Steric and Stereochemical Modulation in Pyridyl- and Quinolyl-Containing Ligands

Abstract

Nitrogen-containing pyridine and quinoline are outstanding platforms on which excellent ionophores and sensors for metal ions can be built. Steric and stereochemical effects can be used to modulate the affinity and selectivity of such ligands toward different metal ions on the coordination chemistry front. On the signal transduction front, such effects can also be used to modulate optical responses of these ligands in metal sensing systems. In this review, steric modulation of achiral ligands and stereochemical modulation in chiral ligands, especially ionophores and sensors for zinc, copper, silver, and mercury, are examined using published structural and spectral data. Although it might be more challenging to construct chiral ligands than achiral ones, isotropic and anisotropic absorption signals from a single chiroptical fluorescent sensor provide not only detection but also differentiation of multiple analytes with high selectivity.

1. Introduction

As nitrogen-containing aromatic compounds, pyridine and quinoline can form complexes with many metal ions because the lone pair electrons on the nitrogen are available for coordination since they are not part of the aromatic systems. The aromatic ring of pyridine or quinoline itself is a rigid platform, which can be incorporated into many achiral and chiral binding pockets to build ligands of different affinities to different metal ions. Chiral pyridyl- or quinolyl-containing ligands and their metal complexes are used as catalysts in asymmetric catalysis [1]. A recent review described the design principle for selective metal ion binding and sensing using many achiral pyridyl-containing ligands [2]. There are some other examples of achiral pyridyl-containing ligands used for metal sensing and this article will discuss the structural features of some of them [3,4,5,6]. This article mainly focuses on the stereochemical approach to achieving selective metal binding and sensing using pyridyl-/quinolyl containing ligands, especially chiral ligands. The structure-activity relationship in the modulation of coordination chemistry and/or signal transduction using such ligands/sensors will be discussed.

2. Steric and Stereochemical Modulation of Binding Affinity and Selectivity

According to Comba [7], the design of a selective ligand for a metal ion must involve a high degree of preorganization for the specific metal ion and also a high degree of “disorganization” or mismatch for competing metal ions. The latter is not an easy task and it has not been addressed in detail. Still et al. pointed out that an important principle in the rational design of synthetic host molecules is using substitution and stereochemistry to reduce the populations of conformations unfavorable to binding [8]. By the same token, substitution and stereochemistry manipulation should be able to reduce the population of conformations favorable to binding, thus enabling the manipulation of selectivity.

2.1. Steric Control of Achiral Ligands’ Selectivity to Zn²⁺ and Cu²⁺

Many ionophores are incorporated in sensors for metal ions and other species. Signal transduction of many sensing events depends on the structure and conformation of the analyte-sensor complexes. In the development of Zn²⁺ chelators and sensors, specificity to zinc is highly desirable from many perspectives. Over the years, some sensitive fluorescent sensors for Zn²⁺, which are mostly N-containing ligands, have been developed and their selectivity against some other metal ions has been investigated [6,9,10,11,12,13,14,15,16,17,18,19,20,21]. However, as predicted by the Irving-Williams series [22], Cu²⁺ complexes with nitrogen donor ligands are typically found to be more stable than Zn²⁺ complexes by several orders of magnitude: Complexation with d¹⁰ metal ion Zn²⁺ offers no ligand field stabilization energy (LFSE) [23]. For example, macrocycles are ideal for the selective coordination of alkali or alkali earth metals, which can be discriminated from each other solely on the basis of their ionic radii and charge. However, they may be less useful in distinguishing Cu²⁺ from Zn²⁺ because the radii of these ions are almost identical. This is reflected in the relative binding affinities of macrocyclic ligands toward Zn²⁺ and Cu²⁺, where the latter is favored by 10–15 orders of magnitude [21,24].

One simple Schiff base sensor exhibits Zn²⁺-chelation enhanced fluorescence, which suffers from interference from Cu²⁺. A subtle structural change can turn such a ligand platform from an enol-imine tautomer to a keto-enamine tautomer, which is much more selectively for Zn²⁺ over Cu²⁺ [25].

