 %) was achieved after a reaction time of only 2 h. The desymmetrization protocol with selective monolithi-

ation of dibromo-substituted TB analogues followed by TSN3 addition and finally reduction, discussed above, gave yields of the monoamino-substituted TB product comparable to those of the Pd-catalyzed reactions (illustrated in Scheme 12).[44] In another approach to amino-substituted TB analogues, Lützen and co-workers obtained a 59 % overall yield (two steps) of the 2,8-diamino-substituted TB analogue by reduction of the 2,8-dinitro-substituted TB analogue,[29c] the latter synthesized by a TB condensation with the corresponding nitroaniline.

Functionalization of the Methanodiazocine Ring Functionalizing the Nitrogen Atoms in the Methanodiazocine Ring The nitrogen atoms of the methanodiazocine ring are clear targets for functionalization. Monoquaternary salts of TB are easily prepared by treatment with the appropriate alkyl or benzyl bromide (Scheme 14, Path A).[18b,52] Quaternization of both nitrogen atoms has been argued to be unfeasible due to the strong negative inductive effect from the first formed quaternary nitrogen atom, which might be expected to eliminate the nucleophilicity of the other tertiary nitrogen atom. However, Lenev and co-workers have recently reported the synthesis of the dimethylated TB analogue on treatment with dimethyl sulfate.[9b] Acetylation and benzoylation were performed in the original elucidation of the structure of TB (1) by Spielman,[2] who obtained the diacylated derivative with the loss of methylene bridge carbon atom as formaldehyde (Scheme 14, Path B). Nitrosation of TB (1) with the loss of the methylene bridge was also reported by Spielman.[2] Cooper and Partridge converted the resulting di-N-nitroso analogue into diamine 16 (Scheme 15) by treatment with CuCl/HCl.[37a] Compound 16 was then condensed with various aldehydes and ketones to afford analogues of TB containing functionalized methylene bridges. This stepwise methodology was later used as a general methodology for the synthesis of different analogues of TB with mono- or disubstituted methylene bridges.[8,37b]

Scheme 14. Alkylation, benzylation, and acylation of TB itself.

Scheme 15. Stepwise synthesis of analogues of TB with substituted methano bridges. 7022

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


Recent progress in the functionalization of the methanodiazo ring has made it possible to avoid the removal of the methano bridge before functionalization. In 2012 Periasamy and co-workers reported exchange reactions of the methano bridge of TB (1) with aromatic carbonyl compounds in the presence of TiCl4 or POCl3 (Scheme 16, Path A).[53] The proposed mechanism for the TiCl4-induced reaction involved cleavage of the methano bridge in situ (Scheme 16, Path B). In addition, Periasamy and coworkers demonstrated that in the presence of POCl3 it was possible to use DMF as the carbonyl partner, to obtain the corresponding TB analogue, under Vilsmeier–Haack conditions. In the same year Reddy and co-workers also reported the regioselective synthesis of dialkylamino-substituted TB analogues under Vilsmeier–Haack conditions.[54] This was achieved by treatment of DMF with various aromatically functionalized methano-TB analogues in the presence of POCl3, yielding arylalkylamino-substituted TB analogues substituted on their respective methano bridges in yields ranging between 53–83 %. Use of POBr3 instead gave similar results for the insertion of N,N-disubstituted amines.[55]

Scheme 16. One-pot syntheses of analogues of TB containing substituted methano bridges.

A different approach for the modification of the methano bridge involving its cleavage was reported by Hamada and Mukai.[9a] They synthesized the ethano-TB analogue by treatment with 1,2-dibromoethane (Scheme 17). It was suggested that the “bridge-replacement” took place through ammonium ion and dibromide intermediates before formation of the ethano-TB analogue along with dibromomethane. The introduction of spiro[4.5]lactone straps onto the TB scaffold was developed by Try and co-workers.[56] It was

achieved by treatment of TB with phthaloyl dichloride and Et3N to yield analogue 17 (Figure 6). No mechanistic aspects were discussed, but the same product was obtained with diazocine 16 (Scheme 15) as a starting material, suggesting a probable cleavage of the methylene bridge of TB as part of the reaction mechanism.

Figure 6. Different methylene-bridge-substituted TB analogues.

Other examples of modifications of the methylene bridge of TB without involvement of its prior cleavage have recently been reported. In 2006, Kim and co-workers reported the synthesis of what they believed to be the unusual [3.3.3]bicyclic analogue 18 (Figure 6) formed by treatment of TB (1) with methyl propiolate in the presence of ZnBr2 in CH3CN.[57] A year later, however, Lenev and co-workers demonstrated by XRD analysis that the actual structure of the product of this reaction was in fact that of the TB analogue 19 (Figure 6).[58] Lacour and co-workers have recently developed two different approaches to the stereoselective synthesis of the configurationally stable ethano-TB analogues.[9c,59] One involved the formation of the quaternary salts 20 of TB (Scheme 18) by treatment with alkyl halides, followed by [1,2]-Stevens rearrangements promoted by basic alumina[9c] to give the ethano-TB analogues 21 with diastereomeric ratios (drs) of ⬎98:2. The other approach involved one-step rhodium(II)-catalyzed reactions proceeding through the formation of electrophilic metal carbenoids formed by reactions between the catalyst and diazo esters.[59] The metal carbenoids attached to the N-lone pair of the TB to give nitrogen ylides that finally rearranged to generate the configurationally stable ethano-bridged TB analogues 22 (Scheme 18). High diastereoselectivities of up to 49:1 were observed, as well as enantioselectivities of up to 99 % ee in the retention of configuration at the 5- and 11-positions when enantiopure TB was used as starting material. In 2012, Lacour and co-workers extended the scope of their chemistry by showing that CuTC [copper(I) thiophene-2carboxylate] catalyzes the same reactions with good diastereo- and enantiocontrol (drs and ees of up to 12:1 and 95 %, respectively) (Figure 18).[60]

Scheme 17. Synthesis of an ethano-TB analogue. Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7023


MICROREVIEW

Scheme 18. Synthesis of substituted ethano-TB analogues.

Scheme 19. Mono- and disubstitution at the exo positions of the benzylic methylenes of TB.

Synthesis of 6-exo- and 6,12-exo,exo-Substituted Analogues of Tröger’s Base Harmata and co-workers were the first to conduct derivatization of the benzylic methylene components in TB,[61] although TB analogues substituted in these positions had been synthesized previously, but not from TB itself.[62] Harmata’s methodology from 1996 involved the metalation of the benzylic methylene groups by treatment of TB (1) with BF3·OEt2 followed by nBuLi and then quenching with an electrophile to give the exo-monosubstituted species. exo,exo-Disubstitution could also be achieved by sequential monometalation and electrophilic quenching (Scheme 19).[63] The downside of Harmata’s procedure is that disubstituted analogues have to be synthesized in a sequential manner, resulting in decreased overall yields.[63] It was demonstrated that exo-6-substituted TB analogues (R = –CH2OH, –CPh2OH, –CH2CPh2OH, or –CH2OCH2CPh2OH) were reasonably good asymmetric inductors in additions of Et2Zn to aromatic aldehydes.[61b]

benzyltriethylammonium chloride (BTEAC) in anhydrous CH2Cl2 at reflux for 9 h, affording monolactam 24 in 25 % yield together with 30 % reclamation of unreacted starting material. The reaction conditions were then optimized to perform a double benzylic oxidation. This was achieved through the use of 9 equiv. of the reagents and a 4 h reaction time, to afford the bis-lactam 25 (Scheme 21) in 28 % yield, as well as 15 % and 5 % yields of the quinazoline side products 26 and 27, respectively.

Oxidation of the Benzylic Methylenes of Tröger’s Base The oxidized TB species 24 (Scheme 20), a lactam analogue of TB, was originally obtained serendipitously by electrophilic quenching by DMF after treatment of 1 with 1 equiv. sBuLi/TMEAD and accidental air oxidation.[64] This led to a joint investigation between the Snieckus, Harmata, and Wärnmark groups to find a robust preparative method for the oxidation.

Scheme 20. Formylation of TB (1) and subsequent oxidation to form the monolactam TB analogue 24.

A variety of oxidizing agents such as different manganese and chromium salts, ruthenium catalysts, and hypervalent iodine were screened. The best results, however, were obtained when 1 was treated with 3 equiv. of KMnO4 and 7024

www.eurjoc.org

Scheme 21. Synthesis of the bis-lactam TB analogue 25 by direct oxidation.

Interestingly, the solid-state structure revealed the bislactam TB analogue 25 to be an example of a twisted amide (Figure 7). The monooxidized analogue 24 was also identified as a twisted amide on the basis of the similar spectroscopic data for 24 and 25. The XRD analysis revealed that the twist angle τ,[65] describing the deviation from co-planarity between the carbon π orbital and the nitrogen lonepair, is –43.7° for TB analogue 25, rather than the values close to 0° and 180° commonly found in unconstrained cisoid and transoid amides, respectively. Furthermore, the overall distortion parameter θ,[66] an additive term that provides a quantitative description of the combined deformation or pyramidality of the nitrogen and carbonyl carbon together with the twist angle, was determined to be 106.1°, whereas that of a simple planar amide is 0°. As a


Eur. J. Org. Chem. 2012, 7015–7041


Scheme 22. A suggested mechanism for the acid-catalyzed hydrolysis of 25 that is consistent with the experimental data.

result of the presence of two such amide functionalities in the molecule, compound 25 is classified as a twisted bisamide, and is, to the best of our knowledge, the first example of such a compound reported to date.

E,Z, and E,E) in approximately 69:33:2 ratio, according to 1 H NMR and GC–MS analysis.

Figure 7. ORTEP representation of the solid-state structure of twisted bis-amide 25.

