Introduction

The 2-aminothiazole moiety is a synthetically flexible and pharmacologically promising scaffold in medicinal chemistry, and a number of drugs containing a 2-aminothiazole-4-subsituted moiety, e.g. cefdinir (antibiotic), mirabegron (β3 adrenergic agonist) or the tyrosine kinase inhibitor dasatinib, are on the market. In recent years, anticancer, antiepileptic, neuroprotective, antidiabetic, antihypertensive, anti-inflammamory, antiviral, antibiotic and antileishmanial properties of 2-aminothiazoles were investigated [1].

N,4-Diaryl substituted 2-aminothiazoles were prepared based on one of the ten scaffolds with antileishmanial properties from a screening of 200,000 compounds [2]. Other microorganisms that are inhibited by this class of compounds include plasmodia [3] and mycobacteria [4]. For mycobacteria, the Tuberculosis Antimicrobial Acquisition and Coordinating Facility discovered an aminothiazole cluster of active compounds that formed the basis of an extensive structure–activity relationship (SAR) study [5]. Makam and Kannan showed that several substituted 2-aminothiazole derivatives exhibited antimycobacterial activity against Mycobacterium tuberculosis, H37Rv, with minimum inhibitory concentration (MIC) values of 6.25–12.50 μM and proposed that these may be targeting the KasA protein in turn disturbing the cell wall biosynthesis by obstructing mycolic acid synthesis [6]. Another study explored structure–activity relationships for 2-aminothiazoles as potassium channel blockers [7], an undesirable pharmacological effect in most cases, which needs to be prevented by modification of the substitution pattern on the 2-aminothiazole moiety. While N-alkyl and N-aryl 2-aminothiazoles are distinguished by therapeutically very desirable acitivities, they are also known to be cytotoxic [5]. This is most likely due to the unsubstituted 5-position, which unspecifically engages in redox reactions in biochemical pathways. In the course of our ongoing investigations towards reduced toxicity of this compound class, we structurally characterized 2-aminothiazoles by X-ray crystallography to gain information on the conformational preferences in the solid-state. The extent of the conjugated system and the steric accessibility of the sulfur atom and 5-position is expected to have an influence on the stability, particularly with regard to oxidation.

Synthesis of 2-amino-4-substituted thiazoles can be accomplished via different routes [8]. Hantzsch synthesis from α-haloketones and thiourea derivatives in polar solvents is a general method [9, 10]. From 1-(4-methylpyridin-2-yl)thiourea (1) and the respective α-bromoketone, we synthesized the two 2-aminothiazoles N-(4-methylpyridin-2-yl)-4-(pyridin-2-yl)thiazol-2-amine (2) and N-(4-methylpyridin-2-yl)-4-(pyrazin-2-yl)thiazol-2-amine (3) through Hantzsch synthesis (Scheme 1). Compound 2 is contained in the Stasis Box (Medicines for Malaria Venture, MMV, Geneva, Switzerland) as MMV006357 and was identified as an antimycobacterial [4, 11] agent and potential drug candidate for eumycetoma [12]. Although coordination compounds bearing 2 as ligand were reported more than 40 years ago [13], to the best of our knowledge and based on a WebCSD search in June 2020 [14], small molecule crystal structures of 2, its pyrazine derivative 3 and the starting thiourea derivative 1 have not been reported so far. Likewise, a search of the Protein Data Bank [15] yielded no structures containing 2 or 3 as ligands. Herein, we describe the structures of two polymorphs of 1, henceforth named 1α and 1β, and the crystal and molecular structures of 2-aminothiazoles 2 and 3, as determined by X-ray crystallography.

Scheme 1
scheme 1

Synthesis of aminothiazoles 2 and 3 from 1 and the respective α-bromoketone

Full size image

Experimental Section

General

Starting materials were purchased and used as received. Solvents were of analytical grade. The syntheses of 1-benzoyl-3-(4-methylpyridin-2-yl)thiourea [16] and 2-bromo-1-(pyridin-2-yl)ethanone hydrobromide [17] can be found in the literature.

Physical Methods

1H and 13C NMR spectra were recorded at room temperature on a Varian INOVA 500 NMR spectrometer. The residual solvent signals of DMSO-d6 (δ1H = 2.50 ppm, δ13C = 39.51 ppm) were used to reference the spectra (s = singlet, bs = broad singlet, d = doublet, dd = double doublet). The high-resolution mass spectrum was measured on a Bruker Daltonics Apex III FT-ICR mass spectrometer.