Rorabacher et al. examined the steric effect on Cu^I/II redox couples [26]. They found that in aqueous solution for 35 different tripodal ligands, many of which are tris(2-pyridylmethyl)amine (TPA, Figure 1) analogs with different substituents, the stability constants of their Cu⁺ complexes are in the relatively narrow range of 10¹²–10¹⁶ although they are hugely different in terms of coordination geometry and donor strength, while those of their Cu²⁺ complexes stretch over 26 orders of magnitude. It was suggested that ligand coordination geometry mainly impacts the complexation of Cu²⁺, while imposing little effect on Cu⁺.

Because of its 3d¹⁰4s⁰ configuration, Zn²⁺, like Cu⁺ [22], is not strongly influenced by constraints in its coordination configuration. Therefore, it is necessary to make a ligand which has a high degree of predisorganization for Cu²⁺ so that it can show better Zn²⁺/Cu²⁺ selectivity. As a d⁹ metal, the bonding in Cu²⁺ complexes is partially covalent. At the same time, Cu²⁺ prefers 4-coordinate square planar and 5-coordinate square pyramidal geometries over tetrahedral and trigonal bipyramidal geometries according to Crystal Field Theory [27]. C₃ or pseudo-C₃ symmetrical N-containing TPA derivatives bind Cu²⁺ much better than Zn²⁺ as predicted by the Irving-Williams series [22]. At the same time, they are naturally trigonal bipyramidal geometry providers and all of their Zn²⁺ complexes are of trigonal bipyramidal configuration to the author’s awareness. To accommodate Cu²⁺, this trigonal bipyramidal configuration is often distorted to resemble a square (bi)pyramidal. This preference is clearly demonstrated in parabenzobis-TPA (PBTPA, Figure 1) [28], one of the tripodal ligands based on TPA. X-ray structures (Figure 2) shows that it forms a trigonal bipyramidal complex with zinc (with N_Py-Zn-N_Py bond angles of 117.3°, 114.9°, and 118.5°) and a square bipyramidal complex with copper(II) (with N_Py-Cu-N_Py bond angles of 85.2°, 95.6°, and 164.4°). Some conformationally mobile chiral TPA analogs also form square pyramidal complexes with Cu²⁺ [29]. To accommodate other metals such as Fe³⁺ [30] and V⁴⁺ [31] complexes, such distortion is also needed.

Karlin et al. found that tris(2-quinolylmethyl)amine (TQA) (Figure 3) and its analogs form C_σ symmetrical square pyramidal complexes with Cu²⁺ (Figure 3) [32]. Other researchers found that steric overcrowding decreases the formation constants between TPA/TQA type of ligands and metal ions. The extent of such a decrease is inversely proportional to the ion sizes. Significant decrease is seen with small ions such as Zn²⁺, while smaller decrease is seen with larger ions such as Pb²⁺ because the Zn-N bond length is distorted in the Zn²⁺-complex with overcrowded ligands [33].

One of the most widely used fluorescent zinc sensors is TSQ [34,35]. Its analogs Zinquin and 2-Me-TSQ (Figure 4) show better Zn²⁺/Cu²⁺ selectivity than most other fluorescent probes, although they still bind Cu²⁺ stronger than Zn²⁺ [36,37]. The overall binding constant of the Cu²⁺ complex (logβ₂ = 18.3 ± 0.05) of Zinquin is only slightly greater than that of its Zn²⁺ counterpart (logβ₂ = 17.54) [12]. A complex between Zn²⁺ and 2-Me-TSQ can be formed with a 1:2 stoichiometry (Figure 4). The methyl groups at the 2-position of each of the two quinolones would clash with each other in a square planar or an octahedral complex, although the bite angle of 83° accommodates these two geometries. As a result, a distorted tetrahedral geometry is formed to relieve such steric hindrance, as shown in its X-ray crystal structure (not shown in Figure 4) [37]. However, it may be energetically less favorable for Cu²⁺, which might account for the lower than expected affinity for Cu²⁺. Therefore, engineering a ligand scaffold that can exert significant steric restrictions upon metal coordination geometry appears to be a promising approach for the design of Zn(II)-selective fluorescence probes.

