Margaret M. Harding

Dynamic combinatorial chemistry (DCC) is a powerful method for the identification of novel ligands for the molecular recognition of receptor molecules. The method relies on self-assembly processes to generate libraries of compounds under reversible conditions, allowing a receptor molecule to select the optimal binding ligand from the mixture. However, while DCC is now an established field of chemistry, there are limited examples of the application of DCC to nucleic acids. The requirement to conduct experiments under physiologically relevant conditions, and avoid reaction with, or denaturation of, the target nucleic acid secondary structure, limits the choice of the reversible chemistry, and presents restrictions on the building block design. This review will summarize recent examples of applications of DCC to the recognition of nucleic acids. Studies with duplex DNA, quadruplex DNA, and RNA have utilized mainly thiol disulfide libraries, although applications of imine libraries, in combination with metal coordination, have been reported. The use of thiol disulfide libraries produces lead compounds with limited biostability, and hence design of stable analogues or mimics is required for many applications.

Antifreeze glycoproteins are an important class of biological antifreezes that have potential applications in many areas of medicine, agriculture and industry in which ice crystal growth is damaging. While the synthesis of antifreeze glycoproteins as pure glycoforms has recently been achieved by using ligation and polymerisation strategies, the routine production of large quantities of pure glycoforms remains challenging. A range of C‐linked analogues that are readily produced by solid‐phase synthesis have delivered novel compounds that are not biological antifreezes, but are potent, non‐cytotoxic, ice‐recrystallisation inhibitors. Structure–activity studies, the identification of cyclic antifreeze glycoproteins and conformational studies have provided further insight into the requirements for antifreeze activity. These results, coupled with significant advances in approaches to the routine synthesis of different glycoproteins and mimics, present opportunities for the design and synthesis of novel ice‐growth‐inhibiting and antifreeze compounds.

The hydrophilic, long-chain diamine PEGda (O,O′-bis(2-aminoethyl)octadeca(ethylene glycol)), when complexed with cis-protected Pt(II) ions afforded water-soluble complexes of the type [Pt(N,N)(PEGda)](NO₃)₂ (N,N = N,N,N′,N′-tetramethyl-1,2-diaminoethane (tmeda), 1,2-diaminoethane (en), and 2,2′-bipyridine (2,2′-bipy)) featuring unusual 62-membered chelate rings. Equimolar mixtures containing either the 16-mer duplex DNA D2 or the single-stranded D2a and [Pt(N,N)(PEGda)]²⁺ were analyzed by negative-ion ESI-MS. Analysis of D2-Pt(II) mixtures showed the formation of 1 : 1 adducts of [Pt(en)(PEGda)]²⁺ , [Pt(tmeda)(PEGda)]²⁺ and the previously-described metallomacrocycle [Pt₂(2,2′-bipy)₂{4,4′-bipy(CH₂)₄4,4′-bipy}₂]⁸⁺ with D2; the dinuclear species bound to D2 most strongly, consistent with its greater charge and aromatic surface area. D2 formed 1 : 2 complexes with the acyclic species [Pt(2,2′-bipy)(Mebipy)₂]⁴⁺ and [Pt(2,2′-bipy)(NH₃)₂]²⁺. Analyses of D2a-Pt(II) mixtures gave results similar to those obtained with D2, although fragmentation was more pronounced, indicating that the nucleobases in D2a play more significant roles in mediating the decomposition of complexes than those in D2, in which they are paired in a complementary manner. Investigations were also conducted into the effects of selected platinum(II) complexes on the thermal denaturation of calf thymus DNA (CT-DNA) in buffered solution. Both [Pt₂(2,2′-bipy)₂{4,4′-bipy(CH₂)₆4,4′-bipy}₂]⁸⁺ and [Pt(2,2′-bipy)(Mebipy)₂]⁴⁺ stabilized CT-DNA. In contrast, [Pt(tmeda)(PEGda)]²⁺ and [Pt(en)(PEGda)]²⁺ (as well as free PEGda) caused negligible changes in melting temperature (ΔT_m), suggesting that these species interact weakly with CT-DNA.

