A different role of Meldrum's acid in the Biginelli reaction

Article abstract of DOI:10.1002/hlca.201000193

A Biginelli-type condensation using Meldrum's acid has been accomplished in refluxing AcOH to give 6-substituted dihydropyrimidine-2,4-(1H,3H)-diones. In contrast to other aldehydes, the three-component reaction with salicylaldehyde led to an oxygen-bridged pyridine. A reaction mechanism is proposed.

Full text of DOI:10.1002/hlca.201000193

Helvetica Chimica Acta – Vol. 94 (2011)
199
A Different Role of Meldrumꢀs Acid in the Biginelli Reaction
by Jan Sve˘tlꢁk* and Lucia Veizerova´
Department of Pharmaceutical Analysis and Nuclear Pharmacy, Comenius University, Odbojarov 10,
SK-83232 Bratislava (e-mail: svetlik@fpharm.uniba.sk)
A Biginelli-type condensation using Meldrumꢀs acid has been accomplished in refluxing AcOH to
give 6-substituted dihydropyrimidine-2,4-(1H,3H)-diones. In contrast to other aldehydes, the three-
component reaction with salicylaldehyde led to an oxygen-bridged pyridine. A reaction mechanism is
proposed.
Introduction. – 2,2-Dimethyl-1,3-dioxane-4,6-dione (1), also known under the
trivial name of Meldrumꢀs acid and being formally a cyclic acylal, can be also classified
as a cyclic malonate (isopropylidene malonate). The high CꢀH acidity (pKa 4.83), flat
ring structure, rigidness, and unique chemical properties render this compound a
versatile reagent for preparative organic chemistry (see [1] for reviews). In terms of
synthesis [2], 1 is referred to as an efficient precursor of a d2-synthon. In the last decade,
widespread attention has been focused on the construction of highly substituted spiro-
heterocycles possessing a Meldrumꢀs acid unit. Due to their remarkable reactivity, these
derivatives have proven to be attractive intermediates in the synthesis of complex
organic molecules, various classes of natural products and their analogs, pharmaco-
logically active agents, and unusual amino acids [1c – 1e][3].
Recently, Lewis acid (NiCl2, CoCl2)-mediated preparation of spiro-heterocycle 2
through Biginelli-like condensation of Meldrumꢀs acid, aldehyde, and urea has been
described [4] (Fig. 1). Similar use of a Brønsted acid additive (AcOH; 1 equiv.) led to
an identical product 2 under otherwise analogous conditions (microwave-assisted
synthesis without solvent) [5a]. However, the same group demonstrated later that the
spiro-heterobicyclic ring system of 2 is produced even without catalyst and solvent at
808 [5b]. These findings differ from our method developed for the synthesis of
pyridinecarboxylates 3 via Hantzsch-like reaction from Meldrumꢀs acid, methyl
acetoacetate, aldehyde, and AcONH4 in boiling EtOH [6]. Evidently, 1 acts in this
case both as a second active CH2 component and a C2-reagent. Later, Verdecia et al., in
order to improve the yield of 3, modified our procedure by using reflux in AcOH [7].
Fig. 1. Structures of compounds 2 and 3
ꢁ 2011 Verlag Helvetica Chimica Acta AG, Zꢂrich

200
Helvetica Chimica Acta – Vol. 94 (2011)
In the light of these facts, it was of interest to revisit the aforementioned unexpected
Biginelli reaction using conventional heating technique in solution. The present study
was also prompted by the recent pharmacological revival of Biginelli dihydropyrimi-
dines which exhibited activity as mitotic kinesin Eg5 inhibitors [8], melanin-
concentrating hormon receptor (MCH1-R) antagonists [9], chemical modulators of
heat shock protein 70 (Hsp 70) [10], hepatitis B-replication inhibitors [11], and
inhibitors of the fatty acid transporter [12]. In particular, a simple 4-aryldihydropyr-
imidine derivative was recently discovered to be a highly effective agent against a
methicillin-resistant strain of Staphylococcus aureus (MRSA) [13]. In pursuing our
work on the Hantzsch- and Biginelli-like heterocyclizations [6][14], we present here a
novel practical application of Meldrumꢀs acid for the synthesis of 6-aryldihydropyr-
imidine-2,4-diones, the so-called 5,6-dihydrouracils.