Synthesis and Crystallization

1-(4-Methylpyridin-2-yl)thiourea (1)

Compound 1 was synthesized by adapting a literature protocol [18]. 7.4 mL of 1 N aqueous NaOH were added to a stirred suspension of 1-benzoyl-3-(4-methylpyridin-2-yl)thiourea (2.00 g, 7.37 mmol) in 15 mL of methanol. The mixture was then heated to reflux for 1 h. After cooling to room temperature, a white solid formed, which was separated by filtration, washed with deionized water and dried over P2O5 in a vacuum desiccator to yield 1.00 g of 1 (5.98 mmol, 81%). Physical properties were in agreement with those reported in the literature [19]. Crystals of 1α suitable for X-ray diffraction were obtained by recrystallization from methanol, and those of 1β were grown from a solution in CDCl3 by slow evaporation of the solvent.

N-(4-Methylpyridin-2-yl)-4-(pyridin-2-yl)thiazol-2-amine (2)

Compound 2 was synthesized following a modified literature procedure [20]. Compound 1 (83 mg, 0.50 mmol) and 2-bromo-1-(pyridin-2-yl)ethanone hydrobromide (140 mg, 0.50 mmol) were dissolved in 5 mL of ethanol and triethylamine (0.1 mL) was added. The reaction mixture was refluxed for 2 h. Subsequently, the solvent was removed using a rotary evaporator. The residue was taken up in 10 mL of a saturated K2CO3 solution and extracted three times with ethyl acetate. The combined organic layers were washed with brine and dried over MgSO4. The solvent was removed under reduced pressure and the crude product recrystallized from methanol. Yield: 54 mg (0.20 mmol, 40%). Spectroscopic properties were in agreement with those reported in the literature [2, 20]. Crystals for X-ray diffraction were taken from the mother liquor.

N-(4-Methylpyridin-2-yl)-4-(pyrazin-2-yl)thiazol-2-amine (3)

Compound 3 was prepared in analogy to 2 from 1 and 2-bromo-1-(pyrazin-2-yl) ethanone hydrobromide [21] (note that the compound was not named hydrobromide by these authors), which was synthesized from acetylpyrazine (0.50 g, 4.00 mmol) using 2-pyrrolidinone hydrotribromide (2.18 g, 4.4 mol) as reagent and used in situ without purification. Yield (based on 1): 75 mg (0.28 mmol, 7%). 1H NMR (500 MHz, DMSO-d6): δ = 11.46 (s, 1H), 9.14 (d, J = 1.5 Hz, 1H), 8.64 (dd, J = 2.5, 1,5 Hz, 1H), 8.55 (d, J = 2.5 Hz, 1H), 8.17 (d, J = 5.2 Hz, 1H), 7.75 (s, 1H), 6.89 (bs, 1H), 6.79 (dd, J = 5.2 Hz, 1H), 2.28 (s, 3H) ppm; 13C NMR (126 MHz, DMSO-d6): δ = 160.5, 151.8, 148.7, 147.7, 146.2, 146.0, 144.4, 143.3, 141.4, 117.7, 111.7, 110.7, 20.7 ppm. HRMS(ESI): calcd. for C13H11N5S [M + H]+ 270.0813, found 270.0806. Crystals for X-ray diffraction were taken from the mother liquor.

Crystal Structure Determination

The X-ray intensity data for 1α and 3 were measured on a Bruker AXS Apex II diffractometer, equipped respectively with an Incoatec IµS microfocus X-ray source and a FR591 rotating anode radiation source. The diffraction data for 2 were collected on an Enraf–Nonius KappaCCD diffractometer with a FR591 rotating anode. The SAINT software was used to perform data reductions [22]. The intensity measurements for 1β were carried out on the P11 beamline at the PETRA III light source (DESY, Hamburg) at an X-ray energy of 22.0 keV. The primary beam intensity was monitored continuously and stored during the experiment. The P11 X-ray optics consisted of a liquid nitrogen-cooled Si(111) and Si(113) double-crystal monochromator and one vertical and two horizontal deflecting X-ray mirrors. The source brilliance at the crystal was 1.7 × 1012 photons per second. The data were collected using a 200 μm beam on a PILATUS 6 M-0109 detector (Dectris Ltd, Baden, Switzerland) [23] at a distance of 163.4 mm from the crystal. The 20-bit dynamic range of the PILATUS 6 M detector allowed for collection of weak high-order and stronger low-order reflections at the same time in one run. The crystal was rotated by 360° in steps of 0.5° with an exposure of 0.250 s per frame with a filter transmission of 0.1 using the P11 Crystallography Control graphical user interface at the P11 beamline [24]. The data were processed with the XDS program package [25]. Absorption corrections were carried out with SADABS [26].