The reactivity of the amide functionality in TB analogue 25 was investigated in kinetic studies of its acid-catalyzed hydrolysis in which the twisted bis-amide 25 was subjected to different concentrations of DCl (0.058 to 0.87 m) in a CD3CN/D2O mixture (5:1) and the hydrolysis was followed in situ by 1H NMR spectroscopy. Pseudo-first-order kinetics were observed at the acid concentrations employed, measured at ambient temperature. Under these conditions, the product of the amide hydrolysis – compound 30 (Scheme 22) – being a normal amide, did not undergo further hydrolysis. Amide 30 was characterized directly in the reaction mixture by NMR and ESI MS spectroscopy. Hydrolysis experiments were also performed with 18Olabeled water in order to investigate the reversibility in the formation of the tetrahedral intermediates. The detection of 18 O-labeled 25 can only be explained by the involvement of the reversible formation of tetrahedral intermediates 28 and 29. Synthesis of 6-endo- and 6,12-endo,endo-Substituted Analogues of Tröger’s Base via its Twisted Amides The unconventional amide reactivity that is manifested in twisted amides[67] was also observed for monoamide 24 and bis-amide 25.[68] Wittig reactions are not possible with normal amides. However, in a collaboration between the Wärnmark, Snieckus, and Harmata groups, a Wittig protocol approach with the bis-amide 25 and Ph3P=CHCO2Et was successfully employed to give the olefinated bis-enamino-substituted TB analogue 31 (Scheme 23) in 77 % yield as a mixture of the three possible diastereomers (Z,Z, Eur. J. Org. Chem. 2012, 7015–7041

Scheme 23. The synthetic route to endo,endo-substituted analogues of TB.

Compound 31 was hydrogenated over Pd/C, resulting in compound 32. NOESY experiment showed through-space coupling of the H-6 and H-12 protons with the protons on the methylene bridge. This correlation is only possible if the ethoxycarbonylmethyl chains are in endo configurations, leaving the hydrogen atoms at the 6- and 12-positions of 32 oriented towards the methylene bridge. The stereoselectivity of the hydrogenation is due to the fact that the convex side of 31 is more accessible to the heterogeneous catalyst than its more hindered concave aromatic cavity. By the same methodology monolactam 24 could be converted into the 6-endo-ethoxycarbonylmethylene-substituted TB, the first reported endo-substituted TB analogue. Although one endo,endo-TB analogue had previously been synthesized in low yield from a 2,5-disubstituted diazocine,[62a] the synthetic sequence described above is, to date, the first rational synthesis of the highly desirable


www.eurjoc.org

7025


MICROREVIEW endo,endo-substituted TB analogues, introducing functional groups oriented towards the aromatic cavity of TB. To demonstrate the utility of endo,endo-substituted TB analogues, diester 32 was reduced to diol 33 (Scheme 23) and further alkylated with diethylene glycol ditosylate, which surprisingly resulted in a dimerization to give the bisTB crown ether analogue 34 in 30 % yield as a mixture of the meso isomer (RR,SS; meso-34) and the racemate (RR,RR and SS,SS; rac-34), as revealed by 1H NMR and MS-ESI analysis. The structure of meso-34 was unambiguously assigned by XRD analysis (Figure 8), which corroborated the endo,endo configurations of compounds 31–34 previously established by NOESY experiments. Interestingly, the crystal structure of meso-34 revealed that the methano bridges of each of the two TB frameworks were directed towards one another. This configuration was also established in solution by NOESY experiments.

mainly non-directional solvophobic effects and van der Waals interactions.[69] Additional interactions such as aromatic stacking and cation–π interactions are feasible if the surface is aromatic. Synthetic molecules with extended concave aromatic surfaces are generally referred to as molecular cleft compounds. TB analogues, with their relatively rigid chiral concave aromatic surfaces, are inherently good structural motifs for the construction of synthetic receptors. In this context, attention has been drawn to the synthesis of fused analogues of TB. A fused TB analogue is a TB that contains more than one methanodiazocine unit and has an aromatic ring fused to two such units (Figure 9). “Bis-”, “tris-”, and “oligo-”fused TB analogues thus refer to molecules containing two, three, or more methanodiazocine units, respectively. Dolenský and co-workers have recently reviewed the progress in fused TB chemistry.[70]

Figure 9. Representative structural unit of a fused TB analogue.

Figure 8. ORTEP representation of the solid-state structure of the bis-TB crown ether analogue meso-34. Hydrogen atoms have been omitted for clarity.

Fused Tröger’s Base Analogues Recognition processes of relatively unfunctionalized molecules often take place on concave surfaces and involve

The first molecule containing two methanodiazocine units fused to the same aromatic ring was synthesized by Pardo in 2001,[71] by an approach similar to the Wilcox procedure for the synthesis of unsymmetrical TB analogues (see Scheme 4). In this line of research, mono-TB analogue 35 (Scheme 24) was extended stepwise to the fused bis-TB analogue 36 (Pathway A) in the forms of the syn (36a) and anti (36b) diastereoisomers. The total yield of TB analogue 36 was 14 % after nine steps from commercially available starting materials. Both symmetrical (R = R⬘) and unsymmetrical (R ⬆ R⬘) analogues could be synthesized. A shorter route to fused bis-TB analogues starting from 1,2-, 1,3-, and 1,4-diaminobenzenes was later reported (Pathway B, Scheme 24).[72] This “simultaneous” formation strategy, originally employed in the synthesis of symmetric TB analogues (36 with

Scheme 24. Stepwise (A) and simultaneous (B) route for the preparation of fused bis-TB analogues. 7026

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


R = R⬘, 38 and 39, Figure 10), was later applied in the synthesis of unsymmetrical fused bis-TB analogues (36 with R ⬆ R⬘),[72a] although it gave lower overall yields than the stepwise approach.

Figure 10. Two examples of the “simultaneous formation” of fused bis-TB analogues. Each diaminobenzene results in specifically one regio- and one diastereoisomer.

Both cyclizations, by the stepwise (A) and the “simultaneous” (B) routes, in the formation of fused bis-TB analogues 36 are regio- and stereoselective. Of the two possible regioisomers, only the one with the 1,2,3,4-substituted central aromatic ring (the angular isomer) was obtained.[71–72] The synthesis of fused bis-TB 36 (R = NO2, R⬘ = CH3) by the two different routes (Scheme 24) gave different relative proportions of the syn and anti diastereoisomers. Whereas the stepwise approach resulted in a 4:1 mixture of isomers 36a and 36b,[71] the “simultaneous” strategy gave a 1:1 mixture.[72b] It is noteworthy that the latter method gave only the anti isomer 36b when the temperature was reduced from 90 °C to 50 °C, albeit in 10 % yield. The relative influence of kinetic and thermodynamic control in Methods A and B was briefly investigated in isomerization studies, in which isomerization experiments were performed on fused bis-TB analogue 36. Subjecting the chemically pure syn isomer 36a (R = NO2, R⬘ = CH3) to the cyclization reaction conditions (formalin/HCl in EtOH, 90 °C, 24 h), for instance, resulted in a mixture with a 4:1 ratio of the syn and the anti isomers.[71] On the other hand, subjecting the anti-bis-TB analogue 36b (R = R⬘ = CH3 or NO2) to similar conditions returned a 1:1 mixture of the two isomers.[72b,72c] This is probably due to the syn isomer being thermodynamically more stable than the anti isomer as a result either of πstacking interaction between the parallel aromatic rings of the syn isomer or of differences in solvation energies between the two isomers.

The first fused tris-TB analogue was synthesized in 2002 by Dolenský and Král,[72a] who built a system in which all three methanodiazocine units share a common central benzene ring (Scheme 25). By the “simultaneous” approach, fused tris-TB analogue 40 was synthesized from 1,3,5-triaminobenzene. Of the two possible diastereoisomers, only the fused tris-TB analogue 40a (R = CH3) was isolated. In a later work from the same group, the synthesis of differently substituted 40a by the same strategy was reported.[73] The one-pot synthesis of fused tris-TB analogue 40 was also achieved by the condensation of 1,3,5-triaminobenzene and p-toluidine in 2 % yield, meaning over 80 % yield “per bond formed”. Remarkably, rapid epimerization of fused tris-TB analogue 40 occurred in TFA at room temperature, which allowed the isolation of fused tris-TB analogue 40b. The structures of both diastereomers were unambiguously assigned by single-crystal XRD analysis. The solid-state structures revealed fused tris-TB analogue 40a to have a “throne” configuration, whereas 40b displayed a “calix” shaped cavity (Figure 11).

Figure 11. ORTEP representations of the throne tris-TB 40a (left) and the calix tris-TB 40b (right). Hydrogen atoms are omitted for clarity.

Naphthalene and anthracene derivatives of throne and calix tris-TB analogues were also prepared by Dolenský and Král.[74] The sizes of the cavities were estimated by molecular modeling. The volume of the cavity of calix tris-TBA (a benzene tris-TB analogue) was 30 % of the volume of the cavity of α-cyclodextrin (α-CD; CDs were used as

Scheme 25. Synthesis of tris-TB 40. Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7027


MICROREVIEW references). The volumes of the cavities of calix-tris-TB-B (a naphthalene tris-TB analogue) and -C (an anthracene tris-TB analogue) were calculated to be 65 % and 124 %, respectively, of the volume of the cavity of α-CD (Figure 12).

Figure 12. Optimized geometries of calix tris-TB derivatives (from the left: calix-A, calix-B and calix-C). A purple cone indicates the calculated cavity volume.

