Supramolecular Chemistry - 2009 - Steed - The Supramolecular Chemistry of Life.pdf

The Supramolecular Chemistry 2 of Life ‘Nature that fram’d us of four elements, Warring within our breasts for regiment, Doth teach us all to have aspiring minds: Our souls, who...

The Supramolecular Chemistry 2 of Life ‘Nature that fram’d us of four elements, Warring within our breasts for regiment, Doth teach us all to have aspiring minds: Our souls, whose faculties can comprehend The wondrous Architecture of the world……’ Christopher Marlowe (1564–1593), Conquests of Tamburlaine  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 50 The Supramolecular Chemistry of Life 2.1 Biological Inspiration for Supramolecular Chemistry Nelson, D. L. and Cox, M. M., Lehninger Principles of Biochemistry. 4th ed.; W. H. Freeman: New York, 2004. Much of the inspiration and origins of supramolecular chemistry comes from the chemistry found in living biological systems. Sometimes incredibly complex, sometimes elegantly simple, Nature has evolved an enormous amount of highly specific, hierarchical, selective and cooperative chemistry that enables living systems to maintain themselves in a dynamic equilibrium with their environment and to feed, respire, reproduce and respond to external stimuli. In biological chemistry, the supramolecu- lar hosts are the receptor sites of enzymes, genes, antibodies of the immune system, and ionophores. The guests are substrates, inhibitors, co-factors, drugs or antigens. These components variously exhibit supramolecular properties such as molecular recognition, self-assembly, self-organisation, self- replication and kinetic and thermodynamic complementarity. The vast majority of these properties rely upon supramolecular interactions such as coordination (ion–dipole) bonds, hydrogen bonds and π–π stacking discussed in Section 1.8. Biological systems are, therefore, supramolecular systems par excellence. A great deal of effort in supramolecular chemistry has been expended in attempts to model or mimic biological processes such as the catalysis of organic chemical reactions by enzymes, or the selective transport of metal cations or molecular substrates such as O2. As part of this process, our un- derstanding of biological systems has grown enormously, but it is also fair to say that the molecular and supramolecular chemistry of human endeavour, as they stand now, are still a long way away in scale, scope and functionality from their biochemical analogues. The fact that Nature exhibits such a rich and efficient natural supramolecular chemistry is, however, an enormous encouragement and motivation to continue to seek ever more sophisticated abiotic (nonbiological) analogues and indeed to attempt to develop synthetic systems capable of carrying out transformations or possessing properties not found naturally. In this chapter, we give a brief and highly selective overview of some of the more important biological chemistry of relevance to supramolecular chemists by way of a very brief introduction to the extensive synthetic systems that we will be looking at in the rest of this book. Subsequent chapters will deal with the ways in which synthetic and model systems mimic these biological processes and how insight has been gained into biochemistry by the study of supramolecular compounds, as well as the enormous diversity of entirely non-biological supramolecular chemistry. 2.2 Alkali Metal Cations in Biochemistry 2.2.1 Membrane Potentials Energy is vital to life. Plants get energy from sunlight (photosynthesis), humans get energy from food, which we oxidise to CO2 and water. Energy is used in respiration, a process by which energy from food is transformed and stored as the chemical bond energy of ATP (adenosine triphosphate). Strictly, ATP has a 4– ionic charge, balanced by alkaline and alkaline earth metal cations. In biological nota- tion this is often omitted. ATP is capable of long-term energy storage and is transported to any areas where energy is needed to drive endergonic (energy-consuming) reactions such as muscle contraction. The energy is released from ATP by a class of enzymes called ATPases, of which Na⫹/K⫹-ATPase is perhaps the most important example. One mole of ATP releases 35 kJ of energy, according to the Supramolecular Chemistry, 2nd edition J. W. Steed and J. L. Atwood © 2009 John Wiley & Sons, Ltd ISBN: 978-0-470-51233-3  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Alkali Metal Cations in Biochemistry 