Results and Discussion. – Since acid catalysis is required in the Biginelli
condensation [15], we chose AcOH as both an acidic promoter and the reaction
medium. Initially, a test experiment with an equivalent of benzaldehyde, Meldrumꢀs
acid, and urea was conducted in AcOH at reflux for 14 h. Although the melting point of
isolated compound 4a (Scheme 1) was close to that reported [5] for spiro heterocycle 2
(Ar¼ Ph), a combustion analysis clearly ruled out this derivative. We, therefore,
considered 1 to be involved as a d2-synthon in the Biginelli reaction forming product 4a.
In view of the above-mentioned pyridine 3, a substituted pyrimidine-2,4-dione 4a, being
partially related to Biginelli derivatives, appears to be a reasonable alternative in this
1
case. Indeed, H- and 13C-NMR spectra are in good agreement with the proposed
structure 4a. Moreover, the structure assignment was confirmed by comparing our
spectroscopic data with published values [16a].
Scheme 1
To explore the scope of the three-component reaction depicted in Scheme 1, we
have evaluated a variety of aldehyde substrates. It was found that benzaldehydes
bearing electron-withdrawing substituents reacted well to give moderate yields of
heterocyclic condensates, whereas those containing electron-donating groups together
with a heterocycle led to only lower yields. Unfortunately, the one-pot condensation

Helvetica Chimica Acta – Vol. 94 (2011)
201
works poorly with aliphatic aldehydes under the same conditions. Furthermore, a
successful application of this method for synthesis of 2-thioxo congeners employing
thiourea as reactant is exemplified by derivative 4h.
In principle, two Knoevenagel conjugates, A and B, are conceivable as precursors of
Biginelli structure C (Fig. 2). Although N-acylimine species was evidenced as a key
intermediate in the Biginelli reaction [15], a pathway involving enone adduct A seems
more probable because of the high reactivity of Meldrumꢀs acid. Also 5-benzylidene
derivatives of type A were reported [5b] to be isolated in spirocyclizations. The bicyclic
derivative C is further converted into desired pyrimidine-2,4-dione 4 by decarbox-
ylative fragmentation of the 1,3-dioxane skeleton under release of CO2 and acetone.
Fig. 2. Two Knoevenagel conjugates, A and B, are conceivable as precursors of Biginelli structure C
Alternatively, from a mechanistic study [17] concerning the transformation of 5-
acyl Meldrumꢀs acids through an a-oxo ketene RꢀCOꢀCH¼C¼O, one may deduce that
a similar highly reactive heterocumulene RꢀCH¼C¼C¼O might be involved as an
intermediate. Nevertheless, this alternative is speculative and remains to be evaluated.
In the light of the above-mentioned results, we continued to examine a cyclo-
condensation with salicylaldehyde, a topic of our continuing interest. Surprisingly, and
in spite of identical reaction conditions, an anomalous course of the three-component
reaction was observed in contrast to the previous heterocyclizations. A product isolated
upon treatment in AcOH and usual workup exhibited a molecular-ion (Mþ) peak at m/
z 203 in its mass spectrum, indicating an odd number of N-atoms [18]. It follows from
this observation that only one N-atom originating from urea was built in the product.
The 1H-NMR spectrum indicated the presence of one Me group, two CH2 groups, one
CH group, an ortho-phenylene ring, and one exchangeable H-atom (most probably
NH). In addition, the 13C-NMR spectrum displayed a resonance at d(C) 82.6 for an sp3-
C-atom and a low-field signal (d(C) 170.2) attributable to a CO group. The chemical-
shift value of the former matches the interval near 80 ppm which is diagnostic for a
C(O)N grouping, i.e., an N,O-acetal moiety [6]. In addition, 2D homo- and
heteronuclear correlations (COSYand HMBC) corroborated a CH2ꢀCHꢀCH2 linkage.
This fact, together with a full NMR analysis, allowed us to ascribe formula 5 to the
isolated compound (Scheme 2).