2.2. Stereochemical Control of Chiral Ligands’ Zn²⁺/Cu²⁺ Selectivity

Although it sounds outlandish to modulate the behavior of achiral metal ions through chiral organic ligands, it is not without precedence. Mother Nature utilizes chiral substances in non-asymmetric processes. Lasalocid is a natural ionophore for Na⁺ with several asymmetric carbons in its skeleton. The metal-binding functions of its stereoisomers depend on their stereochemistry, as do those of some tetrahydropyranoid podand ionophores [8,38,39]. Similar results have been found in another polyether ionophore and metal-binding antibiotic, monensin, its stereoisomers, and analogs [40]. Ca²⁺ is selectively bound by an isomer of the hydroxylated-bistetrahydrofuran skeleton of a potent antitumor agent, annonaceous acetogenins, against other alkali or alkali-earth metal ions [41].

Zinc finger peptides bind zinc exceptionally well, with dissociation constants as low as 5.7 pM [42], because they possess peptidyl domains 25–30 residues in length that form pre-organized metal binding pockets highly selective for divalent zinc. Imperiali et al. explored the “hybrid approach” by constructing and attaching peptidyl domains seven residues in length to the 8-HQ fluorophore, which is structurally similar to TSQ. The peptidyl domain was carefully designed to try to preserve the architecture of the zinc finger motif using minimized size and complexity. One of the reported ligands is shown in Figure 5 [43]. The cysteine residue is key to enhancing the zinc selectivity of the ligand.

The (R,R)- or (S,S)-isomer of a bis-chiral crown ether has been reported to bind selectively sodium against potassium due to the optimized size of the binding pocket, while its optically inactive (R,S)-isomer does not show such selectivity [44].

Balaz et al. used single-labeled pyridylporphyrin–DNA conjugates to sensitively and selectively detect Hg²⁺ in water, although pyridylporphyrin rather than the nucleobase was found to play a crucial role in Hg²⁺ binding and sensing [45].

In a study by Castagnetto et al. recognition of Zn²⁺ by MeBQPA (Figure 6), a chiral derivative of TPA/TQA, benefited from both fluorescence enhancement as well as chiroptical signal increase [46]. However, Cu²⁺ was a significant competitor for Zn²⁺ in that system, as it is for all ligands based on the TPA scaffold [47]. In solution, Canary et al. found that the chiral TQA derivative MeTQA (Figure 6) forms Cu²⁺ complexes with geometries between square pyramidal and trigonal bipyramidal, in some cases even predominantly square pyramidal, as revealed by UV [48].

To improve Zn²⁺/Cu²⁺ selectivity, further preorganization to impose a trigonal bipyramidal coordination geometry is needed. Rigidification is a common approach to preorganization, and in principle it should also work in predisorganization. The orientation and interaction between substitution groups can also exert influences on selectivity as shown in Zinquin complexes. Therefore, one can envision that it is possible to construct a ligand whose structure is chirally synchronized and mechanically rigidified so that its trigonal pyramidal configuration cannot be bent to a planar geometry. Such manipulation would depress its Cu²⁺ affinity, while exerting much smaller compromise, if any, on its Zn²⁺ affinity. In Rorabacher’s words [26], one can construct tripodal ligands which are incapable of adapting to a planar geometry but could readily accommodate tetrahedral or distorted tetrahedral geometries. In this way, the higher affinity of TPA/TQA derivatives to Cu²⁺ over Zn²⁺ might be reversed. Toward this end, a stereochemical control approach was developed to engineer improved Zn²⁺/Cu²⁺ selectivity through controlling ligand stereochemistry: a ring was incorporated into the TPA ligand by connecting two of the arms to give compounds 5 and 6 (Figure 7), in which the piper dine ring reduces conformational mobility by rigidifying the compounds and two chiral centers are introduced to further control the stereochemistry [49].

