Molecular Recognition Studies on Naphthyridine Derivatives

The association constants Kb of three hosts I–III designed to have both enhanced hydrogen bonding donor strength and conformational preorganization with biotin analogues 1–5 are reported. 1H-NMR titrations under two different concentration conditions have been employed to determine the association constants Kb. A statistical analysis using a presence absence matrix has been applied to calculate the different contributions. Hydrogen bond interactions make naphthyridine derivatives II and III potent binders and effective receptors for (+)-biotin methyl ester (1), due to the complex stabilization by additional hydrogen bonds.


Introduction
Amide N-H groups have been used to produce a wide range of receptors capable of coordinating biologically important molecules and anions [1,2]. General reviews covering anion receptors containing amide groups have been published recently [3,4]. Due to their singular stereoelectronic character, they interact with electron deficient centers through the carbonyl group and with electron rich centers through their N-H units; this dual feature has been successfully used for the design of amide-based receptors able to recognize a large variety of guests [1]. On the other hand the recognition OPEN ACCESS capabilities of fused-pyridine and naphthyridine hosts remains an important challenge of supramolecular chemistry [5].

III
Guests 1 [14] and 2-5 [15] were synthesized according to described procedures with slight variations (Figure 2). The first step was the preparation of biotin methyl ester 1 by acid-catalyzed esterification of biotin. Selective reduction of 1 using DIBAL at −78 ºC afforded alcohol 2 in 73% yield. The iodide 3 was prepared from biotin tosylate by halide substitution with NaI in acetone. The reaction of 3 with LiBr in 2-butanone yielded the formation of bromide 4 in 87% yield. Finally, alkyne 5 was obtained in high yield by the substitution reaction of bromide 4 with lithium acetylideethylenediamine in DMSO at 15 ºC.
The ability of receptors containing pyridine or naphthyridine moieties I-III to recognize and bind the aforementioned guests can be evaluated using 1 H-NMR spectroscopy. The stoichiometry of the complexes must be determined beforehand to ensure use of the right equations in the titrations.  We have used the method of continuous variation to generate Job plots [16] by preparing different mixtures of receptor and guest covering the whole range of molar fractions of the host but keeping constant the total concentration of the solutions (10 mM). The plot of the product between the increment in the chemical shift and the receptor concentration versus the molar fraction of the receptor affords a curve. From the value of the maximum, which can be obtained by means of equation X = m/(m+n) [17], the stoichiometry of the complex is determined. For all the compounds used in this study we always obtained a 1:1 stoichiometry (see Figure 3 for an example). NMR titrations [18] were perfomed at two different experimental concentrations [6]: the non saturation titration conditions (nST) in the 0.2-08 range and the saturation titration conditions (ST).

Binding Studies
Association constants K b for hosts I-III with biotin derivatives 1-5 as guests were determined by 1 H-NMR titration experiments in CDCl 3 using the EQNMR software to fit the curves to a 1:1 binding model [19]. The values depicted in Table 1 have been calculated from the chemical shift induced effects of the N-H groups of receptors I-III. Table 1. Association constants K b (M -1 ) for I-III binding 1-5 measured at 300 K in CDCl 3 (Errors ≤ 10%). We have already reported three K b values of Table 1 using the non Saturation Titration conditions (nST): I·1 3,800 ± 500 [6], II·1 35,000 ± 5,250 [6], and III·1 148,000 ± 20,000 [8]. The new values are slightly lower than those determined previously -(0.99 ± 0.04)-fold on average. Addition of biotin analogues 1-5 to I-III in CDCl 3 , results in downfield shifts of the N-H groups of receptors I-III due to host-guest hydrogen bonds interactions. The selectivity displayed by all three receptors I-III is similar to the trend CO 2 Me >> Br ≥ I ≥ acetylide ( Figure 4). The receptors II and III have high association constants (6.7 × 10 4 and 2.5 × 10 5 M -1 , respectively) with (+)-biotin methyl ester (1), presumably due to the formation of additional hydrogen bonds with the receptors involving the methyl ester that could further stabilize the complex. In addition, the biotin analogues with bulky groups (Br, I and acetylide) will destabilize the complex due to steric repulsion and that seems to be the most important factor influencing the relative poor association constants obtained for these guests.
The biotin analogue 2 was insoluble in CDCl 3 , therefore in order to compare its behavior with that of the other guests, the 1 H-NMR titration was performed in a 10% MeOD-90% CDCl 3 solution. As control the (+)-biotin methyl ester (1) was titrated under the same conditions as receptor III. The obtained association constants K b were: 100 M -1 for 1 and 110 M -1 for 2, suggesting that the latter guest interacts with receptor III in a similar manner as it does 1. On the contrary, guest 2 does not interact with receptors I and II.

Data analysis
Finally a more quantitative approach on the association constant values has been attempted. First, we have multiplied by 1.1 (from 100 to 110) all the K b values of Table 1 (Table 2). This procedure corresponds to a weighted mixture of both values. Then we have built a presence/absence matrix (Table 2), known in medicinal chemistry as a Free-Wilson model [20,21]. If we assume that K b = H × G, then, ln K b = ln H + ln G. From Table 2 the following contributions can be calculated: ln HI = 7.7, ln HII = 8.5, ln HIII = 9.0, ln G1 = 2.4 and ln G2 = 2.5 (the other G terms are not significant, ln = 0). These values correspond to HI = 2,253, HII = 4,915, HIII = 7,708, G1 = 11.5 and G2 = 12.4 (other G terms = 1).