Fused tris-TB analogues not sharing common benzenoid rings can variously form linear symmetric, non-symmetric, or non-linear symmetric regioisomers (Figure 13).[75] Pardo and co-workers were the first to introduce linearly fused tris-TB analogues.[76] By a stepwise methodology starting from fused bis-TB analogues 36a and 36b (Scheme 24), they obtained three of the four possible diastereomers of the “non-linear symmetric” fused tris-TB analogue (41–44, R = CH3 and R⬘ = NO2; Figure 14). The syn-syn-TB analogue 41 was envisaged to have interesting host–guest applications due to its cage-like structure but this was never investigated. It was also revealed that fused tris-TB analogue 42 epimerized in 0.8 m HCl solution in EtOH at 80 °C to a 1:2:2:2 isomeric mixture of 41/42/43/44, analogously with fused bis-TB analogues. Our group contributed with another strategy to build fused tris-TB analogues.[75] Both the syn-anti and the antianti non-linear symmetric fused tris-TB analogues 43 and 44 (R = R⬘ = Br; Figure 14) were synthesized from the starting 2,8-dibromo-substituted TB analogue, and isolated by the desymmetrization route. The anti-anti diastereomer 44 is the first example of an isolated fused tris-TB analogue of this configuration.

Figure 13. Regioisomers of fused tris-TB analogues not sharing common benzenoid rings: “linear symmetric”, “non-symmetric”, and “non-linear symmetric”.

Compounds 43 and 44 were subjected to an epimerization study: stirring of 43 in TFA for several days at room temp. gave a mixture of 43 and 44 in an approximately 4:1 ratio, whereas subjection of 44 to the same conditions returned only 44. It was argued that this occurred because of the higher thermodynamic stability of anti-anti 44, probably due to better solvation. Kessler and co-workers recently presented cases in which a guest molecule [tetracyanobenzene (TCB), Figure 14] was captured for the first time in the cavities of fused-TB analogues.[77] Five synthesized TB tweezers were modeled and synthesized for this purpose (Figure 15). These pincers or tweezers were designed to hold and release the guest molecule. Association constants were obtained by 1H NMR titrations and compared with calculated binding energies from computational chemistry modeling. Interestingly, all the synthetic methodologies described above gave exclusively the non-linear symmetric regioisomers of fused tris-TB analogues, with none of the other

Figure 14. The four possible diastereomers of the “nonlinear symmetric” tris-TB, and tetracyanobenzene (TCB). Pardo and co-workers: R = CH3, R⬘ = NO2.[76] Our work: R = R⬘ = Br.[75] 7028

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


Figure 15. The lowest-energy structures of the tweezer/TCB complexes and definition of the ligand–tweezer distances (d1) and the tip dimension (d2, displayed for D only).

plausible regioisomers – the linear symmetric and non-symmetric (Figure 13) – being observed. This can be explained, as in the case of fused bis-TB analogues, in terms of the preference to form 1,2,3,4-substituted aromatic rings, so called angular isomers, during the cyclization.[44] To implement linearity in fused-tris-TB analogues, Král and coworkers used 2,5-dimethylbenzene-1,4-diamine in a one-pot oligomerization reaction,[78] which gave a mixture of fused mono-TB, bis-TB, and tris-TB analogues in 5 %, 10 % and 1 % yields, respectively. The methyl group was used to block other reactive positions. The presence of higher generations of fused TB analogues could be detected by MS analyses, but none of these isomers were ever isolated. To us, the linear symmetric regioisomers of fused tris-TB analogues are the most interesting, due to their cavityshaped binding sites, especially the syn-syn diastereoisomers, which would result in nearly circular geometries of the molecules. At that time, however, such regioisomers had only been obtained in low yields and were synthesized in an uncontrolled manner (vide supra).[78] Our group developed a synthetic strategy for the rational construction of linear symmetrically fused tris-TB analogues by a stepwise approach in which the unwanted reactive positions of the aniline were blocked (Scheme 26). In our 2006 paper we thus reported a condensation between aniline 47 (Scheme 26) and paraformaldehyde in TFA to afford the C2-symmetric dibromo-substituted TB analogue 48,[44] and this was then desymmetrized by application of our previously published conditions.[46,75] TB analogue 48 was thus subjected to a single bromine/lithium exchange (1.1 equiv. of nBuLi), followed by electrophilic Eur. J. Org. Chem. 2012, 7015–7041

quenching with tosyl azide, yielding the asymmetric 2azido-8-bromo-substituted TB analogue 49. Reduction with NaBH4 gave the amino-substituted TB analogue 50 in an excellent yield. Compound 50 was in turn treated with paraformaldehyde in TFA to afford the linear symmetrically fused tris-TB analogue 51 as the only regioisomer. 1H NMR analysis of the crude product of the reaction mixture revealed a diastereomeric ratio of 53:34:13 of syn-anti (51a), anti-anti (51b), and syn-syn (51c) isomers when the TFA used in the reaction was added at –10 °C. All the isomers – 51a–c – could be isolated by preparative TLC. Isomers 51a and 51c was assigned by single-crystal XRD analysis (Figure 16) and the remaining third isomer was then assigned as 51b, supported among other things by its symmetrical NMR spectra. The ratio of diastereomers was affected only to a minor extent by allowing the temperature during the addition of TFA to increase to 0 and 10 °C. This led to lower yields, however, which is in agreement with our previous investigations when we studied how the reaction temperature during the addition of the TFA influenced the yield in the TB condensation.[28a] Interestingly, conducting the reaction at –10 °C in the presence of 10 equiv. of NH4Cl led to an increase in the proportions of the diastereomers containing clefts – 51a and 51c – relative to the cavity compound 51b, with no decrease in total yield, suggesting a templating effect of the ammonium ion through cation–π interactions with the aromatic rings during the reaction to form the fused tris-TB analogues. Recent results from Král and Dolenský,[70] however, might point towards anion–π interactions being responsible for this effect.


www.eurjoc.org

7029


MICROREVIEW

Scheme 26. Synthetic route to the racemic linear tris-TB analogues 51a–c and schematic view of the three possible diastereomers.

Figure 16. A) ORTEP representation of the crystal structure of 51c. Hydrogen atoms have been omitted for clarity. There was residual electron density in the cavity formed by the molecules, probably due to the presence of disordered solvent molecules. B) ORTEP representation of the crystal structure of diastereomer 51a. Hydrogen atoms have been omitted for clarity. The data unambiguously show the molecular structure of 51c to be that of the syn-syn diastereomer and 51a to be the syn-anti diastereomer. C) View along the b axis of the packing of isomer 51c.

All three diastereomers 51a–c were separately subjected to isomerization studies under different acidic conditions and temperatures: each compound was dissolved in neat 7030

www.eurjoc.org

TFA or CHCl3/EtOH (2:1), the latter containing various amounts of HCl in dioxane (4 m) so that the resulting solutions were 3, 0.8, and 0.01 m in HCl. The solutions were stirred at 25, 60, and 95 °C for different periods of time. As a general trend, all three diastereomers 51a–c were reluctant to isomerize under the conditions investigated. From these results, we concluded that the syn-anti 51a is slightly more stable than the anti-anti 51b, which is in turn slightly more stable than syn-syn 51c. Caution must be exercised, however, due to the very low degrees of isomerization observed. We believe that fused tris-TB analogues 51a–c display reluctance to invert at their stereogenic nitrogen atoms under various sets of acidic conditions and temperatures because of the presence of the blocking groups at the ortho-positions (4,10-positions) in their aromatic rings, hampering the inversion of configuration. Such an explanation has also been put forward by Lenev for the slow racemization of optically pure 4,10-dimethyl-substituted TB analogues.[79] This stability towards inversion of configuration makes the above TB analogues attractive building blocks for use in dilute acids and they should be compared to the ethano-TB analogues, which are also configurationally stable.

Heterocyclic Analogues of Tröger’s Base Formation of Tröger⬘s Base Aromatic Heterocycles in OneStep Condensation Reactions The unique structural architecture of TB in combination with the acceptor and donor properties of aromatic hetero-


Eur. J. Org. Chem. 2012, 7015–7041


cycles yield interesting structures that could be beneficial in many fields of application, such as medicinal chemistry. It was not until 1991 that the first aromatic heterocyclic analogue of TB was synthesized, by Yashima and coworkers.[80] They condensed 5-amino-1,10-phenanthroline (52, Figure 17) with formaldehyde (aq. HCl in EtOH) to afford a TB analogue in 22 % yield. The next milestone in the production of heterocyclic TB analogues came in 1993, when Pardo and co-workers,[25a] and Johnson and coworkers reported azolyl-based (compounds 54–56, Figure 17) and 3-picoline-based (compound 64, Figure 17) TB analogues,[37b] respectively, providing early rare examples of direct syntheses of TB analogues bearing σ-electron-withdrawing groups. Later, in 1997, Cudero and co-workers synthesized the first TB analogue in which a heterocyclic aromatic framework was fused directly to the methanodiazonic ring (57, Figure 17).[25b] In that paper they compared the reactivities of π-excessive (azoles) and π-deficient (azines) heterocycles, with the use of either HCl and formaldehyde or of TFA and hexamethylenetetramine as acid and methylene source, respectively (compounds 53, 57–63, Figure 17). Interestingly, when TFA was used as the acid and 57 as

starting material, no reaction occurred. As expected, all the azines were recovered from the reaction mixtures unchanged, due to the low reactivities of π-deficient heterocycles towards electrophilic attack at carbon atoms. Abonia and co-workers later investigated the direct fusion of heterocyclic amines to the methanodiazocine ring, using starting materials 71 and 72 together with acetic acid and formaldehyde.[81] They suggested that the low yields were probably due to the formation of stable side products that did not react further. Some seven years later, Wu and co-workers introduced new members of the class of fivemembered-ring heterocyclic-ring-fused TB analogues by a similar synthetic approach (from compounds 72, 85–87, Figure 17).[82] Since then, numerous heterocyclic TB analogues, summarized in Figure 17, have been produced in one-step condensations.[18e,25,71,81–83] Some of them were synthesized by use of biologically active materials containing free aromatic amine groups (from compounds 68, 73– 79;[84] 81–83,[83c] Figure 17) as starting materials in condensations to afford the TB analogues. In addition to the heterocycles mentioned, Valik and co-workers have produced series of pyrrole TB analogues utilizing the structural features of known antibiotics (vide supra).[85]

Figure 17. Examples of aromatic heterocyclic aniline derivatives that have been condensed to afford TB analogues. The arrows show the sites of reaction with formaldehyde equivalents.[83a,83b] Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7031


MICROREVIEW

Scheme 27. A) Formation of an aromatic heterocyclic analogue through functionalization of the aromatic group of the TB intermediate. B) One-step condensation to form the same analogue.