51 NH2 NH2 N N N + H2O N + H2PO4- N N Mg2+ N N O O O ∆Go = -35 kJ mol-1 O O O O O O O P O P O P O P O P OH OH O O O OH O O OH ATP ADP Scheme 2.1 Energy-releasing dephosphorylation of ATP to ADP and dihydrogen phosphate. Mg2⫹ acts as a catalyst for the reaction. reaction shown in Scheme 2.1. Note that while the ATP molecule is relatively complex, it is only the triphosphate tail that is changed during the course of the reaction. Breaking of the terminal phos- phate ester P—O bond gives the dihydrogen phosphate anion (H2PO4⫺, often referred to as ‘inorganic phosphate’, Pi) and adenosine diphosphate, ADP. The Na⫹/K⫹-ATPase enzyme is an example of a transmembrane enzyme, i.e. an enzyme that exists in the phospholipid membrane (wall) of a biological cell. As part of the process of consuming ATP, the Na⫹/K⫹-ATPase enzyme transports the alkali metal cations Na⫹ and K⫹ from one side of the cell membrane to the other. Effectively, the enzyme scavenges Na⫹ from the inside of the cell and trans- ports it to the outside, against the prevailing concentration gradient. Simultaneously, K⫹ is transported into the cell. Thus, in the intracellular fluid there is a high concentration of K⫹; outside there is a high concentration of Na⫹ (Table 2.1). This uneven distribution of alkali metal cations across the cell mem- brane is a highly important and necessary feature and results in a transmembrane electrical potential, rather like a battery. This potential difference across the cell is used, amongst other things, in informa- tion transfer in nerve cells (Figure 2.1). The actual amount of charge separation across a cell membrane is very small (the number of M⫹ ions is equal on either side of the membrane). Such a potential difference could, in principle, have been set up by separating Na⫹ and Cl⫺ across a membrane. However, such an actual separation of oppositely charged ions would require much more energy because of the large electrostatic forces between such ions. In fact, the resulting chemical potential arising from the different identities of the alkali metal ions (Na⫹ and K⫹) is sufficient to generate the required signal. The most important requirement for utilisation of this kind of ionic diffusion as a means to in- formation transfer is the maintenance of the non-equilibrium ionic concentration gradient. This is a relatively unstable state – it requires energy to counteract the natural entropy-increasing flow back to equilibrium. This is best illustrated by the pump storage model. Ions are actively ‘pumped’ through Table 2.1 Some examples of biochemical Na⫹ and K⫹ distributions. Concentration/mmol kg–1 Location K⫹ Na⫹ Human intracellular fluid (e.g. erythrocytes) 92 11 Human extracellular fluid (e.g. blood plasma) 5 152 Squid nerve (inside) 300 10 Squid nerve (outside) 22 440  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 52 The Supramolecular Chemistry of Life Figure 2.1 Mode of signal transduction of nerve system. Under concentration gradients, generated by Na⫹/K⫹-ATPase, opening of an ion channel causes the passive efflux of K⫹ and the influx of Na⫹ resulting in a small burst of electrical current (nerve impulse) and a change in the membrane potential. At the end of the nerve cell (axon) the electrical signal is transformed into a chemical signal by trig- gering the ejection of a hormone such as acetylcholine (Section 2.7). The hormone, in turn, triggers the opening of a ligand-gated ion channel in the next nerve axon and restarts the nerve impulse as an electrical current by allowing passive flow of K⫹ and Na⫹ across the next membrane. the biological membrane against the concentration gradient until a stationary, non-equilibrium state is reached. Stimulation results in the rapid, passive flow of ions back to equilibrium via the operation of gate functions (Figure 2.2). The importance of maintaining precise concentration gradients is highlighted by the severe effects of metabolic disorders involving alkali metal cations. For example, high sodium intake is linked intimately with the development of high blood pressure; on the other hand, aged Figure 2.2 The pump storage model. Metal ions are actively pumped by Na⫹/K⫹-ATPase (an energy-consuming process) from regions of low concentration to regions of high concentration against the concentration gradient until a dynamic non-equilibrium state is reached in which active pumping is balanced by accidental diffusion. Upon activation by the appropriate hormone, selectively gated ion channels open allowing the passive (and therefore rapid) flow of ions back towards equilibrium, resulting in current flow.  