The structure is fully consistent with that of the heterocyclic product from the
cyclocondensation of 4-(2-hydroxyphenyl)but-3-en-2-one, Meldrumꢀs acid, and
AcONH4 in refluxing EtOH [6]. Structure 5 was confirmed by mixed melting point
with an authentic sample prepared as mentioned above [6]. The O-bridged pyridine 5
can be considered as the result of a Hantzsch-like reaction. Although the sole
formation of the tricyclic compound 5 is rather surprising, Hantzsch-type by-products
under Biginelli conditions has been described in [19]. It is assumed that urea

202
Helvetica Chimica Acta – Vol. 94 (2011)
Scheme 2
decomposes to CO2 and NH3, followed by the heterocyclization step. Structure 5
indicates that both Meldrumꢀs acid and acetone, together with salicylaldehyde and
NH3, take part in the formation of the pyridine ring. The involvement of acetone,
generated from Meldrumꢀs acid, in a cascade mechanism clarifies certain domino
processes with 1, which have recently been proposed [3]. The transformation described
here is, to our knowledge, one of a few examples where an intermediacy of acetone
formed from 1 plays a key role [3][20]. A different result was obtained when treating
salicylaldehyde with Meldrumꢀs acid and urea in boiling EtOH under HCl catalysis,
which is the established Biginelli protocol. Instead of pyridine 5, the product we
isolated was coumarin-3-carboxylic acid (6). Afterwards, we detected traces of 6 in the
mother liquor from the condensation leading to 5. This would imply a different reaction
pathway and consequently another intermediate to be considered when comparing the
discussed foregoing pyrimidine route. Apparently, the OH group of the phenol part
must account for the observed specific behavior. To confirm our assumption that
coumarin derivative 6 might be a key intermediate in the condensation studied, acid 6
was treated with acetone and urea in hot AcOH, and indeed, the O-bridged pyridine 5
was obtained in 80% yield.
Based upon these findings, a reasonable mechanism can be formulated as shown in
Scheme 3. Thus, coumarin-3-carboxylic acid (6) formed in conventional manner [21]
through enone B undergoes Michael addition with acetone to provide species D. The
loss of CO2 may occur in parallel or consecutively. The lactone ring is cleaved by NH3 to
give intermediate amide E, which subsequently again cyclizes to give pyridone F.
Finally, a nucleophilic addition of the phenolic OH group at the activated imine C-atom
of the 2-aza-enone moiety forms the O-bridge in product 5 to terminate the reaction
sequence. A similar mechanism has been reported for the ring transformation of 3-
acetylcoumarin [22].
These results lead to a question regarding the factors that determine which of two
pathways, pyrimidine or pyridine route, will be followed. One speculative explanation
concerning salicylaldehyde is the assumption that the heterocyclization course of the
carboxylic acid 6 proceeds also with urea, while CONH2 linked to the reacting N-atom
is lost during the follow-up reaction to yield pyridine 5. Nevertheless, there is a lack of
data to support this hypothesis.
In conclusion, we have demonstrated a different behavior of Meldrumꢀs acid in
Biginelli-like condensation depending on reaction conditions. In addition to the
formation of spiro-heterocycles reported by other authors, it has been shown that the

Helvetica Chimica Acta – Vol. 94 (2011)
203
Scheme 3
three-component reaction of Meldrumꢀs acid with aldehydes and urea gives 6-
substituted dihydropyrimidine-2,4(1H,3H)-diones. In contrast, salicylaldehyde unex-
pectedly gave an O-bridged pyridine derivative. An unusual incorporation in the
tricyclic molecule of acetone released from the Meldrumꢀs acid fragment has been
established. Some of the prepared compounds are evaluated for anticancer and AIDS
antiviral activity at the National Cancer Institute in Bethesda.
This work was supported by VEGA 1/0320/11. The NMR part of this work was facilitated by a
support of the Slovak National Research and Development Program No. 2003SP200280203.
Experimental Part
General. M.p.: Kofler hot-stage microscope; not corrected. IR Spectra (KBr): Nicolet Impact 400D
spectrometer; nmax in cmꢀ1. 1H- and 13C-NMR spectra: Varian Mercury 300 instrument; in (D6)DMSO at
~
300 and 75 MHz, resp. d in ppm rel. to Me4Si as internal standard, J in Hz. MS: VG 7070E mass
spectrometer; at 70 eV; in m/z. Elemental analyses: Carlo-Erba Elemental Analyzer 1012.