For the cis-piperidine derivative 5, Cu²⁺ and Zn²⁺ complexation gave logβ = 14.8 and 10.1, respectively, and for trans-ligand 6 the numbers were found to be 12.0 and 11.2, respectively [49,50]. The parent compound TPA shows logβ = 16.15 for its Cu²⁺ complex compared to 11.00 for its Zn²⁺ complex. Thus, the ratio of the association constants for the binding of Cu²⁺ over Zn²⁺ for TPA, 5, and 6 is 1.4 × 10⁵, 5 × 10⁴, and 6, respectively. As a pair of diastereomers, piperidine compounds 5 and 6 are expected to exhibit somewhat different affinities toward metal ions. However, the difference here is so large mainly because the trans-piperidine ligand 6 enforces a C₃ coordination environment through the identical stereochemistry at the two chiral centers and the rigid piperidine ring, which makes it less favorable to Cu²⁺ binding. This qualitatively explains its dramatically improved Zn²⁺/Cu²⁺ selectivity.

Semi-empirical calculations and X-ray structures show greater similarity of the [Cu(TPA)Cl]⁺ Cu-N bond lengths in the complex with 5 than in 6 [49]. Thus, trans-ligand 6 appears to distort the coordination sphere of the Cu²⁺ ion, resulting in a less stable complex, while ligand 5 is preferred significantly for Cu²⁺. This agrees with the observation that the binding of Cu²⁺ is quite dependent on ligand stereochemistry while that of Zn²⁺ is not.

The trans-ligand 6 was tagged with naphthalene fluorophores to prepare ligands 7 [49] and 8 [51,52] (Figure 7). The fluorescence of the naphthalene moieties is diminished by photo-induced electron transfer (PET) in the absence of metal ion, but increases nearly 20-fold upon binding Zn²⁺ for compound 7. The sensitivity of compound 8 for Zn²⁺ was found to be nanomolar in 4-(2-hydroxyethyl)-1-pieperazineethanesulfonic acid(HEPES) buffer with 1% methanol at physiological pH. The improved selectivity that had been found for the trans chiral piperidine scaffold was also preserved with compounds 7 and 8. Thus, stereochemical engineering of ligands by constructing an unfriendly environment for Cu²⁺ to depress their Cu²⁺ affinity and enhance their Zn²⁺/Cu²⁺ selectivity has proven to be feasible.

2.3. Steric and Stereochemical Control of Ligands’ Selectivity to Ag⁺

Some pyridyl-containing macrocycles (Figure 8) are used as ionophores for Ag⁺. Their structures and conformations affect their affinities to ions. Compounds 9, 10, and 13 are somewhat planar. Since there is intramolecular N_Py-HN-amide hydrogen bonding in these free ligands, each of their cavities has to undergo significant conformational changes during complexation, resulting in poor affinity to Ag⁺. However, since the steric requirement of methyl/benzyl substituents on amide N in 11 and 12 takes themselves out of the cavities and positions the amide C=O toward their respective cavities, they may undergo fewer conformational changes during complexation, thus showing higher affinities to Ag⁺ over Pb²⁺, Tl⁺, alkali, and alkaline earth cations. The increased spacer length in 14 removes such steric arrangement, resulting in poor selectivity [53].

Proper handling of ligand stereochemistry can lead to improved Ag⁺ affinity in some podand ligands bearing pyridine moieties and two chiral arms (Figure 9, top) [54,55]. In principle, as a pair of diastereomers, (S,S)- and (R,S)-15, respectively, can exhibit different binding abilities toward cations. The (S,S) ligands can extract Ag⁺ more selectively and more slowly in the presence of Pb²⁺, Cu²⁺, Ni²⁺, Co²⁺, and Zn²⁺ than their corresponding meso ligands. Computer modeling and energy calculations showed that in the optimized structures of their Ag⁺ complexes (Figure 9, bottom), (S,S)-15 has a symmetrical arrangement of three pyridine rings for the binding of the Ag⁺ ion, which is equally coordinated by two terminal pyridine rings, while the two terminal pyridine nitrogen atoms coordinate with the Ag⁺ ion in an asymmetrical fashion in the Ag⁺ complex with (S,R)-15. The energy calculations indicate that the Ag⁺-(S,S)-15 complex is more stable by 3.46 kcal/mol than the Ag⁺-(S,R)-15 complex. Such differences are also observed in the derivatives of these two diastereomers. It was concluded that a combination of ligand geometry and stereo-controlled substitution can improve Ag-specificity in this class of ligands [54,55].