The crystal structures were solved with SHELXT-2018/1 [27] and refined with SHELXL-2018/3 [28]. The structure of 1β was refined using aspherical atomic scattering factors [29] and corrected for dispersion according to Kissel and Pratt [30]. Asphericity parameters were generated by the APEX3 software (IDEAL) [31]. Anisotropic displacement parameters were introduced for all non-hydrogen atoms. For 1α, 2 and 3, carbon-bound hydrogen atoms were placed at geometrically calculated positions with Caromatic–H = 0.95 Å and Cmethyl–H = 0.98 Å and refined with the appropriate riding model. Methyl groups were allowed to rotate to match the underlying electron density maxima. Hydrogen atoms attached to nitrogen were localized in difference electron density maps and, for 1α, 2 and 3, refined with the N–H bond lengths restrained to a target value of 0.88(2) Å. Uiso(H) = 1.2 Ueq(C, N) (1.5 for methyl groups) was applied for all hydrogen atoms. Packing indices were calculated with PLATON [32]. Structure pictures were generated with Diamond [33] and Mercury [34]. Crystal data and refinement details for 1α, 1β, 2 and 3 are listed in Table 1.

Table 1 Crystal data and refinement details for 1α, 1β, 2 and 3
Full size table

Results and Discussion

Polymorphic Forms of 1

Two monoclinic crystal forms of 1 were encountered. 1α with one molecule in the asymmetric unit (Z′ = 1) formed upon recrystallization from methanol, and 1β with two crystallographically unique molecules (Z′ = 2) crystallized from a solution of the compound in CDCl3. Figure 1 shows displacement ellipsoid plots of both structures, and Table 2 compares selected bond lengths and torsion angles. The bond lengths are comparable with those in the parent 1-(pyridine-2-yl)thiourea (CSD refcode: HIRPAA) [35]. Both 1α and 1β have in common a six-membered ring intramolecular N–H⋯N hydrogen bond between the primary amino group on the thiourea moiety and the pyridine nitrogen atom, which is in accord with Etter’s second hydrogen bond rule for organic compounds, namely that six-membered ring intramolecular hydrogen bonds form in preference to intermolecular hydrogen bonds [36]. The graph-set assignment for this hydrogen bond motif is S(6). This requires a synperiplanar conformation between the pyridine nitrogen atom and the thiocarbonyl carbon atom C1, and likewise between the pivot carbon atom C2 of the pyridine ring and the primary amino group. The same intramolecular hydrogen bond and molecular conformation was found in HIRPAA. Like in HIRPAA, the molecule is virtually planar in 1α (r.m.s. deviation 0.0234 Å for the non-hydrogen atoms). In contrast, the two unique molecules in 1β deviate markedly from planarity by a tilt of the thiourea group from the plane of the pyridine ring, as revealed by the respective angles between the mean planes of the two groups [molecule 1: 14.16(5)°; molecule 2: 18.08(5)°] and the torsion angles listed in Table 2.

Fig. 1
figure 1

a Hydrogen-bonded association of two symmetry-related molecules in 1α and b of the two crystallographically distinct molecules in 1β. Symmetry code: (a) x, 0.5 − y, 0.5 + z. Displacement ellipsoids are drawn at the 50% probability level. Hydrogen atoms are shown as small spheres of arbitrary radii. Dashed lines represent hydrogen bonds

Full size image
Table 2 Selected bond lengths (Å) and torsion angles (°) for 1α and 1β
Full size table