Compounds 4: General Procedure. A soln. of Meldrumꢀs acid (1; 1.44 g, 10 mmol), the appropriate
aldehyde (10 mmol), and urea (0.6 g, 10 mmol) or thiourea (0.76 g, 10 mmol) in AcOH (20 ml) was
refluxed for 14 h. After removal of the solvent, the oily residue was dissolved in EtOH and left to stand at
r.t. The crystallized products 4 were collected by filtration.
5,6-Dihydro-6-phenylpyrimidine-2,4(1H,3H)-dione (4a): Yield: 0.78 g (41%). White solid. M.p.
219 – 2218 ([16a]: 216 – 2188). 1H-NMR: 2.61 (dd, J ¼ 16.2, 6.9, HaꢀC(5)); 2.84 (dd, J ¼ 16.2, 6.0,
HbꢀC(5)); 4.67 (dt, J ¼ 6.3, 2.4, HꢀC(6)); 7.27 – 7.40 (m, 5 arom. H); 8.00 (s, HꢀN(1)); 10.16 (s, HꢀN(3)).
13C-NMR: 38.3 (CH2); 50.1 (CH); 126.1 (2 arom. C, Co); 127.7 (1 arom. C, Cp); 128.7 (2 arom. C, Cm);
141.2 (1 arom. C, Cip); 153.9 (C(2)¼O); 169.9 (C(4)¼O).
5,6-Dihydro-6-(4-nitrophenyl)pyrimidine-2,4(1H,3H)-dione (4b): Yield: 1.46 g (62%). Brownish
powder. M.p. 257 – 2598 ([16b]: 254 – 2568).
5,6-Dihydro-6-(3-nitrophenyl)pyrimidine-2,4(1H,3H)-dione (4c): Yield: 1.39 g (59%). Beige pow-
der. M.p. 265 – 2678. IR: 3305, 3228 (NH), 1709, 1685 (CO), 1525, 1358 (NO2), 1458 (CH2), 803, 689

204
Helvetica Chimica Acta – Vol. 94 (2011)
1
(¼CH). H-NMR: 2.72 (dd, J ¼ 16.2, 7.2, HaꢀC(5)); 2.89 (dd, J ¼ 16.2, 5.7, HbꢀC(5)); 4.87 (ddd, J ¼ 7.2,
5.7, 2.4, HꢀC(6)); 7.66 – 7.72 (m, 1 arom. H); 7.79 – 7.82 (m, 1 arom. H); 8.14 (s, HꢀN(1)); 8.14 – 8.21 (m, 2
arom. H); 10.26 (s, HꢀN(3)). Anal. calc. for C10H9N3O4 (235.20): C 51.07, H 3.86, N 17.87; found: C 50.89,
H 4.03, N 18.01.
6-(4-Chlorophenyl)-5,6-dihydropyrimidine-2,4(1H,3H)-dione (4d): Yield: 0.67 g (30%). White solid.
M.p. 249 – 2518 ([16c]: 251 – 2538).
5,6-Dihydro-6-(4-methoxyphenyl)pyrimidine-2,4(1H,3H)-dione (4e): Yield: 0.46 g (21%). White
powder. M.p. 230 – 2318 ([16d]: 2288).
5,6-Dihydro-6-(thiophen-2-yl)pyrimidine-2,4(1H,3H)-dione (4f): Yield: 0.31 g (16%). White crys-
tals. M.p. 242 – 2448 ([16a]: 243 – 2458).
5,6-Dihydro-6-(2-methylpropyl)pyrimidine-2,4(1H,3H)-dione (4g): Yield: 0.15 g (9%). White solid.
M.p. 216 – 2188 ([16e]: 215 – 2178).
3,4,5,6-Tetrahydro-6-phenyl-2-thioxopyrimidin-4(1H)-one (4h): Yield: 0.68 g (33%). White solid.
M.p. 235 – 2378 ([16f]: 2388).