3. Steric and Stereochemical Modulation of Signal Transduction

Many ionophores are incorporated in sensors for metal ions and other species. Signal transduction of many sensing events depends on the structure and conformation of the analyte-sensor complexes.

3.1. Modulating Fluorescence or Luminescence

Many metal ions sensors employ fluorescence as their signal output. Generally “switch-on” fluorescent sensors are preferred to “switch-off” ones. There have been significant endeavors to make fluorescent sensors for Hg²⁺. Although a number of reversible “switch-on” fluorescent sensors for Hg²⁺ have been reported [56,57,58,59], many other sensors exhibit fluorescence “switch-off” upon binding Hg²⁺, which as a heavy metal turns to quench its sensors’ fluorescence through spin-orbit coupling. Hancock et al recently shed some light on the structural requirements for Hg²⁺ sensors that exhibit chelation enhanced fluorescence (CHEF) when their photo-induced electron transfer (PET) processes are handicapped by Hg²⁺ [60]. The secret is that the formation of a π-complex between the heavy metal and the fluorophore needs to be disrupted or eliminated. By examining some pyridyl-containing sensor, including N-(9-anthracenylmethyl)-N-(2-pyridylmethyl)-2-pyridinemethanamine (ADPA) (Figure 10), and other nitrogen-containing sensors from a few research groups, it is generalized that the Hg²⁺ ion should be held far enough away from the fluorophore, or covalently binding donor atoms, such as S and Br, need to be employed to limit the strength of interaction between the Hg²⁺ ion and the fluorophore. This might be able to account for the “switch-on” fluorescence of chiroptical mercury sensors that contain quinolyl and methionine/S-methylcysteine moieties made by the author of this article [61].

Steric crowding in ligands can compromise the (CHEF) effect by small metal ions such as Zn²⁺ ion as compared to larger ions such as Cd²⁺ ion [33]. Steric crowding distorts the Zn-N bond length, which allows some quenching of fluorescence by the PET mechanism. The steric crowding increases in the following sequence for tripodal pyridyl-/quinolyl-containing ligands: TPA < TQA < tris(6-methyl-2-pyridyl)amine (TMPA). In a complex formed between Zn²⁺ and TQA, X-ray crystallography shows that the Zn-N bonds are all of normal lengths (Figure 11), which means the level of steric crowding in TQA is not severe enough to cause significant Zn-N bond length distortion. As a result, there is larger enhancement of TQA fluorescence by Zn²⁺ than by Cd²⁺, in contrast to similar but more sterically crowded TMPA where Cd²⁺ induced CHEF effect is stronger. The CHEF effect for TQA increases with the decrease in metal ions sizes.

Open chain TPA analogs with two chiral arms (Figure 12) have been used to tune lanthanide luminescence [62]. Lanthanide complexes formed with (R,S)-16 ligand give stronger fluorescence than their corresponding cousins formed with (R,R)-16 ligand. Lanthanide complexes of another pyridyl-containing ligand with two chiral centers also exhibited interesting properties in luminescence [63].

More recently, a chiral pyridyl/quinolyl-containing tripodal ligand was demonstrated to form a series of lanthanide complexes exhibiting multiple anion-sensing profiles, which can be explained by the presence of a fluorescent quinoline and a stereocontrolled methyl group resulting in differences in fluorescence, CD, and Ln(III)-luminescence signals of the anion-bound complex, which are controlled by the nature of the targeted anions [64]. It is more specific than regular fluorescence sensing.

3.2. Modulating Chiroptical Signals

Compared with achiral ligands, chiral ones can yield additional spectroscopic information such as chiroptical signals [65]. Zinc greatly enhances the fluorescence of MeBQPA [46], while other metal ions induce fluorescence responses. More interestingly, the ligand generates strong signals in exciton-coupled circular dichroism (ECCD) upon formation of complexes with some metal ions (Zn²⁺, Cu²⁺), while complexes with octahedral metal ions (Cd²⁺, Fe²⁺) do not give strong CD signals (Figure 13). Both isotropic (fluorescence) and anisotropic absorption (CD) signals from the optical response of this single chiral ligand are employed to provide not only detection but also differentiation of multiple analytes: Zn²⁺ (strong fluorescence and ECCD response), Cu²⁺ (strong ECCD but no fluorescence), Cd²⁺ (strong fluorescence but no ECCD) and Fe²⁺ (neither fluorescence nor ECCD).