The syntheses of heterocyclic derivatives through a onestep condensation reactions in which the aromatic heterocyclic molecule is directly fused into the methanodiazonic ring still need more study to provide an appropriate general synthetic methodology. The TFA method that we developed, for instance, cannot be generally applied in the synthesis of heterocyclic TB analogues. The synthetic approach for each analogue thus has to be developed individually to produce high yields. The scope for improvements in the synthesis of heterocyclic TB analogues is illustrated in a recent publication by Dolenský and co-workers in which they fused the heterocyclic [2-aminotetraarylporphyrinato(2–)]nickel system directly to the methanodiazonic ring.[86] They manipulated the condensation by varying the formaldehyde source and, particularly, the solvent, which made it possible to drive the reaction selectively towards the desired product, a [tetraarylporphyrinato(2–)]nickel TB analogue. Formation of Heterocyclic Tröger’s Base Analogues after TB Condensation As discussed above, the π-deficient properties of many heterocycles hamper the production of heterocyclic TB analogues by one-step condensation protocols. A solution can be found, however, by attaching the heterocyclic functional group to a formed TB analogue. Goswani and Ghosh synthesized aminopyridine analogues by both approaches – Routes A and B in Scheme 27 – to form the hydrogen-bond receptor TB analogue 88.[87] Approaches similar to Route A (Scheme 27) have been developed for production of heterocyclic TB analogues (e.g., bis-pyridinium analogues).[88] In addition, benzodiazolium TB analogues have been synthesized by an approach in which the heterocyclic functional group is introduced after the TB condensation has taken place.[89]

Applications of Tröger’s Base Analogues The remarkable structural features of TB were underexploited until the 1980s, before which it was used solely as a standard for the evaluation of new chiral chromatographic 7032

www.eurjoc.org

techniques. However, the groundbreaking report by Wilcox in 1985,[16] introducing new TB analogues as a potential unit for chiral hosts and metal ion ligands, represented the resurrection of TB as a hot research topic. Since then, TB analogues have found applications as building blocks in the fields of, among others, supramolecular chemistry, molecular recognition, catalysis, enzyme inhibitors, and new materials.

Hydrogen-Bonding Tröger’s Base Analogues as Receptors In 1989, Wilcox and Adrian described the synthesis of carboxylic-acid-substituted TB derivatives (compounds 89, Figure 18) designed to form four hydrogen bonds simultaneously with cyclic urea derivatives and adenine base moieties.[90] The binding process was studied by NMR and UV/ fluorescence spectroscopy techniques in solvents with different polarities and hydrogen bonding capabilities. The same system was used for determining the effect of water on the binding abilities of similar hosts.[18d]

Figure 18. Diacid- and aminopyridene-derived TBs as hydrogenbonding receptors.

The design and the synthesis of the TB amidopyridine analogues 88 and 90 were reported by Goswami and Ghosh


Eur. J. Org. Chem. 2012, 7015–7041


(Figure 18).[83d,87] Complexation studies of the recognition of dicarboxylic acids of different lengths showed the TB analogue 88 to be a selective host for suberic acid in the presence of other α,ω-diacids. The more flexible TB receptor 90 exhibited weaker binding towards the same dicarboxylic acid and the loss of binding selectivity. Our group investigated the use of the C2-symmetric bis(crown ether) TB analogue 91 (Figure 19) for recognition studies of achiral and chiral primary bisammonium salts.[91] Experiments conducted on the dihydrochloride salts of α,ωdiaminoalkanes showed very similar binding affinities for bisammonium salts n = 6–8. Enantiomeric discrimination by receptor 91 was explored with the dihydrochloride salts of the methyl esters of l-cystine and l-lysine. The diastereoselectivity of the complexation for l-cystine was estimated from the 1H NMR spectra to be 62:38, whereas no enantiomeric discrimination was observed for l-lysine.

cycle of this series contained a diphenylmethane unit linked to a TB with an alkyl tether containing secondary ammonium groups (compound 94; Figure 21). The association of TB macrocycle 94 with several benzenoid substrates was analyzed in acidic aqueous media and showed a preference for aromatic molecules with electron-withdrawing groups.[94a,94b]

Figure 19. A TB-bis(crown ether) receptor for bisammonium salts.

Macrocyclic Tröger’s Base Analogues as Receptors The use of macrocyclic frameworks to construct receptors is especially suitable for relatively unfunctionalized TB analogues, with the cavity of the receptor providing scope for solvophobic effects and van der Waals interactions. The use of a TB scaffold in the synthesis of cyclophanes was demonstrated by Inazu and Fukae in 1984.[92] The dimeric TB analogue 92 (Figure 20) was prepared in 45 % yield and the meso and rac forms were isolated by fractional recrystallization. However, the limited size of the cavity hampered any inclusion phenomena. The same research group was the first to incorporate crown ethers into TB analogues (compounds 93, n = 3–5; Figure 20).[93] Despite the differences in length of the polyethylene chains, binding studies revealed similar binding affinities for all cations studied, with the high degree of flexibility impeding any selectivity.

Figure 20. The first macrocyclic and crown-ether-derived TB analogues, respectively.

Wilcox and co-workers developed a series of water-soluble cyclophanes based on the TB scaffold as chiral receptors for small and neutral organic molecules.[94] The first macroEur. J. Org. Chem. 2012, 7015–7041

Figure 21. TB-based cyclophanes.

The optically pure macrocyclic TB bis-sulfone 95 demonstrated binding of isomeric menthols with reasonable selectivity (Figure 21).[94c,94d] In a more recent contribution, a cyclophane consisting of two TB moieties and bearing a mercaptoimidazole group on the alkyl linker was designed to bind biologically relevant substrates. 1H NMR studies showed that TB analogue 96 preferably bound 4-nitrophenyl phosphate over its non-phosphate analogue (Figure 21).[94e]

Porphyrin-Derived TB Analogues as Receptors The coordination abilities of metal-containing porphyrins have also been exploited when fused to a TB scaffold. Crossley and co-workers synthesized different TB analogues in which the methanodiazocine ring is fused to two tetraarylporphyrins, forming a well-defined chiral cleft molecule.[95] These series of bisporphyrino-TB analogues with inserted metal ions exhibited strong affinities with different selectivities towards diamine guests such as α,ω-diaminoalkanes of different lengths,[95a] as well as towards histidine and lysine esters.[95b] TB analogue 97 (Figure 22), in addition to binding simple diaminoalkanes, encapsulated a tetramine dendrimer, forming a spherical cage.[95c] Specifically, in a recent report based on similar structures, the Crossley group demonstrated ditopic binding of an α,ω-dicarboxylic acid in the cavity when Sn(OH)2 was coordinated to the porphyrin rings.[96]


www.eurjoc.org

7033


MICROREVIEW

up to 67 % in the amine-promoted aziridation of chalcones,[104] whereas (–)-1 gave rise to a 57 % ee when used as an additive in the 1,4-addition of alkyllithium species to α,β-unsaturated tert-butyl esters.[105]

Figure 22. A porphyrin-containing TB analogue.

Tröger’s Base Analogues as a Molecular Torsion Balances Arguably the most elegant application of TB analogues is in the molecular torsion balances designed by Wilcox and co-workers. Use of structures such as TB analogues 98 (Figure 23), allowed for the quantification of weak forces (e.g., aromatic edge-to-face interactions,[97] or CH–π interactions[98]) believed to play an important role in protein folding. Later improvements resulted in water-soluble structures, thus avoiding the need for corrections for the change in the direction of the dipoles between folded and unfolded conformers.[99] Further developments by Diederich and coworkers provided molecular balances for study of the interactions between “organic” fluorine atoms and an amide group.[100] This work was followed by a new generation of molecular torsion balances based on an indole fragment, which provided evidence of the existence of a favorable orthogonal dipolar interaction between the C–F bond and an amide carbonyl group.[101]

Tröger’s Base Analogues in Selective Catalysis The rigid chiral structure of TB and the presence of a transition-metal binding sites (the nitrogen atoms) make TB and its analogues good candidates as ligands in asymmetric catalysis. Applications in this field, however, have to date barely been investigated. The first use of TB as a ligand for catalytic applications was demonstrated in the hydrosilylation of terminal alkynes.[102] The complex TB·2 RhCl3 (99, M = Rh, Figure 24) showed catalytic activity, giving rise to the thermodynamically less stable cis products with selectivity up to 95 %. Enantiopure (+)-1 was employed as an additive in the Pt/Al2O3 hydrogenation of ethyl pyruvate, resulting in ethyl lactate with 65 % ee.[103] The same enantiomer gave ees of

Figure 24. Examples of TB analogues used as ligands in catalysis.