(ꢁ)-(2RS,6SR)-2,3,5,6-Tetrahydro-2-methyl-4H-2,6-methano-1,3-benzoxazocin-4-one (5). a) This
compound was prepared from salicylaldehyde (1.1 ml, 10 mmol), Meldrumꢀs acid (3.0 g, 21 mmol),
and urea (0.6 g, 10 mmol) following the above protocol. Yield: 1.37 g (67%). M.p. 258 – 2598 ([6]: 257 –
2588). 1H-NMR: 1.60 (s, Me); 2.02 (ddd, J ¼ 13.2, 1.8, 1.7, HeqꢀC(11)); 2.14 (dd, J ¼ 13.2, 1.8, HaxꢀC(11));
2.26 (br. d, J ¼ 17.4, HeqꢀC(5)); 2.63 (dd, J ¼ 17.4, 4.8, HaxꢀC(5)); 3.18 – 3.20 (m, HꢀC(6)); 6.74 (d, J ¼ 7.8,
arom. HꢀC(10)); 6.89 (t, J ¼ 7.8, arom. HꢀC(8)); 7.13 (t, J ¼ 7.8, arom. HꢀC(9)); 7.18 (d, J ¼ 7.8, arom.
HꢀC(7)); 8.37 (br. s, NH). 13C-NMR: 26.9 (Me); 28.9 (C(6)); 32.1 (C(11)); 40.5 (C(5)); 82.6 (C(O)N);
116.7 (arom. C(10)); 120.7 (arom. C(8)); 125.6 (arom. C(6a)); 128.1 (arom. C(9)); 129.3 (arom. C(7));
151.3 (arom. C(10a)); 170.2 (C(4)¼O). EI-MS: 203 (Mþ).
b) Alternative procedure: a mixture of coumarin-3-carboxylic acid (6; 0.6 g, 3.16 mmol), urea
(0.19 g, 3.16 mmol), and acetone (3 ml) in AcOH (25 ml) was refluxed for 24 h. The solvent was
evaporated, and the oily residue was triturated with EtOH. The crystalline product was filtered off. Yield:
0.52 g (80%). NMR Data are identical to those given above.
REFERENCES
[1] a) H. McNab, Chem. Soc. Rev. 1978, 7, 345; b) B.-C. Chen, Heterocycles 1991, 32, 529; c) J.
´
´
Gerencser, G. Dorman, F. Darvas, QSAR Comb. Sci. 2006, 25, 439; d) A. S. Ivanov, Chem. Soc. Rev.
2008, 37, 789; e) V. V. Lipson, N. Y. Gorobets, Mol. Diversity 2009, 13, 399.
[2] J. Fuhrhop, G. Li, ꢃOrganic Synthesis: Concepts and Methodsꢀ, 3rd ed., Wiley-VCH, Weinheim,
2003, pp. 1 – 6; R. W. Hoffmann, ꢃElements of Synthesis Planningꢀ, Springer Verlag, Berlin-
Heidelberg, 2009, pp. 9 – 15.
[3] D. B. Ramachary, N. S. Chowdari, C. F. Barbas III, Angew. Chem., Int. Ed. 2003, 42, 4233; B. Jiang,
W.-J. Hao, J.-P. Zhang, S.-J. Tu, F. Shi, Org. Biomol. Chem. 2009, 7, 2195.
[4] A. V. Narsaiah, K. Nagaiah, Indian J. Chem., Sect. B 2004, 43, 2482.
[5] a) A. Shaabani, A. Bazgir, Tetrahedron Lett. 2004, 45, 2575; b) A. Shaabani, A. Bazgir, H. R.
Bijanzadeh, Mol. Diversity 2004, 8, 141.
˘
ˇ
[6] J. Svetlꢄk, I. Goljer, F. Turecek, J. Chem. Soc., Perkin Trans. 1 1990, 1315.
´
´
[7] Y. Verdecia, M. Suarez, A. Morales, E. Rodrꢄguez, E. Ochoa, L. Gonzales, N. Martin, M. Quinteiro,
C. Seoane, J. L. Soto, J. Chem. Soc., Perkin Trans. 1 1996, 947.
[8] V. Sarli, S. Huemmer, N. Sunder-Plassmann, T. U. Mayer, A. Giannis, ChemBioChem 2005, 6, 2005;
E. Klein, S. DeBonis, B. Thiede, D. A. Skoufias, F. Kozielski, L. Lebeau, Bioorg. Med. Chem. 2007,
15, 6474.