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Alkali Metal Cations in Biochemistry 53 organisms have some difficulty in preventing the excretion of the very labile K⫹ because of disturbed membrane permeabilities. 2.2.2 Membrane Transport So how does an alkali metal get from the inside of the cell to the outside? A cell membrane consists of hydrophilic (water soluble) phosphate head groups attached to a long lipid (fatty) tail and is thus an example of an amphiphile (Section 13.2.1). In the body’s aqueous environment, the hydrophilic head groups are attracted to the surrounding medium (e.g. via hydrogen bonding and dipolar interactions) while the organic tail is repelled. This results in a bilayer arrangement in which the organic components are all hidden away from the solvent while the hydrophilic portions face out. Anything that is to pass through the cell wall must therefore be able to pass this lipophilic (fat-soluble) region (Figure 2.3). Sodium and potassium cations are not at all lipophilic. They cannot effectively diffuse through the cell wall unless something makes them lipophilic or a nonlipophilic pathway is created for them. There are two main possible methods of such passive cation transport along a concentration gradient: transport by some kind of lipophilic carrier, or controlled passage through a hydrophilic channel in the membrane (Figure 2.4). Transport of metal ions via the carrier mechanism involves a carrier ligand that is able both to bind selectively to the metal cation and to shield it from the lipophilic region of the membrane. Such ion carriers are termed ionophores. The natural products valinomycin (2.1) and nonactin (2.2) are among the best known. Valinomycin was first isolated from the bacterium Steptomyces fulvissimus in 1955, and was established in 1967 to catalyse the exchange of K⫹ and H⫹ across the membrane of mitochon- dria within cells via a carrier mechanism without affecting Na⫹ concentration. Chemically, valino- mycin is a cyclic depsipeptide made up of a threefold repetition of four amino acid residues: L-valine (Val), D-hydroxyisovaleric acid (Hyi), D-valine and L-lactic acid (Lac) (2.3). Hydrogen bonding of type N–H…O⫽C to both ester and amide carbonyl groups plays an important role in the conformation of valinomycin, where it helps the peptide chain wrap around the metal cation, contributing to its degree of preorganisation and stabilising the bound conformation. Valinomycin and nonactin are both selective for K⫹ because they are able to fold in on themselves in order to produce an approximately octahedral array of hard (i.e. non-polarisable, according to the hard and soft acids and bases (HSAB) theory; see Section 3.1.2) carbonyl oxygen atom donors exactly suited to fit an ion of the size of K⫹. Rubidium and caesium are too large, whereas the ionophore cannot contract tightly enough to bind to Na⫹. The X-ray crystal structure of the K⫹ complex of valinomycin is shown in Figure 2.5. It can Figure 2.3 Schematic diagram of a phospholipid biological membrane (5–6 nm in width).  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 54 The Supramolecular Chemistry of Life Figure 2.4 (a) Carrier, (b) channel and (c) gated channel mechanisms of ion transport across a bio- logical membrane. The carrier encapsulates the alkali metal cation, stripping away most or all of the water molecules bound to it. The carrier then presents a lipophilic surface, moving across the mem- brane and releasing the ion back into aqueous solution at the other end. The channel is, primarily, an aqueous hole running through the membrane. Ions can traverse swiftly through the channel without losing their solvent sphere throughout most of their journey, as long as the gate (activated by potential changes or hormones) is open and the ion can pass through the selectivity filter (which discriminates between Na⫹ and K⫹). (Reproduced by permission of John Wiley & Sons, Ltd). Figure 2.5 X-ray crystal structure of the K⫹ complex of valinomycin. For a summary of the technique of X-ray crystallography, see Box 2.1.  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Alkali Metal Cations in Biochemistry 55 Box 2.1 X-ray Crystallography of Supramolecular Compounds Massa, W. and Gould, R. O., Crystal Structure Determination. 