Metal ion detection by a multimode switchable chiroptical fluorescent sensor containing both S-methylcysteine and quinoline moieties through both fluorescence enhancement and anisotropic absorption distinguish even more metal ions [61]. We will discuss these sensors in more detail in the “Modulating Switchable Binding Pockets in One Ligand” section.

A pyridyl-containing bidentate Hg²⁺ sensor with a chiral-center near its naphthalene chromophore has been reported more recently (Figure 14). The sensor exhibits significant CD changes when titrated with Hg²⁺, while many other metal ions do not induce such changes. Cold ESI-MS and ¹HNMR data suggest that Hg²⁺ and the ligand form a 1:1 coordination polymer, which gives a negative exciton coupled CD (ECCD) signal, suggesting that the naphthalene units in the coordination polymer are arranged in a counterclockwise screw sense in solution. The complex’s X-ray crystallography indeed shows a polymer-like structure, although the naphthalene units in the solid state are arranged in an ECCD-inactive eclipsed form, likely caused by a packing effect [66].

There are other chiroptical sensors for metal ions that show changes in CD. However, they do not contain pyridyl or quinolyl groups [67,68,69] and will not be discussed in further detail.

A new approach, differential circularly polarized fluorescence excitation (CPE), to metal ion sensing using fluorescence-detected circular dichroic detection was developed, which integrates fluorescence and exciton coupled circular dichroism methods to give better contrast than can be achieved in either of the two parent methods. This approach uses the ΔF (ΔF = F_L − F_R, F_L, F_R = fluorescence with left and right circularly polarized excitation, respectively) [70] component of fluorescence-detected circular dichroism (FDCD) for metal sensing. The contrast in ΔF signals between a sample with both a large quantum yield and a large CD and a sample with both a small quantum yield and a small CD will be much larger than the contrast in either fluorescence or CD signals. The corresponding spectra of one of the sensors employed, compound 9 ((S,S) form, structure shown in Figure 7), titrated with Zn(II), are shown in Figure 15 [52]. Apparently, measurements in ΔF gave greatly enhanced contrast over other spectroscopic methods. On the coordination chemistry front, stereochemical control and rigidification employed in these sensors ensures improved Zn²⁺/Cu²⁺ selectivity. On the photophysical signal transduction front, the CPE approach has the potential to improve contrast and diminish interference from background fluorescence, such as that from the protein lyzozyme which contains tryptophan (a common source of background fluorescence in cells) (Figure 16).

4. Modulating Two Binding Pockets in One Ligand

4.1. Modulating Ditoptic Ligands for Zn²⁺

There can be two or more binding pockets in one ionophore. The Zhu group designed a series of fluorescent sensors based on a pyridyl-containing platform bearing two binding pockets with different affinities to Zn²⁺ (Figure 17) [71,72,73,74,75,76,77,78]. Through the modulation of PET, internal charge transfer (ICT), conformation rigidification, and substitution, such ditopic ligands can bind low concentration Zn²⁺ through the high-affinity pocket (bis- or tris-(2-methylpyridyl)amine) to give fluorescence enhancement in one wavelength channel and bind high concentration Zn²⁺ through both the high-affinity and low-affinity(2,2′-bipyridyl) pockets to result in fluorescence enhancement at another wavelength channel. Such sensors have been proven to be useful in live-cell imaging of free Zn²⁺ over a concentration range of six orders of magnitude [79].

4.2. Modulating Switchable Binding Pockets in One Ligand

Sauvage et al. made a molecular muscle system [80,81] employ a rotaxane dimer 17: its Cu⁺ complex 18²⁺ is the extension state in which the bidentate phenanthrolinyl of a macrocyle are pulled near to the bidentate phenanthrolinyl in the middle of the molecule by Cu⁺, and its Zn complex 19⁴⁺ is the contraction state in which the bidentate phenanthrolinyl of the macrocycle is pulled near the tridentate terpyridyl at the two ends of the molecule by Zn²⁺ (Figure 18).