[9] A. M. Basso, N. A. Bratcher, K. B. Gallagher, M. D. Cowart, C. Zhao, M. Sun, T. A. Esbenshade,
M. E. Brune, G. B. Fox, M. Schmidt, C. A. Collins, A. J. Souers, R. Iyengar, A. Vasudevan, P. R.
Kym, A. A. Hancock, L. E. Rueter, Eur. J. Pharmacol. 2006, 540, 115.
´
[10] S. Wisen, J. Androsavich, C. G. Evans, L. Chang, J. E. Gestwicki, Bioorg. Med. Chem. Lett. 2008, 18,
60.

Helvetica Chimica Acta – Vol. 94 (2011)
205
[11] K. Deres, C. H. Schrçder, A. Paessens, S. Goldmann, H. J. Hacker, O. Weber, T. Krꢅmer, U.
Niewçhner, U. Pleiss, J. Stoltefuss, E. Graef, D. Koletzki, R. N. A. Masantschek, A. Reimann, R.
Jaeger, R. Groß, B. Beckermann, K.-H. Schlemmer, D. Haebich, H. Rꢂbsamen-Wagmann, Science
2003, 299, 893.
[12] C. Blackburn, B. Guan, J. Brown, C. Cullis, S. M. Condon, T. J. Jenkins, S. Peluso, Y. Ye, R. E.
Gimeno, S. Punreddy, Y. Sun, H. Wu, B. Hubbard, V. Kaushik, P. Tummino, P. Sanchetti, D. Y. Sun, T.
Daniels, E. Tozzo, S. K. Balani, P. Raman, Bioorg. Med. Chem. Lett. 2006, 16, 3504.
[13] E. E. Wyatt, W. R. J. D. Galloway, G. L. Thomas, M. Welch, O. Loiseleur, A. T. Plowright, D. R.
Spring, Chem. Commun. 2008, 4962.
˘
˘
´
[14] J. Svetlꢄk, V. Kettmann, B. Zaleska, Tetrahedron Lett. 2005, 46, 5511; J. Svetlꢄk, L. Veizerova, V.
Kettmann, Tetrahedron Lett. 2008, 49, 3520.
[15] C. O. Kappe, Acc. Chem. Res. 2000, 33, 879.
[16] a) L. Strekowski, R. A. Wattson, M. A. Faunce, Synthesis 1987, 579; b) N. G. Chernova, A. Y. Berlin,
Zh. Obshch. Khim. 1962, 32, 596; Chem. Abstr. 1963, 58, 4564; c) ꢆ. ꢇalis¸, M. Kçksal, Arzneim.-
Forsch. 2001, 51, 523; d) T. B. Johnson, J. E. Livak, J. Am. Chem. Soc. 1936, 58, 299; e) D. Keglevic,
A. Kornhauser, Croat. Chem. Acta 1959, 31, 47; Chem. Abstr. 1960, 54, 13142c; f) L. S. Sandakhchiev,
V. P. Mamaev, Izv. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. 1961, 72; Chem. Abstr. 1962, 56, 2448g.
[17] F. Xu, J. D. Armstrong III, G. X. Zhou, B. Simmons, D. Hughes, Z. Ge, E. J. J. Grabowski, J. Am.
Chem. Soc. 2004, 126, 13002.
[18] F. W. McLafferty, F. Turecek, ꢃInterpretation of Mass Spectraꢀ, 4th ed., University Science Book,
Mill Valley, 1993, pp. 37 – 38.
[19] J. S. Yadav, B. V. S. Reddy, P. T. Reddy, Synth. Commun. 2001, 31, 425; R. Osnaya, G. A. Arroyo, L.
´
Parada, F. Delgado, J. Trujillo, M. Salmon, R. Miranda, Arkivoc 2003, 112.
[20] A. Shaabani, A. H. Rezayan, S. Keshipour, A. Sarvary, S. W. Ng, Org. Lett. 2009, 11, 3342.
[21] A. Song, X. Wang, K. S. Lam, Tetrahedron Lett. 2003, 44, 1755.
[22] C. N. OꢀCallaghan, T. B. H. McMurry, J. Chem. Res., Synop. 1989, 329.
Received May 12, 2010