Different TB analogues have been screened as inductors of asymmetry in catalytic processes. Harmata studied the effect of different TB chiral ligands in the additions of Et2Zn to aromatic aldehydes.[61b] Although enantiopure parent TB (1) gave poor enantioselectivity in the resulting alcohol, 6-exo-substituted TB analogue 100 (Figure 24) afforded up to 86 % ee in the alcohol product. In addition, substituted pyrazole analogues of TB (compounds 101; Figure 24) were used as organocatalysts in one-pot Mannich reactions between aromatic aldehydes, aniline derivatives, and cyclohexanone in aqueous media, resulting in very good yields and remarkable anti/syn stereoselectivities in the products.[82] The dimeric dipalladium complex 102, based on the same pyrazole TB architecture, was used as a catalyst in the Mizoroki–Heck C–C coupling reaction, displaying high catalytic activity and considerable selectivity towards the formation of trans-stilbenes with 89–93 % conversion (Figure 24).[106] TB analogues containing thiourea chains linked to the aromatic rings showed activity as catalysts in Michael additions of malonate derivatives to nitroolefins, but no enantioselectivity was observed.[107]

Figure 23. TB molecular torsion balances for the quantification of aromatic edge-to-face interactions.[97,98,101] 7034

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


In recent years TB-embedded materials have been employed in different heterogeneous catalytic processes with encouraging results. A nanoporous polymer containing covalently bonded TB analogues, for instance, showed reasonable catalytic activity in the addition of diethylzinc to 4chlorobenzaldehyde, affording 1-(4-chlorophenyl)propan-1ol in 60–56 % yield.[108] The catalytic efficiency was maintained after at least three catalytic cycles. Investigations conducted with mesoporous organosilicas containing TB (1) revealed their activities in promoting reactions such as the Knoevenagel reaction, S-arylation of aryl iodides, and azide–alkyne cycloaddition.[109]

Figure 26. The chirality of acridine TB analogue 106 in relation to different DNA conformers.[114]

Medicinal Properties of Heterocyclic Tröger’s Base Analogues TB analogues containing aromatic heterocyclic rings have shown great ability to interact with DNA. Yashima and co-workers introduced the first TB analogue of this family (compound 103; Figure 25), a phenanthroline analogue that exhibited higher affinity towards DNA than the parent 1,10-phenanthroline.[80] Demeunynck and Lhomme have widely explored this field.[18e,83f,84a,110] The proflavine TB analogue 104 (Figure 25) is a representative example. Compound (–)-104 binds in a sequence-selective fashion to calf thymus B-DNA.[83f,84a] Sequences with motifs such as 5⬘-GTT·AAC or 5⬘-ATGA·TCAT bound preferably to compound (–)-104. In another example, a DNA interaction assay involving the non-symmetric proflavine-phenanthroline TB 105[110c] (Figure 25) revealed that the acridine ring intercalates between the DNA pairs whereas the phenanthroline moiety resides in one groove of the DNA.

Figure 27. Molecular modeling of binding of acridine TB analogue 106 to DNA. One of the acridine rings intercalates between neighboring base pairs whereas the other acridine ring is positioned in the minor or major groove.[114]

Figure 25. Phenanthroline and proflavine analogues of TB.

The well-known B-DNA is a double helix and is twisted in a right-hand direction. However, studies have shown the presence of another conformation, Z-DNA, in which the helix is twisted in the left-hand direction (Figure 26). The existence and biological function of this conformer in vivo is still unclear, however, and requires further studies.[111] Historically, only chiral metal complexes have been investigated for “enantioselective”[112] recognition of DNA conformations. However, when coupled to acridine, the geometry of TB 106 (Figure 26) gives rise to a helical shape of the molecule, which can be either similar or opposite to that of DNA, thus resulting in “enantioselective” binding (Figures 27 and 28).[113] Eur. J. Org. Chem. 2012, 7015–7041

Figure 28. Molecular modeling of binding of acridine TB analogue 106 to DNA. Both acridine rings are positioned in one of the grooves.[114]

Tatibuet and co-workers reported that the (2)-(7R,17R) enantiomer of an acridine analogue of TB[113] – compound (–)-106 – showed preferential binding to calf-thymus B-


www.eurjoc.org

7035


MICROREVIEW DNA (i.e., the conformer in which the double helix is twisted in a right-hand direction) and hence the potential to serve as an “enantioselective” molecular DNA probe. Molecular modeling predicted selective binding of acridine TB analogues to DNA. This can occur by different mechanisms, as illustrated in Figures 27 and 28.[115] In a recent work, Veale and Gunnlaugsson reported the synthesis of a small library of fluorescent 1,8-naphthalamide analogues of TB (compounds 107; Figure 29).[83h] It was predicted that at physiological pH the cationic amino termini of TB analogues 107 would strongly bind to the phosphate backbone of DNA, and all three compounds indeed showed high affinities. Fluorescence imaging studies demonstrated the rapid uptake of TB analogues 107 by cancer cells and that the compounds become localized within the nuclei.

of the N-methylpyrrole TB analogues 109 (Figure 30), which mimic the building blocks of natural antibiotics such as distamycin and netopsin.[85b] The fusion of the N-methylpyrrole rings with the methanodiazocine ring resulted in compounds with binding ability to DNA, although to the best of our knowledge no biological activity studies for those compounds have been reported. Recently, TB analogues bearing the cytotoxic acronycine motif, such as compound 110 (Figure 30), have been reported to be cytotoxic against L-1210 leukemia and KB-3–1 solid tumor cell lines.[116] In 2012, Ananya and co-workers published an investigation about the use of benzimidazole-derived TB analogues 111 (Figure 30) as cancer treatment candidates, based on their inhibition of teleomerase activity.[117] Inhibition of telomerase activity leads to cell death. In cancer cells the telomerase is over-expressed whereas in normal somatic cells it is has undetectable activity, and this is the basis for cancer treatment based on telomerases. The TB analogues 110 inhibited human telomerase activity by acting as G-quadruplex ligands that fold guanidine-rich DNA into G4DNA (G-quadruplex DNA) and hence inhibit telomerase activity.

Figure 29. Fluorescent analogues of TB.

Other Applications of Tröger’s Base

TB analogues containing two 3-picolyl groups (compounds 108, Figure 30) showed inhibition activity towards the enzyme thromboxane A2 synthase (TxA2).[37b] The employment of the picoline functional group on the TB structure was based on the same structural characteristics as the known TxA2 inhibitor sodium furegrelate (Figure 30), which also bears a 3-picoline functional group. An attempt to increase the activity by adding substituents in the diazocine methylene bridge in fact significantly decreased the activity. Dolenský and co-workers described the synthesis

The rigid V-shaped geometry of Tröger’s base has been exploited in the construction of elaborated metallomacrocycles. In the first example of these structurally beautiful complexes, reported by Mirkin, a phosphane/thioether ligand containing the TB motif formed a dimeric metallocycle when coordinated to CuI, whereas with RhI a mixture of trimer and tetramer was obtained (Figure 31).[118] TB analogues have been studied in supramolecular selfassembly, a broad field ranging from the DNA helix to inclusion compounds. It is generally regarded as the process

Figure 30. Biologically active sodium furegrelate, together with the pyridyl- and acronycine-, N-methylpyrroline-, and benzimidazolebased TB analogues. 7036

www.eurjoc.org


Eur. J. Org. Chem. 2012, 7015–7041


Figure 31. Left) TB ligand, and right) solid-state structure of its RhI-complexed tetramer. Hydrogen atoms are omitted for clarity.

Figure 32. Left) TB ligand, and right) the solid-state structure of the PtII-coordinated cage. Hydrogen atoms are omitted for clarity.

through which individual molecules form defined aggregates, which can then self-organize to form higher-order structures. Lützen has made two very important observations in the field of self-assembled TB metallohelicates. The first relates to the presence or absence of diastereoselectivity in the self-assembly process. He showed that several TB ligands bearing 2,2⬘-bipyridine or 2-pyridylmethanimine moieties self-assembled into dinuclear double-stranded helicates upon coordination to AgI and CuI.[29c] The complex-

ation of the metal ion by the racemic ligand is diastereospecific in a self-recognition manner, giving rise to diastereomerically pure complexes.[29c,119] In contrast, coordination to FeIII and ZnII results in triple-stranded helicates. Whereas FeIII resulted in diastereomerically pure complexes, ZnII displayed no diastereoselectivity.[29c] In a similar approach, Lützen and our own group built a metallomacrocycle from a racemic bis(4-pyridyl-alkyne)-derived TB analogue (Figure 32) and (dpp)Pt(OTf)2 in a diastereoselective

Figure 33. TB analogues used in optical and optoelectronic applications.

Eur. J. Org. Chem. 2012, 7015–7041


www.eurjoc.org

7037


MICROREVIEW self-discrimination process, which formed exclusively the heterochiral R,R,S,S isomer.[88] Our group showed that the resulting cage (Figure 32, right) displayed greatly enhanced fluorescence relative to the TB ligand alone. The fluorescence was quenched upon addition of C60. The quenching might be due to the interaction of π-accepting C60 and the lone pair of the TB nitrogen atoms, because no evidence of C60 inclusion was found. Lützen’s second observation, this time together with Piguet, was in an experimental and theoretical analysis of metal–metal interactions in solution. They point to two effects that oppose each other: Coulombic interactions produce large intermetallic repulsion at short distances, and solvation effects result in large intermetallic attractions for small pseudo-spherical ions with short intermetallic separations.[120] Incipient applications of TB analogues as optical and optoelectronic materials have been reported in recent years and were recently reviewed by Yuan and co-workers.[121] The bis-pyridinium-derived TB analogue 112 (Figure 33) exhibits aggregation-induced light emission in the solid state but is virtually non-emissive in solution, a feature not observed in a planar counterpart.[122] Photophysical and electroluminescent properties have been examined in fluorene-derived TB analogues.[123] Compounds such as TB analogue 113 (Figure 33) exhibit strong fluorescence emission in dilute solutions and in aggregated states. Organic lightemitting diodes (OLEDs) fabricated with these TB analogues show high brightness, high efficiency, and low turnon voltage. Benzothiazolium- and para-nitrophenyl-derived TB analogues 114 and 115, respectively (Figure 33), show second-order nonlinear optic properties, the former being the more active.[89,124]

Conclusions Since the beginning of the 21st century synthetic TB chemistry has developed rapidly, allowing for the positioning of functional groups in virtually any position of the TB core and for the formation of TB analogues fused with phenylene rings. In addition, resolution of TB analogues into their antipodes by chiral HPLC on a semipreparative scales has become almost routine. The essential properties of TB – the chiral, rigid, and aromatic cavity, together with the synthetic entries available to the scientific community – are bound to inspire new applications of TB analogues. One could argue that no real groundbreaking application of molecules containing the TB skeleton has yet arrived, but with the molecular torsion balance being a possible exception. However, the reason for this shortage of applications might be that the synthetic tools needed to make new building blocks for applications are quite recent. We argue that the properties of the TB skeleton discussed above are especially suited for applications in the fields of asymmetric catalysis and of sensors for biomolecules with diagnostic applications. Hopefully, the very recently introduced methodologies for the insertion of endo-substituted groups reaching 7038

www.eurjoc.org

the chiral cavity of TB, the variety of methods to modify the bridging methano bridge of TB now available, and finally the introduction of configurationally stable ethanobridged or 4,10-disubstituted TB analogue might inspire new entries into the applied fields.