2nd ed.; Springer: Heidelberg, 2004. The primary objective of an X-ray diffraction experiment is to obtain a detailed picture of the contents of the crystal at the atomic level, as if you were viewing it through an extremely powerful microscope. Experi- mentally this process consists of a series of intensity measurements of X-ray beams diffracted from a small, single crystal sample. In the best case the experiment results in a detailed knowledge of the positions of all the atoms in the molecule and hence detailed knowledge of the molecular structure as a whole, including bond lengths, angles and intermolecular contacts. The experiment also gives insight into thermal motion in the solid state. These results are of particular importance to supramolecular chemists because they give direct information on intermolecular interactions such as hydrogen bonding and ion–dipole interactions, as well as the steric fit of a receptor and a substrate (host and guest). Because X-ray crystallography relies upon diffraction by the electron density in the crystal, however, it is not very good at locating hydrogen atoms accurately (after all, they have only one electron each). These are important in hydrogen bonding studies and, if extremely precise information is required, they are usually located using single crystal neu- tron diffraction (see Box 8.1). So, why don’t we just use a very powerful microscope and view atoms directly? The answer is that atoms are simply too small. The wavelength of visible light falls between 400 and 700 nm. In comparison, a typical bond distance between two atoms (e.g. carbon) is about 0.15 nm, so we need radiation of a much smaller wavelength, comparable to that of the interatomic separations: X-ray radiation. The problem with radiation of very short wavelength is that it is impossible to manufacture a lens powerful enough to refocus it. This is referred to as the phase problem. Another way of describing the phase problem is that crystal structure determination calculations require information about the amplitude of diffracted X-rays (the so-called structure factor amplitude, F) which is proportional to the square root of the observed X-ray intensity. Unfortunately F can be both positive and negative and there is no way of directly determin- ing the sign of F from a measurement of intensity, i.e. F 2 because 冨–F冨2 is the same as 冨⫹F冨2. This loss of ⫹/⫺ phase information is the phase problem. The phase problem is solved by developing a mathematical model of the structure based on clues from the experimental data and chemical intuition. This is used to calculate the molecules’ X-ray diffraction patterns. The calculated prediction is compared with the observed experimental data and improved by least-squares refi nement. The experiment is complete when the agreement between calculated and observed data is as good as possible. This agreement is measured by the R (residual) factor and the standard uncertainties (errors, often referred to as estimated standard deviations) on the individual bond lengths and angles. For a good structure determination, the R factor should be around 5 % or less. A Typical Experiment 1. Prepare single crystal sample (slow evaporation, recrystallisation etc.): Homogeneous single crystal without defects or cracks. Size 0.1–0.8 mm edge length, ideally spherical or at least equidimensional. 2. Examine under optical microscope to check for imperfections. 3. Preliminary X-ray photographs (polaroid film or electronic area detector). Check crystal quality. Determine unit cell dimensions and symmetry information (the unit cell is the basic building block of the crystal). 4. Collect intensity data (1000–100 000 data points depending on molecular size). 5. Develop an approximate model of the structure (referred to as solving the structure). Application of Patterson or Direct methods to ‘guess’ approximate phases. Stereochemistry of molecule in outline. Bond lengths to ⫾ 0.1Å (1Å (Ångstrom) ⫽ 10⫺10 m ⫽ 0.1 nm). (continued)  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 56 The Supramolecular Chemistry of Life Box 2.1 (Continued) 6. Refine structure by least squares method. Optimise model (set of atom coordinates and thermal parameters) to get best fit. Bond lengths to ⫾ 0.005Å. Precise stereochemistry. 7. Convert coordinates to tables of bond lengths and angles, and draw picture. 