4.3. Modulating Switchable Chiroptical Sensors for Metal Oxidation States

Pyridyl and phenanthrolinyl groups were incorporated in catenane [82] and rotaxane [83,84] systems with two binding pockets that can switchably bind copper ions at different oxidation states by turning a bischelating macrocycle containing both bidentate (phenanthroline) and tridentate (terpyridine) moieties around a bidentate (phenanthroline) axel.

Pyridyl groups have also been used in the design of chiral sensors whose coordination chemistry and signal transduction are sensitive to the metal ions’ oxidation states.

Shanzer et al. reported such a sensor which employs a triple-stranded system that accommodated a single metal ion in one of two sites, either a “hard” binding cavity bearing three hydroxamate moieties preferable to Fe³⁺ or a “soft” cavity with three bipyridyl moieties preferable to Fe²⁺ (Figure 19) [85]. A split CD spectrum in the UV region was three times more intense for the Fe²⁺ than for Fe³⁺, suggesting exciton interactions involving the bipyridyl groups, which originate from the helical arrangement of each of the three strands. Since switching metal oxidation states can be achieved through redox processes, such processes can be monitored by switchable chiroptical sensors. Chiral tripodal ligands bearing a chiral arm and two achiral 2-methylquinolyl arms can form propeller-like metal complexes whose configuration is dictated by the chiral centers in such ligands as proven in many crystallographic structures in the solid state and ECCD in solution. Such a ligand, N,N-Bis(2-quinolylmethyl)-l-methionine (l-MethBQA) (Figure 20) [86], derived from the amino acid methionine forms a tetradentate complex with Cu²⁺ involving three nitrogen atoms and a carboxylate. Upon binding Cu⁺, the ligand reorganizes and the sulfide moiety replaces the carboxylate from coordination. Binding Cu²⁺ and Cu⁺ produce opposite helical orientation of the two quinolyl moieties, resulting in mirror images in the ECCD spectra for the Cu⁺ vs. Cu²⁺ complexes. Other such tripodal derivatives of methioninol and S-methylcysteine can also sense the oxidation states of copper ions following the same mechanism (Figure 21) [87,88,89,90].

4.4. Modulating Switchable Chiroptical Sensors for Multiple Metal Ions

l-MehtBQA and a similar compound N,N-Bis(2-quinolylmethyl)-l-S-methylcysteine (l-CysBQA) (Figure 20), were found to be multimode switchable chiroptical fluorescent sensors for multiple ions including but not limited to Hg²⁺, Cu^2+, and Zn²⁺ [61]. Quinolyl groups serve as the fluorophore and possess nitrogen lone pairs capable of chelating metal ions. Upon exposure to Hg²⁺ or Zn²⁺ these sensors show signal enhancement in fluorescence in 30:70 acetonitrile/water. It is likely that the interaction between the Hg²⁺ ion and the fluorophore is limited by the covalently binding S atom, disrupting the heavy atom effect and resulting in fluorescence enhancement [60]. However, Cu²⁺ quenches their fluorescence. l-CysBQA complexes with Hg²⁺, giving rise to an exciton-coupled circular dichroism spectrum with a positive couplet (a positive Cotton effect at a longer wavelength followed by a negative Cotton effect at a shorter wavelength). However, Cu²⁺ or Zn²⁺ complexation produces a negative ECCD couplet (Figure 22). This remarkable differentiation of Hg²⁺ from Cu²⁺ and Zn²⁺ stems from the different structures of the CD active products. The Cu²⁺ ion binds with the tertiary amine, the two quinolones, and the carboxylate moieties of the ligand. The two quinoline groups form a propeller whose orientation is dictated by the stereocenter of the S-methyl cysteine arm. However, Hg²⁺ prefers coordination by the sulfur atom. As shown in Figure 22, for the sulfide to bind to the metal center, the amino acid arm must pivot about the C-N bond, which inverts the orientation of the quinoline moieties, leading to an exciton coupled CD with the opposite sign. Although crystals of l-CysBQA complexes with these metal ions are not available, Zn²⁺ and Cu²⁺ coordinate with the carboxylate instead of the sulfide in the crystals of their complexes with l-MethBQA and other similar compounds [88,89,90]. In solution, the soft Cu⁺ ion coordinates with the sulfur atoms of such ligands in solution as demonstrated by NMR data and other evidence [88,91], which indicates that Hg²⁺ should coordinate the sulfur atom since it is also soft.