Acknowledgments K. W. thanks the Swedish Research Council, The Royal Physiographic Society in Lund, the Crafoord Foundation, and the Swedish Foundation for Strategic Research for financial support. Ö. V. R. thanks the Olle Engkvist Byggmästare foundation for a postdoc fellowship. We would like to acknowledge Martine Demeunynck (Figures 26, 27, and 28),[114] the Taylor & Francis publication (Figure 12),[74] and Wiley publications (Figure 15) for permission to use these illustrations. Furthermore, we thank Per Ola Norrby, University of Gothenburg, and Anders Sundin, Lund University, for discussions and calculations, respectively, relating to the anomeric effect in TB. [1] [2] [3] [4] [5]

J. Tröger, J. Prakt. Chem. 1887, 36, 225–245. M. A. Spielman, J. Am. Chem. Soc. 1935, 57, 583–585. E. Wagner, J. Am. Chem. Soc. 1935, 57, 1296–1298. V. Prelog, P. Wieland, Helv. Chim. Acta 1944, 27, 1127–1134. S. B. Larson, C. S. Wilcox, Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1986, 42, 224–227. [6] S. H. Wilen, J. Z. Qi, P. G. Williard, J. Org. Chem. 1991, 56, 485–487. [7] a) Y. Miyahara, K. Izumi, A. A. Ibrahim, T. Inazu, Tetrahedron Lett. 1999, 40, 1705–1708; b) S. Sergeyev, F. Diederich, Chirality 2006, 18, 707–712; c) U. Kiehne, T. Bruhn, G. Schnakenburg, R. Fröhlich, G. Bringmann, A. Lützen, Chem. Eur. J. 2008, 14, 4246–4255; d) D. Didier, B. Tylleman, N. Lambert, C. M. L. Vande Velde, F. Blockhuys, A. Collas, S. Sergeyev, Tetrahedron 2008, 64, 6252–6262; e) C. Benkhaeuser-Schunk, B. Wezisla, K. Urbahn, U. Kiehne, J. Daniels, G. Schnakenburg, F. Neese, A. Lützen, ChemPlusChem 2012, 77, 396–403. [8] A. Greenberg, N. Molinaro, M. Lang, J. Org. Chem. 1984, 49, 1127–1130. [9] a) Y. Hamada, S. Mukai, Tetrahedron: Asymmetry 1996, 7, 2671–2674; b) D. A. Lenev, D. G. Golovanov, K. A. Lyssenko, R. G. Kostyanovsky, Tetrahedron: Asymmetry 2006, 17, 2191– 2194; c) C. Michon, A. Sharma, G. Bernardinelli, E. Francotte, J. Lacour, Chem. Commun. 2010, 46, 2206–2208. [10] O. Trapp, V. Schurig, J. Am. Chem. Soc. 2000, 122, 1424–1430. [11] Á. Révész, D. Schröder, T. A. Rokob, M. Havlík, B. Dolenský, Angew. Chem. Int. Ed. 2011, 50, 2401–2404. [12] B. M. Wepster, Recl. Trav. Chim. Pays-Bas 1953, 72, 661–672. [13] B. M. Wepster, Recl. Trav. Chim. Pays-Bas 1952, 71, 1159– 1170. [14] Comment: The change in bond length between the respective nitrogen atom and the methylene carbon is more pronounced than other changes in bond lengths upon protonation of one of the nitrogen atoms of TB. [15] Calculations: DFT calculations were at the B3LYP/6–31G* level with use of the SM8 solvation model for EtOH. [16] C. S. Wilcox, Tetrahedron Lett. 1985, 26, 5749–5752. [17] a) F. Vögtle, Fascinating molecules in Organic Chemistry, Wiley, Chichester, UK, 1992; b) B. G. Bag, Curr. Sci. 1995, 68, 279– 288; c) M. Demeunynck, A. Tatibouët, in: Progress in Heterocyclic Chemistry, vol. 11 (Eds.: W. G. Gordon, L. G. Thomas), Elsevier, 1999, pp. 1–20; d) M. Valík, R. M. Strongin, V. Král, Supramol. Chem. 2005, 17, 347–367; e) B. Dolenský, J. Elguero, V. Král, C. Pardo, M. Valík, in: Adv. Heterocycl. Chem. vol. 93 (Ed.: F. R. S. Alan, R. Katritzky), Academic Press, 2007, pp. 1– 56; f) S. Sergeyev, Helv. Chim. Acta 2009, 92, 415–444.


Eur. J. Org. Chem. 2012, 7015–7041

Tröger’s Base’s 125th Anniversary: Synthesis and Applications of Analogues [18] a) T. R. Miller, E. C. Wagner, J. Am. Chem. Soc. 1941, 63, 832– 836; b) M. Häring, Helv. Chim. Acta 1963, 46, 2970–2982; c) B. G. Bag, U. Maitra, Synth. Commun. 1995, 25, 1849–1856; d) J. C. Adrian, C. S. Wilcox, J. Am. Chem. Soc. 1992, 114, 1398– 1403; e) H. Salez, A. Wardani, M. Demeunynck, A. Tatibouet, J. Lhomme, Tetrahedron Lett. 1995, 36, 1271–1274; f) A. Hansson, J. Jensen, O. F. Wendt, K. Warnmark, Eur. J. Org. Chem. 2003, 3179–3188. [19] Z. Li, X. Xu, Y. Peng, Z. Jiang, C. Ding, X. Qian, Synthesis 2005, 1228; Z. Li, X. Xu, Y. Peng, Z. Jiang, C. Ding, X. Qian, Synthesis 2005, 1230. [20] D. P. Becker, P. M. Finnegan, P. W. Collins, Tetrahedron Lett. 1993, 34, 1889–1892. [21] a) S. Satishkumar, M. Periasamy, Tetrahedron: Asymmetry 2006, 17, 1116–1119; b) S. Satishkumar, M. Periasamy, Tetrahedron: Asymmetry 2009, 20, 2257–2262. [22] H. Wu, P. Zhang, Y. Shen, F.-R. Zhang, Y. Wan, D.-Q. Shi, Synlett 2007, 336–338. [23] M. D. H. Bhuiyan, A. B. Mahon, P. Jensen, J. K. Clegg, A. C. Try, Eur. J. Org. Chem. 2009, 687–698. [24] E. Vardelle, A. Martin-Mingot, M.-P. Jouannetaud, J.-C. Jacquesy, J. Marrot, Tetrahedron Lett. 2009, 50, 1093–1096. [25] a) L. Cerrada, J. Cudero, J. Elguero, C. Pardo, J. Chem. Soc., Chem. Commun. 1993, 1713–1714; b) J. Cudero, C. Pardo, M. Ramos, E. Gutierrez-Puebla, A. Monge, J. Elguero, Tetrahedron 1997, 53, 2233–2240. [26] J. Jensen, K. Wärnmark, Synthesis 2001, 1873–1877. [27] A. Hansson, J. Jensen, O. F. Wendt, K. Wärnmark, Eur. J. Org. Chem. 2003, 3179–3188. [28] a) J. Jensen, M. Strozyk, K. Wärnmark, J. Heterocycl. Chem. 2003, 40, 373–375; b) S. Sergeyev, M. Schär, P. Seiler, O. Lukoyanova, L. Echegoyen, F. Diederich, Chem. Eur. J. 2005, 11, 2284–2294; c) S. P. Bew, L. Legentil, V. Scholier, S. V. Sharma, Chem. Commun. 2007, 389–391. [29] a) M. D. H. Bhuiyan, K.-X. Zhu, P. Jensen, A. C. Try, Eur. J. Org. Chem. 2010, 4662–4670; b) Q. M. Malik, S. Ijaz, D. C. Craig, A. C. Try, Tetrahedron 2011, 67, 5798–5805; c) U. Kiehne, T. Weilandt, A. Lützen, Org. Lett. 2007, 9, 1283–1286. [30] D. Didier, S. Sergeyev, Eur. J. Org. Chem. 2007, 3905–3910. [31] W. V. Farrar, J. Appl. Chem. 1964, 14, 389–399. [32] E. C. Wagner, J. Org. Chem. 1954, 19, 1862–1881. [33] C. A. M. Abella, M. Benassi, L. S. Santos, M. N. Eberlin, F. Coelho, J. Org. Chem. 2007, 72, 4048–4054. [34] Y. Wan, R. Yuan, W.-C. Zhang, Y.-h. Shi, W. Lin, W. Yin, R.c. Bo, J.-j. Shi, H. Wu, Tetrahedron 2010, 66, 3405–3409. [35] T. H. Webb, C. S. Wilcox, J. Org. Chem. 1990, 55, 363–365. [36] F. C. Cooper, M. W. Partridge, J. Chem. Soc. (resumed) 1955, 991–994. [37] a) F. C. Cooper, M. W. Partridge, J. Chem. Soc. (resumed) 1957, 2888–2893; b) R. A. Johnson, R. R. Gorman, R. J. Wnuk, N. J. Crittenden, J. W. Aiken, J. Med. Chem. 1993, 36, 3202–3206. [38] U. Maitra, B. G. Bag, J. Org. Chem. 1992, 57, 6979–6981. [39] U. Maitra, B. G. Bag, P. Rao, D. Powell, J. Chem. Soc. Perkin Trans. 1 1995, 2049–2056. [40] a) M. D. H. Bhuiyan, P. Jensen, A. C. Try, Acta Crystallogr., Sect. E: Struct. Rep. Online 2010, 66, O4393–U1521; b) J. Jensen, M. Strozyk, K. Wärnmark, Synthesis 2002, 2761–2765; c) U. Kiehne, T. Weilandt, A. Luetzen, Org. Lett. 2007, 9, 1283– 1286. [41] U. Kiehne, A. Lützen, Synthesis 2004, 1687; U. Kiehne, A. Lützen, Synthesis 2004, 1695. [42] a) F. Hof, M. Schär, D. M. Scofield, F. Fischer, F. Diederich, S. Sergeyev, Helv. Chim. Acta 2005, 88, 2333–2344; b) C. Solano, D. Svensson, Z. Olomi, J. Jensen, O. F. Wendt, K. Wärnmark, Eur. J. Org. Chem. 2005, 3510–3517. [43] D. Didier, S. Sergeyev, Tetrahedron 2007, 63, 3864–3869. [44] J. Artacho, P. Nilsson, K. E. Bergquist, O. F. Wendt, K. Wärnmark, Chem. Eur. J. 2006, 12, 2692–2701. [45] J. Artacho, K. Wärnmark, Synthesis 2009, 3120–3126. Eur. J. Org. Chem. 2012, 7015–7041