8. Deposit the results in CIF format in the Cambridge Structural Database (CSD, Section 8.4) This process becomes more complicated as the size of the molecule or array to be studied increases. Most supramolecular compounds are of a size intermediate between traditional ‘small molecules’ and large biomolecules such as proteins and it is usually possible to obtain precise information on them fairly read- ily. As the size of the molecule of molecular assembly increases, however, the number of data measured and parameters that must be fitted rises dramatically. This complexity is often accompanied by increasing problems in obtaining good-quality crystalline samples. Large molecules, particularly those encountered in supramolecular chemistry that contain cavities or are of awkward shapes and fit together poorly, often contain occluded solvent molecules. These can diffuse out of the crystal lattice during the X-ray experiment causing loss of crystallinity and hence loss of diffraction intensity. They can also move around resulting in a ‘smeared’ (disordered) averaged electron density, which is difficult to model. Even in cases where solvent is not present, poorly packed crystals result in weak diffraction. The X-ray crystallography of even higher molecular weight samples, such as proteins, is much more complicated, requiring the collection of a number of closely related sets of data, often from many crystals. In protein work, problems are encountered with X-ray damage to the sample, weak diffraction, and difficulty in identifying the various molecular fragments. Both supramolecular and protein crystallography have been revolutionised by the advent of area detectors such as charge-coupled device (CCD) instruments (Figure 2.6) (X-ray detectors that are capable of measuring many data points simultaneously), which enhance vastly both the speed and sensitivity of the experiment. Moreover synchrotron sources such as Diamond in the UK are becoming relatively accessible and are able to obtain single crystal X-ray data on extremely small samples often only a few tens of microns in size. In addition modern X-ray diffraction work, especially that involving hydrogen bonded or solvated supramolecular species, is generally carried out at very low temperatures (100–150 K) to reduce atomic motion and prevent diffusional solvent loss. Figure 2.6 A modern CCD diffractometer. Note the circular area detector on the left, which acts as a very sensitive electronic, reusable equivalent of photographic film, allowing many data points to be collected simultaneously. (Photograph courtesy of Nonius.).  10.1002/9780470740880.ch2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/9780470740880.ch2 by Uva Universiteitsbibliotheek, Wiley Online Library on [16/09/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Alkali Metal Cations in Biochemistry 57 be seen clearly that the interaction of the hydrophilic carbonyl oxygen atoms with the central K⫹ ion causes the lipophilic iso-propyl groups to point outwards thus exposing a primarily hydrocarbon coated outer sheath to the surrounding medium. The remaining amide functionalities act to ‘zip up’ the molecule via intramolecular hydrogen bonds, ensuring the K⫹ ion is encased entirely in a lipophilic exterior as it crosses the membrane. Both valinomycin and nonactin are potent antibiotics because of their ability to perturb transmembrane ionic balance in bacteria. O O O N O O O H O O O D-Hyi D-Val L-Lac L-Val O N N H H O O O H H O N N O nonactin O O O = O valinomycin O O O O O 3 O H O H N N O O H D-Hyi O O O O O Me O N N O O O 3 O R 2.3 O 2.1 2.2 Enniatin A: R = N-methyl-L-isoleucine Enniatin B: R = N-methyl-L-valine Enniatin C: R = N-methyl-L-leucine Baeuvericin: R = N-methyl-L-phenylalanine Related closely to valinomycin are the enniatins (2.3), which are made up of only half the number of amino acid binding units (see Box 2.2 for amino acid structure). The enniatins transport alkali metal cations and alkaline earth metal ions, although they are much less selective than valinomycin. Note that the amide nitrogen atoms are methylated, precluding the possibility of hydrogen bonding. Binding constants for a range of naturally occurring ionophores are given in Table 2.2. Note that many of the ionophores bind strongly to K⫹ while only monesin actually binds Na⫹ with any selectivity. We will return to the origins of the high affinity of these ligands for K⫹ and the factors affecting selectivity in general in the next chapter. Table 2.2 Log K11 values for alkali metal ion binding by naturally occurring ionophores in methanol solvent at 25 oC. Ligand Li⫹ Na⫹ K⫹ Rb⫹ Cs⫹ Valinomycin