1α and 1β exhibit distinctly different intermolecular hydrogen bonding patterns. Two self-complementary N–H⋯S hydrogen bonding sites on each molecule, provided by the sulfur atom as hydrogen bond acceptor and each of the primary and secondary amino groups as hydrogen bond donor sites, can lead to three possible combinations with \({\text{R}}_{2}^{2}(8)\) motifs (Scheme 2) [37]. As shown in Fig. 1, 1α exhibits the asymmetric combination involving primary and secondary amino groups (type I), whereas 1β shows the symmetric combination formed solely by the primary amino groups (type II). Type I is also observed for the structure of HIRPAA. Table 3 lists geometric parameters of the hydrogen bonds in 1α and 1β, which are as expected [35]. In 1α, this hydrogen bond arrangement results in polymeric tapes extending through glide symmetry in the [001] direction (Fig. 2). In contrast, the secondary amino groups in 1β are directed towards sulfur atoms of adjacent hydrogen-bonded dimers in the crystal structure with N⋯S distances of ca. 3.4 Å and significantly smaller H⋯S–C angles than in 1α. The symmetric dimer in 1β exhibits local inversion symmetry in the crystal structure. Frustration between competing intermolecular interactions, such as hydrogen bonds, and close packing has been put forward as a possible explanation for the Z’ > 1 phenomenon [38]. The crystal structures of both 1α and 1β belong to the same space group type (No. 14), which is available for densest packing of molecules of arbitrary shape. The packing index is 69.6% for 1α and 70.4% for 1β, revealing a dense crystal packing in both forms [39]. It is worth noting that in spite of the different hydrogen bonding patterns, the calculated crystallographic density is almost the same for 1α and 1β (Table 1).

Scheme 2
scheme 2

Possible N–H⋯S hydrogen bond associations of 1 (the black circle represents 4-methylpyridin-2-yl) with \({\text{R}}_{2}^{2}(8)\) motifs. 1α exhibits type I and 1β shows type II. Type III is not seen here

Full size image
Table 3 Hydrogen bond parameters (Å, °) for 1α and 1β
Full size table
Fig. 2
figure 2

Part of the crystal structure of 1α, showing hydrogen-bonded tapes of the molecules extending parallel to the crystallographic c axis

Full size image

Crystal and Molecular Structures of 2 and 3

Compound 2 crystallizes with two molecules in the asymmetric unit (Z′ = 2), as shown in Fig. 3. Selected bond lengths, bond angles and torsion angles are given in Table 4. In both molecules the thiazole sulfur atom S1 and the pivot carbon atom C10 of the pyridine ring as well as the pivot carbon atom of the thiazole ring C2 and the picoline nitrogen atom N1 are in a synperiplanar arrangement. The intramolecular N1⋯S1 distance is ca. 2.8 Å in both molecules and the C5–S1⋯N1 angles are 161.5 and 160.1° in molecule 1 and 2, respectively. From the structural point of view, this can be interpreted as chalcogen bonds between the picoline nitrogen lone pairs and the σ hole at the sulfur atoms opposite to the C5–S1 σ bonds [40, 41]. Interestingly, all 15 crystal structures of 2-aminothiazoles with N-bonded heteroaromatic substituents containing a nitrogen atom in the 2-position in the Cambridge Structural Database (June 2020) [42] exhibit planar conformations with intramolecular N⋯S distances of 2.70(4) Å (mean) in spite of different crystal environments, including structures of dasatinib and nine of its solvates [43, 44], as well as thiazovivin, a small molecule tool for stem cell research [45]. In contrast, 41 crystal structures of 2-aminothiazoles with variously substituted N-phenyl groups contain molecules in which the two moieties are randomly orientated to one another. The absence of any classical or weak hydrogen bonds towards the picoline nitrogen N1 corroborates this view. The dihedral angles between the mean planes through the thiazole rings and those through the pyridine rings attached to C4 is 22.96(8)° for molecule 1 and 37.53(6)° for molecule 2. Both molecules are thus significantly non-planar in this region. Clearly, the tilt between the pyridine ring and the thiazole ring should be disadvantageous for π electron delocalisation, but appears to be outweighed by the formation of strong intermolecular N–H⋯N hydrogen bonds between the amino group and the pyridine nitrogen atom N4 in the solid-state (Table 5). Molecule 1 and molecule 2 each form distinct hydrogen-bonded zigzag tapes extending in the [001] direction through glide symmetry (Fig. 4), resulting in a polar c axis. The packing index is 69.0%.