[46] J. Jensen, J. Tejler, K. Wärnmark, J. Org. Chem. 2002, 67, 6008– 6014. [47] S. Sergeyev, F. Diederich, Angew. Chem. 2004, 116, 1770; Angew. Chem. Int. Ed. 2004, 43, 1738–1740. [48] Y. Ishida, H. Ito, D. Mori, K. Saigo, Tetrahedron Lett. 2005, 46, 109–112. [49] D. Didier, S. Sergeyev, Eur. J. Org. Chem. 2007, 3905–3910. [50] M. Faroughi, P. Jensen, A. C. Try, ARKIVOC (Gainesville, FL, U.S.) 2009, 269–280. [51] a) M. Kosugi, M. Kameyama, T. Migita, Chem. Lett. 1983, 927–928; b) A. S. Guram, S. L. Buchwald, J. Am. Chem. Soc. 1994, 116, 7901–7902; c) F. Paul, J. Patt, J. F. Hartwig, J. Am. Chem. Soc. 1994, 116, 5969–5970; d) J. Louie, J. F. Hartwig, Tetrahedron Lett. 1995, 36, 3609–3612; e) A. S. Guram, R. A. Rennels, S. L. Buchwald, Angew. Chem. 1995, 107, 1456; Angew. Chem. Int. Ed. Engl. 1995, 34, 1348–1350. [52] a) E. Weber, U. Müller, D. Worsch, F. Vögtle, G. Will, A. Kirfel, J. Chem. Soc., Chem. Commun. 1985, 1578–1580; b) D. R. Bond, J. L. Scott, J. Chem. Soc. Perkin Trans. 2 1991, 47–51; c) C. Michon, M.-H. Gonçalves-Farbos, J. Lacour, Chirality 2009, 21, 809–817. [53] M. Periasamy, S. Suresh, S. Satishkumar, Tetrahedron: Asymmetry 2012, 23, 108–116. [54] M. B. Reddy, A. Manjula, B. V. Rao, B. Sridhar, Eur. J. Org. Chem. 2012, 312–319. [55] Q. M. Malik, A. B. Mahon, D. C. Craig, A. C. Try, Tetrahedron 2011, 67, 8509–8514. [56] A. B. Mahon, D. C. Craig, A. C. Try, Synthesis 2009, 636; A. B. Mahon, D. C. Craig, A. C. Try, Synthesis 2009, 642. [57] K. Y. Lee, S. Gowrisankar, J. N. Kim, Synlett 2006, 1389–1393. [58] D. A. Lenev, I. I. Chervin, K. A. Lyssenko, R. G. Kostyanovsky, Tetrahedron Lett. 2007, 48, 3363–3366. [59] A. Sharma, L. Guénée, J.-V. Naubron, J. Lacour, Angew. Chem. Int. Ed. 2011, 50, 3677–3680. [60] A. Sharma, C. Besnard, L. Guenee, J. Lacour, Org. Biomol. Chem. 2012, 10, 966–969. [61] a) M. Harmata, K. W. Carter, D. E. Jones, M. Kahraman, Tetrahedron Lett. 1996, 37, 6267–6270; b) M. Harmata, M. Kahraman, Tetrahedron: Asymmetry 2000, 11, 2875–2879. [62] a) W. Metlesics, R. Tavares, L. H. Sternbach, J. Org. Chem. 1966, 31, 3356–3362; b) A. Albert, H. Yamamoto, J. Chem. Soc. B 1966, 956–963. [63] M. Harmata, K.-O. Rayanil, C. L. Barnes, Supramol. Chem. 2006, 18, 581–586. [64] J. Artacho, E. Ascic, T. Rantanen, J. Karlsson, C.-J. Wallentin, R. Wang, O. F. Wendt, M. Harmata, V. Snieckus, K. Wärnmark, Chem. Eur. J. 2012, 18, 1038–1042. [65] F. K. Winkler, J. D. Dunitz, J. Mol. Biol. 1971, 59, 169–182. [66] A. J. Kirby, I. V. Komarov, K. Kowski, P. Rademacher, J. Chem. Soc. Perkin Trans. 2 1999, 1313–1316. [67] A. J. Kirby, I. V. Komarov, P. D. Wothers, N. Feeder, Angew. Chem. 1998, 110, 830; Angew. Chem. Int. Ed. 1998, 37, 785– 786. [68] J. Artacho, E. Ascic, T. Rantanen, C.-J. Wallentin, S. Dawaigher, K.-E. Bergquist, M. Harmata, V. Snieckus, K. Wärnmark, Org. Lett. 2012, 14, 4706–4709. [69] a) A. Marmur, J. Am. Chem. Soc. 2000, 122, 2120–2121; b) J. W. Steed, J. L. Artwood, Supramolecular Chemistry, 2nd ed., Wiley, Chichester, England, 2000. [70] B. Dolenský, M. Havlík, V. Král, Chem. Soc. Rev. 2012, 41, 3839–3858. [71] C. Pardo, E. Sesmilo, E. Gutiérrez-Puebla, A. Monge, J. Elguero, A. Fruchier, J. Org. Chem. 2001, 66, 1607–1611. [72] a) M. Valík, B. Dolensky, H. Petríckova, V. Král, Collect. Czech. Chem. Commun. 2002, 67, 609–621; b) T. Mas, C. Pardo, F. Salort, J. Elguero, M. R. Torres, Eur. J. Org. Chem. 2004, 1097–1104; c) T. Mas, C. Pardo, J. Elguero, ARKIVOC (Gainesville, FL, U.S.) 2004, 86–93. [73] M. Valík, J. Cejka, M. Havlík, V. Král, B. Dolenský, Chem. Commun. 2007, 3835–3837.


