Design of DNA minor groove binding diamidines that recognize GC base pair sequences: a dimeric-hinge interaction motif

Abstract

The classical model of DNA minor groove binding compounds is that they should have a crescent shape that closely fits the helical twist of the groove. Several compounds with relatively linear shape and large dihedral twist, however, have been found recently to bind strongly to the minor groove. These observations raise the question of how far the curvature requirement could be relaxed. As an initial step in experimental analysis of this question, a linear triphenyl diamidine, DB1111, and a series of nitrogen tricyclic analogues were prepared. The goal with the heterocycles is to design GC binding selectivity into heterocyclic compounds that can get into cells and exert biological effects. The compounds have a zero radius of curvature from amidine carbon to amidine carbon but a significant dihedral twist across the tricyclic and amidine-ring junctions. They would not be expected to bind well to the DNA minor groove by shape-matching criteria. Detailed DNase I footprinting studies of the sequence specificity of this set of diamidines indicated that a pyrimidine heterocyclic derivative, DB1242, binds specifically to a GC-rich sequence, -GCTCG-. It binds to the GC sequence more strongly than to the usual AT recognition sequences for curved minor groove agents. Other similar derivatives did not exhibit the GC specificity. Biosensor-surface plasmon resonance and isothermal titration calorimetry experiments indicate that DB1242 binds to the GC sequence as a highly cooperative stacked dimer. Circular dichroism results indicate that the compound binds in the minor groove. Molecular modeling studies support a minor groove complex and provide an inter-compound and compound-DNA hydrogen-bonding rational for the unusual GC binding specificity and the requirement for a pyrimidine heterocycle. This compound represents a new direction in the development of DNA sequence-specific agents, and it is the first non-polyamide, synthetic compound to specifically recognize a DNA sequence with a majority of GC base pairs.

Figure 1

Compound structures and DNA oligomer sequences used in this study.

Figure 2

DNase I footprinting titration experiments. The pBS plasmid was isolated and purified from E. coli using Qiagen columns. The 265 bp DNA fragment was prepared by 3′-[³²P]-end labeling of the EcoRI-PvuII double digest of the pBS plasmid (Stratagene) using α-[³²P]-dATP and AMV reverse transcriptase. (A) The products of the DNase I digestion were resolved on an 8% polyacrylamide gel containing 8 M urea. Drug concentrations are at the top of the lanes. Tracks labeled G represent dimethylsulfate piperidine markers specific for guanines. Differential cleavage plots compare the susceptibility of the DNA to cutting by DNase I in the presence of (B) DB75 and DB1242 (C) DB75, DB1111, DB1164 and DB1228. Deviation of points toward the lettered sequence (negative values) corresponds to a ligand-protected site and deviation away (positive values) represents enhanced cleavage. The vertical scale is in units of ln(fa) - ln(fc), where fa is the fractional cleavage at any bond in the presence of the drug and fc is the fractional cleavage of the same bond in the control.

Figure 3

SPR sensorgrams for DB1242 with (A) -GCTCG- and (B) -AATT- hairpin DNA (figure 1). The compound concentrations were 0.1, 0.3, 0.5, 1.0, 1.4, 1.8, 2.2, 2.4, 2.8, 3.5, 4.0 to 4 μM from bottom to top in (A) and (B). The experiments were carried out in cacodylic acid buffer at 25 °C.

Figure 4

RU values from the steady state region of SPR sensorgrams are plotted against the unbound compound concentration, C_f (flow solution): (A) DB1242 and DB1164 binding to -GCTCG-DNA hairpin and (B) DB1242 and DB1111 with the AATT DNA hairpin. The data in A were fitted to a two site model and that in B were fitted to one site model using equation 1.

Figure 5

ITC curves (12μM hairpin duplex) for the binding of DB1242 to the (A) –GCTCG- and (B) -AATT-hairpin; DB1111 to the (C) -GCTCG- and (D) -AATT- hairpin; DB1164 to the (E) -GCTCG- and (F) -AATT- hairpin. In each panel the top plot is the baseline corrected experimental data. For the lower plots results were converted to molar heats and plotted against the compound to DNA molar ratio. The same buffer conditions were used as in Figure 3.

Figure 6

Difference CD spectra for DB1242, DB1111 and DB1164. In each panel a CD spectrum for DNA is shown along difference CD spectra for the titration of compound into the DNA solution. (A) DB1242 with -AATT-: ratios of the compound to DNA hairpin from bottom to top at 300 nm are 0.5, 2.0, 2.5 and 3.0. (B) DB1242 with -GCTCG-: ratios of compound to DNA from bottom to top are 0.4, 0.8, 1.2, 1.6, 2.0, 2.4, 2.8 (C) DB1111and -AATT-: ratios of compound to DNA from bottom to top are 0.5, 1.0, 1.5 (D) DB1111 with -GCTCG-: ratios of compound to DNA from bottom to top are 0.5, 1.0, 2.0, 3.0 (E) DB1164 with -GCTCG-: ratios of compound to DNA from bottom to top are 1.0, 1.5, 2.0, 2.5, 3.0. The same buffer conditions were used as in Figure 3. Because of the relatively

Figure 7

(A) Geometry-optimized models for DB1242. A color coded electron density map is on the left and a space-filling model with atom colors is displayed on the right (carbon-black, nitrogen-blue, and hydrogen-white). (B) Stacked dimer models for DB1242: (i) stick model and (iii) space filling model. (C) Flexidock generated DB1242 dimer (green) in complex with DNA minor groove binding site-GCTCG-. A representative low energy structure is shown for this DNA dimer complex. The 3′G is presented in magenta and the 5′G is shown in yellow; key hydrogen bonds are displayed in white. Note that two hydrogen bonds are formed between the molecules of the dimer. (D) Flexidock generated DB1111 dimer (green) in complex with DNA minor groove binding site-GCTCG-. Example structures are shown for this DNA complex with hydrogen bonds in white. Note that there are no hydrogen bonds between the dimer molecules.

Figure 7

(A) Geometry-optimized models for DB1242. A color coded electron density map is on the left and a space-filling model with atom colors is displayed on the right (carbon-black, nitrogen-blue, and hydrogen-white). (B) Stacked dimer models for DB1242: (i) stick model and (iii) space filling model. (C) Flexidock generated DB1242 dimer (green) in complex with DNA minor groove binding site-GCTCG-. A representative low energy structure is shown for this DNA dimer complex. The 3′G is presented in magenta and the 5′G is shown in yellow; key hydrogen bonds are displayed in white. Note that two hydrogen bonds are formed between the molecules of the dimer. (D) Flexidock generated DB1111 dimer (green) in complex with DNA minor groove binding site-GCTCG-. Example structures are shown for this DNA complex with hydrogen bonds in white. Note that there are no hydrogen bonds between the dimer molecules.

Figure 7

(A) Geometry-optimized models for DB1242. A color coded electron density map is on the left and a space-filling model with atom colors is displayed on the right (carbon-black, nitrogen-blue, and hydrogen-white). (B) Stacked dimer models for DB1242: (i) stick model and (iii) space filling model. (C) Flexidock generated DB1242 dimer (green) in complex with DNA minor groove binding site-GCTCG-. A representative low energy structure is shown for this DNA dimer complex. The 3′G is presented in magenta and the 5′G is shown in yellow; key hydrogen bonds are displayed in white. Note that two hydrogen bonds are formed between the molecules of the dimer. (D) Flexidock generated DB1111 dimer (green) in complex with DNA minor groove binding site-GCTCG-. Example structures are shown for this DNA complex with hydrogen bonds in white. Note that there are no hydrogen bonds between the dimer molecules.