General
Unless otherwise reported, all reactions were carried out under dry and deoxygenated argon atmospheres. Solvents were freshly distilled and dried before use by standard methods. 1 H-and 13 C-NMR spectra were recorded on a Bruker DRX 400 spectrometer (9.4 Tesla, operating at 400.13 MHz for 1 H and 100.62 MHz for 13 C, respectively) with a 5-mm inverse detection H-X probe equipped with a z-gradient coil, at 300 K. Chemical shifts (δ, in ppm) are given from internal solvent CDCl 3 (7.26 for 1 H and 77.0 for 13 C) and DMSO-d 6 (2.49 for 1 H and 39.5 for 13 C). 2D gs-COSY ( 1 H-1 H) and 2D inverse proton detected heteronuclear shift correlation spectra [gs-HMQC ( 1 H-13 C) and gs-HMBC ( 1 H-13 C)] were acquired and processed using standard Bruker NMR software and in non-phase-sensitive mode and were carried out to assign the 1 H and 13 C signals without ambiguity. (1) [14]. A suspension of D-(+)-biotin (1g, 4.1 mmol) and H 2 SO 4 (3 mL) in methanol (20 mL) was heated under reflux for 12 h. Solvents were evaporated under reduced pressure and the residue was poured into ice/water. The precipitated material was filtered, washed with water and dried, to give 1 as a white solid (850 mg, 81%). (2) [15]. Compound 1 (500 mg, 1.93 mmol) was dissolved in CH 2 Cl 2 (30 mL). The mixture was cooled to −78 ºC and DIBAL (1.0 M, 6.8 mL, 6.78 mmol) was added; the resulting solution was stirred for 2 h at r.t. The mixture was then cooled to -78 ºC and quenched with MeOH. The solvent was evaporated under reduced pressure and the residue was extracted with EtOH using a Soxhlet, to give 2 as a white solid (320 mg, 73%). 1 [3,4-d]imidazol-2(3H)-one (3) [15]. This compound was prepared from biotin tosylate as reported [22]. The biotin tosylate (280 mg, 0.75 mmol) and NaI (216 mg, 0.15 mmol) were stirred in acetone (20 mL) for 24 h. The solvent was removed under reduced pressure and the residue was dissolved in CH 2 Cl 2 and the organic layer was successively washed with sodium thiosulfate (1.0 N) and water, dried over anhydrous Na 2 SO 4 , and concentrated under vacuum. Purification of the crude material by column chromatography on silica gel with CH 2 Cl 2 /MeOH (5%) gives 3 as a white solid (335 mg, 78%). 1 [3,4-d] imidazol-2(3H)-one (4) [15]. A solution of 3 (800 mg, 2.35 mmol) and LiBr (1.03 g, 11.76 mmol) in 2-butanone (30 mL) was stirred at 80 ºC for 24 h. After cooling at room temperature, 20 mL of 10% NaHSO 3 were added to quench the reaction. The organic layer was washed with water, dried (MgSO 4 ), and evaporated to dryness. The residue was purified by chromatography on silica gel with CH 2 Cl 2 /MeOH (5%) to give 4 as a white solid (595 mg, 87%). 1 [3,4-d]imidazol-2(3H)-one (5) [15]. Lithium acetylide-ethylendiamine complex (194 mg, 2.1 mmol) was suspended in DMSO (5 mL) and cooled to 15 ºC. A solution of 4 (200 mg, 0.68 mmol) in DMSO (3 mL) was added and the mixture was stirred for 1.5 h. The reaction mixture was poured over an ice/brine and the crude material extracted with CH 2 Cl 2 . The organic layer was washed with water, dried (MgSO 4 ), and evaporated to dryness. The residue was purified by chromatography on silica gel with CH 2 Cl 2 /MeOH (5%) to give 5 as a white solid (146 mg, 90%). 1

NMR titrations
Association constants were calculated through 1 H-NMR titration experiments in CDCl 3 (and in some cases by adding a 10% MeOD to the CDCl 3 ) using the EQNMR software to analyze the results [19]. Changes in the chemical shifts of the N-H groups of receptors were used. With respect to the non Saturation Titration conditions (nST) we proceeded as described already by ourselves [6][7][8][9]. To perform the measurement of K b , the host and guest solutions were prepared in a volumetric flask of the appropriate volume (2 mL to 5 mL). 0.5 mL of the host solution I-III was taken, put in a NMR tube and successive additions of solutions of guests 1-5 aliquots were made. The number of additions was continued until saturation titration conditions (ST) or until reached the 0.8 value of saturation (nST). We present here a typical example of the experimental data for ST and nST conditions (Table 3 and  Table 4).

Conclusions
The naphthyridine receptors II and III exhibit a high selectivity towards the (+)-biotin methyl ester (1), due to the stabilization of the complex by additional hydrogen bonds, with an association constant K b = 6.7 × 10 4 and 2.5 × 10 5 M -1 , respectively, in CDCl 3 . These studies illustrate the capability to modulate the association constant of receptors depending on the biotin type analogue. We have found that the selectivity follows the sequence: CO 2 Me >> Br ≥ I ≥ acetylide. The simple quantitative model (Table 2) allows to go a step further. The hosts classify in the order: III (our host) [8] > II (Goswami host) [13] > I (Thummel host) [12]. Concerning the guests, the order is G2 ≥ G1 ((+)-biotin methyl ester) >> Br ≈ I ≈ acetylide. Although G2 is insoluble in chloroform, it has an affinity for host III slightly better than G1. Possibly the terminal OH anchors to the host. This make G2 a promising candidate for further studies, for instance, modifying the polymethylene chain length.