A series of cationic, ditopic N-donor ligands based on 4,4′-bipyridine (4,4′-bipy), 3-aminopyrazine (apyz) and 2-aminopyrimidine (apym), each incorporating two positively-charged N-heterocycles linked by a conformationally-flexible spacer unit, have been synthesised and treated with palladium(II) or platinum(II) precursors [M(2,2′-bipy)(NO₃)₂] (M = Pd(II) or Pt(II)) to form highly cationic metallocyclic species. Treatment of 1,6-bis(4,4′-bipyridinium)hexane nitrate with [M(2,2′-bipy)(NO₃)₂] in aqueous solution, followed by the addition of KPF₆, resulted in the formation of the [2+2] species [M₂(2,2′-bipy)₂{4,4′-bipy(CH₂)₆4,4′-bipy}₂](PF₆)₈. Treatment of [Pd(PhCN)₂Cl₂ ] with 1,3-bis(4,4′-bipyridinium)propane hexafluorophosphate in MeCN afforded [Pd₂Cl₄{4,4′-bipy(CH₂)₃4,4′-bipy}₂](PF₆)₄. When the cationic apyz or apym ligands were used in aqueous solution, the analogous metallomacrocycles did not form. Instead, deprotonation of the exocyclic amino group occurred upon coordination of the ligand to afford a tetranuclear [4+2] species in the case of platinum(II), with Pt(II)⋯Pt(II) bonding supported by strong UV-vis absorption at λ = 428 nm which was assigned to a metal-metal-to-ligand charge transfer (MMLCT) band. Thus, treatment of 1,6-bis(3-aminopyrazinium)hexane nitrate with [Pt(2,2′-bipy)(NO₃)₂], followed by the addition of KPF₆, led to the formation of the red species [Pt₄(2,2′-bipy)₄{apyz(CH₂)₆apyz–2H}₂](PF₆)₈. No related products could be identified with palladium(II), consistent with the low propensity for this metal ion to form strong Pd(II)⋯Pd(II) bonding interactions.

Homogeneous glycopeptide analogues of fish antifreeze glycoproteins of discrete oligomeric length have been synthesised using a native chemical ligation–desulfurisation strategy.

Fish antifreeze proteins and glycoproteins (AF(G)Ps) prevent ice crystal growth and are able to protect mammalian cells and tissues from hypothermic damage in the sub-zero Polar oceans. This protective mechanism is not fully understood, and further data is required to explain how AF(G)Ps are able to stabilize lipid membranes as they pass through their phase transition temperatures. Solid-state NMR spectroscopy was used as a direct method to study the interaction of the 37-residue α-helical type I AFP, TTTT, and the low molecular weight fraction glycoprotein, AFGP8, with dimyristoylphosphatidylcholine membranes above and below the gel-fluid phase transition temperature. In contrast to previous studies in fluid phase bilayers these experiments have provided direct information regarding both the mobility of the phosphate headgroups and perturbation of the acyl chains at a range of temperatures under identical conditions on the same sample. At 5°C changes in the ²H and ³¹P spectra and a dramatic increase in the ³¹P T₁ relaxation times were consistent with a significant disruption of the membrane by TTTT. Heating to 30°C appeared to expel the peptide from the lipid and re-cooling showed that the interaction of TTTT was not reversible. By contrast, ³¹P spectra of the membranes with AFGP8 were consistent with interaction with the phosphate headgroups at both 5 and 30°C. Although both peptides interact with the phospholipid bilayer surface, which may stabilize the membrane at lower temperatures, the longer ³¹P T₁ values and the ²H NMR data obtained for TTTT compared with AFGP8 suggest that TTTT causes a greater reduction of phosphate headgroup mobility and has a greater effect on the lipid acyl chains at 5°C.

The encapsulation of the aquated forms of molybdocene dichloride and titanocene dichloride by cucurbit[n]uril (Q[n], where n = 7 and 8) at different pD values has been studied by ¹H NMR spectroscopy and molecular modelling. ¹H NMR titration experiments indicate that both metallocenes form 1 : 1 host–guest complexes with both Q[7] and Q[8]. In these complexes, both the cyclopentadienyl ligands and metal centre are positioned deep within the cucurbituril cavity. In vitro cell proliferation studies using the cancer cell lines MCF-7 and 2008 showed that the encapsulated molybdocene complex was more active than the corresponding free metallocene, with GI₅₀ values of 210 and 400 μM respectively. However, unexpectedly the encapsulation of Cp₂MoCl_2(aq)at pD 7 catalysed significant degradation of the cucurbituril framework in the presence of oxygen. Encapsulation of Cp₂TiCl_2(aq) by Q[7] greatly slowed the protonolysis of the cyclopentadienyl ligands in aqueous phosphate buffer (pD 7), while encapsulation in Q[8] only slightly retarded the hydrolytic degradation of the metallocene.

The α-helix is the most abundant secondary structural element in proteins and is an important structural domain for mediating protein–protein and protein–nucleic acid interactions. Strategies for the rational design and synthesis of α-helix mimetics have not matured as well as other secondary structure mimetics such as strands and turns. This perspective will focus on developments in the design, synthesis and applications of α-helices and mimetics, particularly in the last 5 years. Examples where synthetic compounds have delivered promising biological results will be highlighted as well as opportunities for the design of mimetics of the type I α-helical antifreeze proteins.