Serendipity at work: unexpected ring transformations of 4-aminopyrazolidin-3-ones into N -aminohydantoins

Catalytic hydrogenation of (4 R *,5 R *)-4-benzyloxycarbonylamino-5-phenylpyrazolidin-3-one 2 in the presence of Pd-C furnished an unexpected ‘ring switching’ transformation product, 3-amino-5-benzylimidazolidine-2,4-dione 4 . Furthermore, heating of azomethine imines 3a , b (derived from 2 and aromatic aldehydes) afforded the corresponding ( Z )-5-benzylidene-3-[( E )- benzylideneamino]imidazolidine-2,4-diones 8a , b as ring transformation products. Both reactions are explainable by cleavage of the C(5)–N(1) single bond in substrates 2 and 3 followed by cyclocondensation of the amide nitrogen to the carbamate carbonyl group. The structure of hydantoin 8a was confirmed by X-ray diffraction.


Scheme 1
The structures of the pyrazolidinone 2 and azomethine imines 3a and 3b were determined by NMR.First, the (4R*,5R*)-configuration of 2 was established on the basis of vicinal coupling constant, 3 J H4-H5 .The coupling constant, 3 J H4-H5 = 11.1 Hz, fitted within the typical order of magnitude, 3 J H4-H5 ~ 10-12 Hz, measured previously in closely related 3-pyrazolidinones 5,9 with the trans-configuration around the C(4)-C(5) bond.The structures of dipoles 3a,b were determined by NOESY spectroscopy.NOE between the 1'-H and 5-H and NOE between 5-H and NH in compounds 3a and 3b were in agreement with the (Z)-configuration around the exocyclic C=N double bond and trans-configuration around the C(4)-C(5) bond.Other characteristic NMR data of compounds 3a,b were also in agreement with the data for closely related dipoles (Figure 1).

Figure 1
In order to prepare the unprotected (4R*,5R*)-4-amino-5-phenyl-3-pyrazolidinone 5, compound 2 was catalytically hydrogenated under 3 bar of H 2 in ethanol in the presence of 10% Pd-C.The conversion was surprisingly slow and required 30 hours to completion.To our surprise, the isolated product was 3-amino-5-benzylimidazolidine-2,4-dione 4 and not the expected compound 5.A possible explanation for the formation of hydantoin 4 could be the following.Catalytic hydrogenation of benzyl carbamate 2 produces the amine 5 and CO 2 as the primary products.Next, hydrogenolytic cleavage of the benzylic C(5)-N(1) bond in 5 gives the open-chain intermediate 6.Under slightly elevated pressure (3 bar), the amine 6 and CO 2 are in equilibrium with the carbamic acid 7, which cyclizes into N-amino hydantoin 4. The proposed mechanism is supported by known, closely related examples of cyclisations of Nbenzyloxycarbonyl-α-amino acid hydrazides 11 and α-semicarbazidoacetates 12 into 3aminoimidazolidine-2,4-diones.Besides, the above transformation is also related to Bucherer's synthesis of hydantoins, which proceeds in a closed vessel under slightly elevated pressure utilizing CO 2 (or carbonate) as a C 1 -synthon (Scheme 2). 13 Finally, heating of azomethine imines 3a and 3b in anisole produced 3-benzylideneimino-5benzylideneimidazolidine-2,4-diones 8a and 8b in 83% and 82% yield, respectively.Also here, the observed ring transformation is feasible only by cleavage of the C(5)-N(1) bond.Thus, a plausible rationale for this reaction includes rearrangement of heterocyclic enol 3' to give the open-chain α,β-unsaturated hydrazide 9, followed by intramolecular cyclocondensation to the benzyl carbamate residue to furnish the title compound 8 (Scheme 3).It is noteworthy, that the synthesis of compound 8a and some of its close analogues by condensation of 5-unsubstituted 3-(benzylideneamino)hydantoins with aromatic aldehydes has already been reported previously. 14

Scheme 3
The structures of novel compounds 2, 3a, 3b, and 8b and known compounds 4 and 8a were determined by spectroscopic methods (IR, 1 H and 13 C NMR, NOESY spectroscopy, and MS) and by elemental analyses for C, H, and N. Compounds 2, 3a, and 3b were not obtained in analytically pure form.Their identities were confirmed by 13 C NMR and/or EI-HRMS.
Physical and spectral data for known compound 4 were in agreement with the literature data.11a,12c On the other hand, physical and spectral data for 8a were not consistent with the literature data (see also Experimental). 14However, spectral data of 8a were in agreement with the literature data for closely related 5-alkylidenehydantoins. 15 Finally, the structure of compound 8a was unambiguously determined by X-ray diffraction (Figure 2).

Experimental Section
General.Melting points were measured on a Stanford Research Systems MPA100 OptiMelt automated melting point system.IR spectra were recorded on a Perkin-Elmer Spectrum BX FTIR spectrometer. 1 H and 13 C NMR spectra were recorded at 300 and 75.5 MHz, respectively, on a Bruker Avance DPX 300 instrument with DMSO-d 6 and CDCl 3 as solvents and TMS as internal standard.Mass spectra were recorded on an AutoSpecQ spectrometer.Elemental analyses for C, H, and N were obtained using Perkin-Elmer CHN Analyzer 2400 II.Catalytic hydrogenation was performed on a Parr Hydrogenation Apparatus 500 ml 3916EF.Column chromatography (CC) was performed on silica gel (Fluka, Silica gel 60, particle size: 0.035-0.070mm).Anisole, benzaldehyde, 2,6-dichlorobenzaldehyde, hydrazine hydrate, and 10% palladium on charcoal were purchased from Sigma-Aldrich and used without further purification.Methyl 2-(benzyloxycarbonylamino)-3-phenylacrylate 1 was prepared following the literature procedure. 8

General procedure for the preparation of azomethine imines (3a,b)
A mixture of 2 (1.557 g, 5 mmol), aromatic aldehyde (6 mmol), and ethanol (20 mL) was stirred at r.t. for 5 min.Then, trifluoroacetic acid (10 drops) was added and stirring at r.t. was continued for 3 h.The precipitate was collected by filtration and washed with ethanol (5 mL) and ether (2×5 mL).

X-Ray structure analysis for compound (8a)
Single crystal X-ray diffraction data of compound 8a were collected at room temperature on a Nonius Kappa CCD diffractometer using the Nonius Collect Software. 16DENZO and SCALEPACK 17 were used for indexing and scaling of the data and the structure was solved by means of SIR97. 18Refinement and plotting were done using Xtal3.6 19 program package.Crystal structure was refined on F values using the full-matrix least-squares procedure.The nonhydrogen atoms were refined anisotropically, while the positions of hydrogen atoms were geometrically calculated and their positional and isotropic atomic displacement parameters were not refined.Crystallographic data (excluding structure factors) for compound 8a have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication number CCDC 787638.Copy of the data can be obtained, free of charge, on application to CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK [fax: +44(0)-1223-336033 or e-mail: deposit@ccdc.cam.ac.uk].

Figure 2 .
Figure 2. ORTEP plot of the dimeric unit of 8a in the crystal structure.Ellipsoids are plotted at 50% probability, hydrogen bonds are depicted as dashed lines, and atom labeling of one asymmetric unit is shown.