www.eurjoc.org

7039


MICROREVIEW [74] M. Havlik, B. Dolensky, J. Kessler, I. Cisarova, V. Kral, Supramol. Chem. 2012, 24, 127–134. [75] A. Hansson, T. Wixe, K.-E. Bergquist, K. Wärnmark, Org. Lett. 2005, 7, 2019–2022. [76] a) T. Mas, C. Pardo, J. Elguero, Mendeleev Commun. 2004, 14, 235–237; b) T. Mas, C. Pardo, J. Elguero, Helv. Chim. Acta 2005, 88, 1199–1207. [77] J. Kessler, M. Jakubek, B. Dolensky, P. Bour, J. Comb. Chem. 2012, 33, 2310–2317. [78] B. Dolenský, M. Valík, D. Sýkora, V. Král, Org. Lett. 2005, 7, 67–70. [79] D. A. Lenev, K. A. Lyssenko, D. G. Golovanov, V. Buss, R. G. Kostyanovsky, Chem. Eur. J. 2006, 12, 6412–6418. [80] E. Yashima, M. Akashi, N. Miyauchi, Chem. Lett. 1991, 1017– 1020. [81] R. Abonia, A. Albornoz, H. Larrahondo, J. Quiroga, B. Insuasty, H. Insuasty, A. Hormaza, A. Sanchez, M. Nogueras, J. Chem. Soc. Perkin Trans. 2 2002, 1588–1591. [82] H. Wu, X.-M. Chen, Y. Wan, L. Ye, H.-Q. Xin, H.-h. Xu, C.h. Yue, L.-L. Pang, R. Ma, D.-Q. Shi, Tetrahedron Lett. 2009, 50, 1062–1065. [83] a) M. Demeunynck, C. Fontaine, J. Lhomme, Magn. Reson. Chem. 1999, 37, 73–76; b) M. Demeunynck, C. Moucheron, A. K. De Mesmaeker, Tetrahedron Lett. 2002, 43, 261–264; c) T. Gaslonde, S. Leonce, A. Pierre, B. Pfeiffer, S. Michel, F. Tillequin, Tetrahedron Lett. 2011, 52, 4426–4429; d) S. Goswami, K. Ghosh, S. Dasgupta, J. Org. Chem. 2000, 65, 1907–1914; e) T. Kobayashi, T. Moriwaki, M. Tsubakiyama, S. Yoshida, J. Chem. Soc. Perkin Trans. 2 2002, 1963–1967; f) A. Tatibouët, M. Demeunynck, C. Andraud, A. Collet, J. Lhomme, Chem. Commun. 1999, 161–162; g) E. B. Veale, D. O. Frimannsson, M. Lawler, T. Gunnlaugsson, Org. Lett. 2009, 11, 4040–4043; h) E. B. Veale, T. Gunnlaugsson, J. Org. Chem. 2010, 75, 5513– 5525. [84] a) C. Bailly, W. Laine, M. Demeunynck, J. Lhomme, Biochem. Biophys. Res. Commun. 2000, 273, 681–685; b) B. Baldeyrou, C. Tardy, C. Bailly, P. Colson, C. Houssier, F. Charmantray, M. Demeunynck, Eur. J. Med. Chem. 2002, 37, 315–322; c) H. Salez, A. Wardani, M. Demeunynck, A. Tatibouet, J. Lhomme, Tetrahedron Lett. 1995, 36, 1271–1274; d) A. Tatibouet, M. Demeunynck, C. Andraud, A. Collet, J. Lhomme, Chem. Commun. 1999, 161–162. [85] a) M. Valik, B. Dolensky, E. Herdtweck, V. Kral, Tetrahedron: Asymmetry 2005, 16, 1969–1974; b) M. Valik, B. Dolensky, H. Petrickova, P. Vasek, V. Kral, Tetrahedron Lett. 2003, 44, 2083– 2086. [86] A. Tatar, B. Dolenský, H. Dvoráková, V. Král, Tetrahedron Lett. 2012, 53, 6015–6017. [87] S. Goswami, K. Ghosh, Tetrahedron Lett. 1997, 38, 4503–4506. [88] C. S. Arribas, O. F. Wendt, A. P. Sundin, C.-J. Carling, R. Wang, R. P. Lemieux, K. Warnmark, Chem. Commun. 2010, 46, 4381–4383. [89] V. Lemaur, J. Cornil, D. Didier, A. Mujawase, S. Sergeyev, Helv. Chim. Acta 2007, 90, 2087–2095. [90] J. C. Adrian, C. S. Wilcox, J. Am. Chem. Soc. 1989, 111, 8055– 8057. [91] A. P. Hansson, P.-O. Norrby, K. Wärnmark, Tetrahedron Lett. 1998, 39, 4565–4568. [92] M. Fukae, T. Inazu, J. Incl. Phenom. Macrocycl. Chem. 1984, 2, 223–229. [93] A. A. Ibrahim, M. Matsumoto, Y. Miyahara, K. Izumi, M. Suenaga, N. Shimizu, T. Inazu, J. Heterocycl. Chem. 1998, 35, 209–215. [94] a) C. S. Wilcox, M. D. Cowart, Tetrahedron Lett. 1986, 27, 5563–5566; b) M. D. Cowart, I. Sucholeiki, R. R. Bukownik, C. S. Wilcox, J. Am. Chem. Soc. 1988, 110, 6204–6210; c) T. H. Webb, H. Suh, C. S. Wilcox, J. Am. Chem. Soc. 1991, 113, 8554–8555; d) C. S. Wilcox, J. C. Adrian, T. H. Webb, F. J. Zawacki, J. Am. Chem. Soc. 1992, 114, 10189–10197; e) M. Miyake, C. S. Wilcox, Heterocycles 2002, 57, 515–522. 7040

www.eurjoc.org

[95] a) M. J. Crossley, T. W. Hambley, L. G. Mackay, A. C. Try, R. Walton, J. Chem. Soc., Chem. Commun. 1995, 1077–1079; b) M. J. Crossley, L. G. Mackay, A. C. Try, J. Chem. Soc., Chem. Commun. 1995, 1925–1927; c) J. N. H. Reek, M. J. Crossley, A. P. H. J. Schenning, A. W. Bosman, E. W. Meijer, Chem. Commun. 1998, 11–12. [96] P. R. Brotherhood, R. A. S. Wu, P. Turner, M. J. Crossley, Chem. Commun. 2007, 225–227. [97] S. Paliwal, S. Geib, C. S. Wilcox, J. Am. Chem. Soc. 1994, 116, 4497–4498. [98] E.-i. Kim, S. Paliwal, C. S. Wilcox, J. Am. Chem. Soc. 1998, 120, 11192–11193. [99] B. Bhayana, C. S. Wilcox, Angew. Chem. 2007, 119, 6957; Angew. Chem. Int. Ed. 2007, 46, 6833–6836. [100] F. Hof, D. M. Scofield, W. B. Schweizer, F. Diederich, Angew. Chem. 2004, 116, 5166; Angew. Chem. Int. Ed. 2004, 43, 5056– 5059. [101] F. R. Fischer, W. B. Schweizer, F. Diederich, Angew. Chem. 2007, 119, 8418; Angew. Chem. Int. Ed. 2007, 46, 8270–8273. [102] Y. Goldberg, H. Alper, Tetrahedron Lett. 1995, 36, 369–372. [103] B. Minder, M. Schürch, T. Mallat, A. Baiker, Catal. Lett. 1995, 31, 143–151. [104] Y.-M. Shen, M.-X. Zhao, J. Xu, Y. Shi, Angew. Chem. 2006, 118, 8173; Angew. Chem. Int. Ed. 2006, 45, 8005–8008. [105] F. Xu, R. D. Tillyer, D. M. Tschaen, E. J. J. Grabowski, P. J. Reider, Tetrahedron: Asymmetry 1998, 9, 1651–1655. [106] F. Cuenú, R. Abonia, A. Bolaños, A. Cabrera, J. Organomet. Chem. 2011, 696, 1834–1839. [107] D. Didier, S. Sergeyev, ARKIVOC (Gainesville, FL, U.S.) 2009, 124–134. [108] X. Du, Y. Sun, B. Tan, Q. Teng, X. Yao, C. Su, W. Wang, Chem. Commun. 2010, 46, 970–972. [109] a) J. R. Cabrero-Antonino, T. García, P. Rubio-Marqués, J. A. Vidal-Moya, A. Leyva-Pérez, S. S. Al-Deyab, S. I. Al-Resayes, U. Díaz, A. Corma, ACS Catal. 2011, 1, 147–158; b) E. Poli, E. Merino, U. Díaz, D. Brunel, A. Corma, J. Phys. Chem. C 2011, 115, 7573–7585. [110] a) A. Tatibouët, M. Demeunynck, J. Lhomme, Synth. Commun. 1996, 26, 4375–4395; b) A. Tatibouet, M. Demeunynck, H. Salez, R. Arnaud, J. Lhomme, C. Courseille, Bull. Soc. Chim. Fr. 1997, 134, 495–501; c) B. Baldeyrou, C. Tardy, C. Bailly, P. Colson, C. Houssier, F. Charmantray, M. Demeunynck, Eur. J. Med. Chem. 2002, 37, 315–322. [111] G. W. Gribble, T. L. Gilchrist, Progress in Heterocyclic Chemistry, vol. 11 (a critical review of the 1998 literature preceded by two chapters on current heterocyclic topics), Elsevier, 1999. [112] Comment: more correct: “conformationally” selective binding. [113] A. Tatibouët, M. Demeunynck, C. Andraud, A. Collet, J. Lhomme, J. Chem. Soc., Chem. Commun. 1991, 161–162. [114] A. Tatibouët, M. Demeunynck, J. Lhomme, P. Vatton, D. Charapoff, M. L. Dheu, in: Electronic Conference on Heterocyclic Chemistry, 29–24 July 1998, Electronic Conference, 1998. [115] Y. Coppel, C. Coulombeau, C. Colulombeau, J. Lhomme, M. L. Dheu-Andries, P. Vatton, J. Biomol. Struct. Dyn. 1994, 12, 637–653. [116] T. Gaslonde, S. Léonce, A. Pierré, B. Pfeiffer, S. Michel, F. Tillequin, Tetrahedron Lett. 2011, 52, 4426–4429. [117] A. Paul, B. Maji, S. K. Misra, A. K. Jain, K. Muniyappa, S. Bhattacharya, J. Med. Chem. 2012, 55, 7460–7471. [118] M. S. Khoshbin, M. V. Ovchinnikov, C. A. Mirkin, J. A. Golen, A. L. Rheingold, Inorg. Chem. 2006, 45, 2603–2609. [119] a) U. Kiehne, T. Weilandt, A. Lützen, Eur. J. Org. Chem. 2008, 2056–2064; b) T. Weilandt, U. Kiehne, J. Bunzen, G. Schnakenburg, A. Lützen, Chem. Eur. J. 2010, 16, 2418–2426; c) T. Weilandt, U. Kiehne, G. Schnakenburg, A. Luetzen, Chem. Commun. 2009, 2320–2322.


Eur. J. Org. Chem. 2012, 7015–7041

Tröger’s Base’s 125th Anniversary: Synthesis and Applications of Analogues [120] N. Dalla Favera, U. Kiehne, J. Bunzen, S. Hytteballe, A. Lützen, C. Piguet, Angew. Chem. 2010, 122, 129; Angew. Chem. Int. Ed. 2010, 49, 125–128. [121] C. Yuan, Q. Xin, H. Liu, L. Wang, M. Jiang, X. Tao, Sci. China-Chem. 2011, 54, 587–595. [122] C.-X. Yuan, X.-T. Tao, Y. Ren, Y. Li, J.-X. Yang, W.-T. Yu, L. Wang, M.-H. Jiang, J. Phys. Chem. C 2007, 111, 12811– 12816.

Eur. J. Org. Chem. 2012, 7015–7041

[123] Q. Xin, X.-T. Tao, F.-Z. Wang, J.-L. Sun, D.-C. Zou, F.-J. Wang, H.-J. Liu, Z. Liu, Y. Ren, M.-H. Jiang, Org. Electron. 2008, 9, 1076–1086. [124] S. Sergeyev, D. Didier, V. Boitsov, A. Teshome, I. Asselberghs, K. Clays, C. M. L. Vande Velde, A. Plaquet, B. Champagne, Chem. Eur. J. 2010, 16, 8181–8190. Received: September 20, 2012 Published Online: November 20, 2012


www.eurjoc.org

7041

Anniversary of the Tröger’s Base Molecule: Synthesis and

Recommend Documents