The design of such sensitive and selective chiroptical fluorescent sensors for metal ions includes innovations on both coordination chemistry and signal transduction. Take the above mentioned mercury sensors for example. On the coordination chemistry front, ionophores are equipped with two sets of coordination “teeth”, switchable by exposure to different metal ions. One set of “teeth” offers high affinity for Hg²⁺, determining the sensitivity of the probe; the other set preferably binds other metal ions. On the photophysical signal transduction front, the binding between Hg²⁺ and its preferred set of “teeth” leads to fluorescence enhancement and a positive exciton coupled CD by design; and the coordination of the other set of “teeth” with other metal ions, such as Zn²⁺ and Cu²⁺, triggers a fluorescence change (enhanced or quenched) and a negative exciton coupled CD. Other metal ions produce other combinations of fluorescence and exciton coupled CD. In this way, metal ion sensing by such chiroptical fluorescent sensors through both fluorescence and anisotropic absorption distinguishes (Figure 23), for example, Hg²⁺ (enhanced fluorescence with strong positive exciton coupled CD (ECCD)), Zn²⁺ (enhanced fluorescence and strong negative ECCD), Cu²⁺ (strong negative ECCD but quenched fluorescence), Ni²⁺ (strong positive exciton coupled CD but quenched fluorescence), Pb²⁺ (quenched fluorescence but no ECCD), Cd²⁺ (enhanced fluorescence but no ECCD). More ions such as Cd²⁺(enhanced fluorescence and no exciton coupled CD), Ag⁺ (no fluorescence change and strong positive exciton coupled CD) and alkali metal ions (no change in fluorescence or CD) can be added to the list [92]. l-MethBQA offers similar advantages. These results further illustrate that recognition involving both isotropic and anisotropic detection tools may be utilized to maximize the information transmitted by a single sensor molecule [46].

Interestingly, a pyridyl analog l-MethBPNaph (Figure 20) shows fluorescence enhancement as well as significant spectral red-shift in emission upon exposure to HgCl₂ [93]. However, its CD spectrum does not changed upon addition of HgCl₂, ZnCl₂, CuCl₂, Zn(ClO₄)₂, Cu(ClO₄)₂, Ni(ClO₄)₂ or Pb(ClO₄)₂.

Aside from metal ions, pyridyl containing systems have been used as chiroptical sensors for anions [64,94] and other species, some of which have been recently reviewed [93,95,96]. Recently, a pyridyl-containing homochiral, square-shaped, D₂ symmetrical metal-linked macrocycle has been shown to be a selective chiroptical and electrochemical sensor for ferrocene in the presence of other species [97].

5. Conclusions

Pyridyl- or quinolyl-containing compound are excellent platforms to build selective ionophores and sensors for metal ions. Steric and stereochemical effects can be used to modulate such ionophores and sensors in terms of coordination chemistry and/or signal transduction. On the coordination chemistry front, a ligand’s affinity and selectivity toward metal ions can be systemically fine-tuned or switched through modification of ligand structures by introducing/removing steric crowding or adding chiral handles, which creates a high degree of preorganization for the specific metal ion and/or a high degree of ‘disorganization’ or mismatch for competing metal ions. On the photophysical signal transduction front, steric effect can be used to engineer “turn-on” fluorescent sensors for heavy metal ions and control the extent of chelation induced enhancement of fluorescence. Stereochemical modification can bring additional spectroscopic information such as chiroptical signals into the signal transduction part of sensing, which is capable of making sensing events more specific. Sensing strategies employing both isotropic and anisotropic absorption signals from a single chiral sensory molecule provide not only detection but also differentiation of multiple analytes.