Discrimination Between T/A and A/T Base Pairs of Pyrrole‐Imidazole Polyamides Substituted with Chiral Β‐hydroxy‐γ‐aminobutyric Acid/β‐Alanine Pairs
Wen Zhang,Shi-Kun Jiang,Yan-Ling Wu,Chuan-Xin Guo,Hong-Fei Zhang,Hiroshi Sugiyama,Xing-Lai Chen
DOI: https://doi.org/10.1002/cbic.201100675
IF: 3.2
2011-01-01
ChemBioChem
Abstract:Hairpin polyamides are a class of small synthetic molecules that currently have the greatest potential to become gene-targeting drug candidates. Recently, studies on in vivo gene regulation by polyamides have attracted considerable attention. k, 2] They are mainly composed of N-methylpyrrole (Py), N-methylimidazole (Im), and N-methyl-3-hydroxypyrrole (Hp) amino acids, and aliphatic chain compounds including aliphatic amino acids. To date all four Watson–Crick base pairs can be recognized by using the “discrimination code” developed by Dervan and colleagues, that is, a Py/Im pair specifying C·G and an Im/Py pair for G·C. A Py/Py pair non-preferentially binds T·A and A·T, while an Hp/Py pair is able to discriminate T·A from A·T. But there are some disadvantages to using the replacement of Py with Hp for specifying T. Although some excellent groups (such as the Dervan group) have made considerable attempts to explore an optimal substitute for Hp in the design of polyamides, such as hydroxybenzimidazole (Hz), no T-specific recognition element better than Hp has been found. On the other hand, polyamides composed of more than five contiguous aromatic rings are too curved to fit the curvature and twist of the DNA minor groove, and this leads to loss of affinity and specificity of binding. This is a dominant factor which excludes longer polyamides from discriminating longer DNA sequences (i.e. , those seldom repeated within the 3 billion basepair human genome). Therefore, solving this problem is another research focus in this area. Incorporation of flexible b-alanine (b) into polyamides allows longer crescent-shaped polyamides that match the curvature and twist of longer DNA sequences (containing 16 base pairs), with association equilibrium constant (KA) of 3 10 10 m . However, T·A and A·T still cannot be discriminated by either b/b or Py/b (or b/Py) pairings. In order to find a recognition subunit that not only acts as a T-specific recognition element, but also acts like “spring” to regulate length and curvature of the polyamides to fit the DNA minor groove, some chirally modified aliphatic amino acids, such as (S)-isoserine (Is), (R)-isoserine (Is), (S)-a-fluoro-balanine (Fb), and (S)-b-aminoalanine (Aa) have been incorporated into polyamides and assessed as T recognition elements by Dervan’s group. Their results indicated that Is/b favorably binds T·A over A·T, with a modest specificity of 7.2-fold. In addition, a great effect for chiral (R/S)-a/b-NH2-substituted gturns on recognition affinity and specificity (to T·A, A·T, C·G, and G·C) at the turn position was found. 10] Also, in our studies, the significant binding preference of an (R/S)-a-hydroxyl-gaminobutyric acid (gRO or gSO) or (R/S)-a,g-diaminobutyric acid (gRN or gSN) residue pairing with b-alanine to T·A, A·T, C·G and G·C in the middle positions of hairpin polyamides and at the turn position was revealed by surface plasmon resonance (SPR) and molecular simulation (MSim) techniques. Collectively, the reported data demonstrate that the stereochemistry and nature of aliphatic amino acids residues significantly influence binding characteristics, including affinity and specificity, in polyamide–DNA recognition events. The above exciting findings inspired us to further explore the possibility of (R/S)-b-hydroxyl-g-aminobutyric acids (gb-OH, gb-OH) as T-specific recognition elements. Earlier studies on (R/ S)-a-hydroxyl-g-aminobutyric acids as T recognition and as Gand T-specific recognition elements, 3c] found hydrogen bond formation between the Im N3 and guanine N2, and between Hp C3-OH and thymine O2, respectively; furthermore, the amount of covalent bonding from the C terminus to N terminus and the hydroxyl position for the gb-OH moieties is fully in accord with the backbone carbon properties of Hp, facing the core of DNA as shown in Scheme 1. Consequently, we have designed and synthesized five novel polyamides containing b/b (1), b/gb-OH (2), b/ gb-OH (3), gb-OH/b (4) and gb-OH/b (5) pairings. We studied their binding properties to predetermined DNA sequences, including binding affinity, specificity, stoichiometry, kinetics, and thermodynamics; we used equilibrium analysis (association equilibrium constant (KA) or dissociation equilibrium constant (KD)) and kinetic analysis (association rate constant (ka) or dissociation rate constant (kd)) by using Biacore SPR and MSim. Surprisingly, to our knowledge, there have been no reports on the application of these two chiral amino acids in DNA-specific recognition by polyamides. In order to evaluate the selective recognition ability of novel polyamides 1 with a b/b pairing and 2–5 with a chiral gb-OH/b or b/gb-OH pairing to Watson–Crick base pairs T·A, A·T, C·G, [a] Prof. W. Zhang, S.-K. Jiang, C.-X. Guo, H.-F. Zhang, X.-L. Chen Department of Biopharmaceutical Sciences College of Pharmaceutical Sciences, Zhejiang University of Technology 18 Chaowang Road, Hangzhou 310014 (China) E-mail : wzhang63@zjut.edu.cn [b] Prof. Y.-L. Wu Virus Inspection Department Zhejiang Provincial Center for Disease Control and Prevention 630 Xincheng Road, Hangzhou 310051 (China) E-mail : yanlingw@yahoo.cn [c] Prof. H. Sugiyama Department of Chemistry, Graduate School of Science Kyoto University Sakyo-ku, Kyoto 606-8502 (Japan) Supporting information for this article is available on the WWW under http ://dx.doi.org/10.1002/cbic.201100675.