Fig. 3
figure 3

Asymmetric unit of 2. Displacement ellipsoids are drawn at the 50% probability level. Hydrogen atoms are shown as small spheres of arbitrary radii

Full size image
Table 4 Selected bond lengths, bond angles and torsion angles (Å, °) for 2 and 3
Full size table
Table 5 Hydrogen bond parameters (Å, °) for 2 and 3
Full size table
Fig. 4
figure 4

Part of the crystal structure of 2, viewed approximately along the a axis direction and b along the c axis direction. Hydrogen atoms have been omitted for clarity. Dashed lines represent hydrogen bonds. Red and blue colour indicate distinct hydrogen-bonded tapes resulting from unique molecules 1 and 2, respectively (cf. Fig. 3)

Full size image

Figure 5 shows the molecular structure of 3, bearing a pyrazinyl group at C4 of the thiazole ring instead of a 2-pyridinyl group as in 2. Selected bond lengths, bond angles and torsion angles are listed in Table 4. Like 2, the thiazole sulfur atom S1 and the pivot carbon atom C10 of the pyridine ring as well as the pivot carbon atom of the thiazole ring C2 and the picoline nitrogen atom N1 in 3 are in a synperiplanar orientation with an intramolecular S1⋯N1 distance of ca. 2.7 Å with a C5–S1⋯N1 angle of 162.5°. Thus, the observed conformation appears to be a characteristic trait of the N-(4-methylpyridin-2-yl)thiazol-2-amine moiety and is possibly supported by an intramolecular chalcogen bond (vide supra). In contrast to 2, the dihedral angle between the mean plane through the thiazole ring and that through the pyrazin ring attached to C4 is only 7.25(6)°. The supramolecular structure of 3 in the crystal is distinctly different from that in 2. Instead of the polymeric N–H⋯N hydrogen-bonded assemblies observed in 2, the crystal structure of 3 is characterized by centrosymmetric dimers (Fig. 6). The amino group of each molecule forms a hydrogen bond to the pyrazine nitrogen atom in meta position to the thiazole ring, affording a \({\text{R}}_{2}^{2}(16)\) motif. The dimers so formed are laterally connected via C–H⋯N interactions involving the opposite pyrazine nitrogen atom. Although the H⋯A distance is longer than the sum of the van der Waals radii, the < (DHA) angle (Table 5) supports the presence of weak cooperative C–H⋯N hydrogen bonds [46]. The N–H⋯N hydrogen bonds and long C–H⋯N contacts generate layers of molecules parallel to (10–1), which are π⋯π stacked in the third dimension. The packing index for 3 of 71.4% reveals a denser crystal packing than in 2. Since 2 and 3 differ only by one heteroatom site, it is possible that the weak cooperative C–H⋯N hydrogen bonds in 3 may have a structure-directing influence.

Fig. 5
figure 5

Molecular structure of 3 in the crystal. Displacement ellipsoids are drawn at the 50% probability level. Hydrogen atoms are represented by small spheres of arbitrary radii

Full size image
Fig. 6
figure 6

Part of the crystal structure of 3, viewed towards plane (10–1), showing two adjacent N–H⋯N hydrogen-bonded dimers laterally connected via long C–H⋯N contacts. Hydrogen bonds are shown by dashed lines

Full size image

Conclusions

We synthesized two pharmacologically relevant N,4-diaryl substituted 2-aminothiazoles via Hantzsch synthesis and elucidated their structures. Structural characterization of the starting thiourea derivative 1 revealed two crystalline forms, which exhibit distinctly different intermolecular N–H⋯S hydrogen bonding patterns. For 2-aminothiazoles 2 and 3, the encountered synperiplanar conformation of the N-(4-methylpyridin-2-yl)thiazol-2-amine moiety suggests the existence of supportive intramolecular N⋯S chalcogen bonds. Replacement of a pyridyl group in 2 by a pyrazinyl group in 3 has a significant effect on the supramolecular structures of 2 and 3 in the solid-state. Whereas 2 forms polymeric N–H⋯N hydrogen-bonded tapes, 3 forms N–H⋯N hydrogen-bonded cyclic dimers, which may further associate by additional weak intermolecular C–H⋯N hydrogen bonds. Since the number of structurally characterized N,4-diaryl substituted 2-aminothiazoles is hitherto limited, the structural insight gained from this study provides impetus for further exploration of this compound class in medicinal chemistry.

Supplementary Material

CCDC 2008777–2008780 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures.