1704
G. Pellegrino et al. / Tetrahedron Letters 51 (2010) 1702–1705
experimental conditions on the solution phase. Among the differ-
ent set of experimental conditions explored, the best results were
obtained using a 10% mixture of HMPA in DMF as the solvent and
DIPEA as the base under microwave irradiation. Unfortunately, car-
rying out the same reaction on the solid phase the overall yields
decreased dramatically (down to 40%). The problem was solved
by increasing the base strength of the used amine. In fact, by using
DBU instead of DIPEA the yields calculated after cleavage from the
resin and purification by preparative TLC were in the range 67–
95%. The next step, that is the formation of the enaminoketones,
required a full optimization on the solid phase, since there are sev-
eral known procedures for this reaction on the solution phase, to
the best of our knowledge, no valuable synthetic protocol has been
so far published for the same reaction on the solid phase. Gener-
ally, the preparation of a phenoxyenaminoketone in the solution
phase is accomplished by heating the starting ketone in neat
DMFDMA. Unfortunately, the same procedure cannot be trans-
ferred on the solid phase given the bad swelling properties of this
reagent. To gain the best compromise between the amount of
DMFDMA needed and the swelling of the resin, DMF was used as
solvent, and a 4–1 DMF/DMFDMA ratio was found to afford an
acceptable swelling of the resin. The reaction was performed under
microwave irradiation at 120 °C for 1 h affording, after cleavage
and purification by preparative TLC, a pure compound in almost
quantitative yield. The last chemical step, that is the cyclization
reaction, was obviously different since diverse reagents were used
for this chemical transformation. As illustrated in Scheme 1 the
cyclization reaction was carried out using three suitably substi-
tuted guanidines and hydrazines, and hydroxylamine, as reagents.
The optimization of this crucial step was very difficult, since it was
impossible to find out a unique procedure to accomplish all the
cyclization reactions in high yields. In fact, the best experimental
conditions settled out for the synthesis of pyrimidine derivatives
did not give satisfactorily results in the preparation of isoxazole
and pyrazole derivatives. For this reason, two different methods
were developed to carry out the cyclization reactions, one for the
synthesis of pyrimidines and another for the preparation of isoxaz-
oles and pyrazoles. For the former, the best results (69–73%, yields)
were obtained using 10 equiv of BEMP in a 10% mixture of HMPA in
DMF under microwave irradiation at 150 °C for 30 min (Table 1).
As far as the synthesis of isoxazole and pyrazole derivatives is
concerned, the best experimental conditions settled for accom-
plishing this critical transformation were quite different compared
to the previous ones, since the final cyclization was performed in a
different solvent mixture (DMF/i-PrOH, 4:1) and without base be-
cause the desired products were only obtained using the hydro-
chloric salts of the corresponding hydroxylamine or hydrazines.
In fact, the use of DMF/HMPA as well as the use of DIPEA, or also
DBU, gave only modest yields of the desired products, whereas the
use of a 4:1 mixture of DMF/i-PrOH under microwave exposure
afforded in 30 min isoxazoles 19–22 and pyrazoles 23–40 deriva-
tives at 90 °C and 100 °C, respectively (path 2, Scheme 1 and Table
2). The yields assessed for this synthetic step were 50–70% for isox-
azole and 40–85% for pyrazole derivatives.
irradiation11,12 in each step significantly reduced the reaction
times and improved the yields, the purity and the final work-up
of the desired products. Our strategy proved to be very versatile
to introduce a variety of molecular fragments and functional
groups for an expected productive binding at the targeted enzy-
matic binding sites.
Acknowledgment
The authors gratefully acknowledge the financial support from
European Commission (‘CancerGrid’ STREP project, FP VI, Contract
LSHC-CT-2006-03755).
References and notes
1. Zhang, S. Structure/ligand-based Drug Design and Structure Bioinformatics: Basics,
Concepts, Methods, and Applications; VDM: Saarbrucken, 2009.
2. Zartler, E. R.; Shapiro, M. J. Fragment-Based Drug Discovery: A Practical Approach;
John-Wiley: New-York, 2008.
3. Gubernator, K.; Bohm, H. J. Structure-based Ligand Design; Wiley-VCH:
Weinheim, 1998.
4. Schreiber, S. L. Science 2000, 287, 1964–1969.
5. Gillet, V. J.; Nicolotti, O. Perspect Drug Discovery Des. 2000, 20, 265–287.
6. (a) Lapenna, S.; Giordano, A. Nat. Rev. Drug Disc. 2009, 8, 547–566; (b)
Hartmann, J. T.; Haap, M.; Kopp, H. G.; Lipp, H. P. Curr. Drug Metab. 2009, 10,
7. (a) Anzini, M.; Rovini, M.; Cappelli, A.; Vomero, S.; Manetti, F.; Botta, M.;
Sautebin, L.; Rossi, A.; Pergola, C.; Ghelardini, C.; Norcini, M.; Giordani, A.;
Makovec, F.; Anzellotti, P.; Patrignani, P.; Biava, M. J. Med. Chem. 2008, 51,
4476–4481; (b) Fitzgerald, G. A.; Patrono, C. N. Engl. J. Med. 2001, 345, 433–442.
8. (a) Radi, M.; Falciani, C.; Contemori, L.; Petricci, E.; Maga, G.; Samuele, A.;
Zanoli, S.; Terrazas, M.; Castria, M.; Togninelli, A.; Esté, J. A.; Clotet-Codina, I.;
Armand-Ugón, M.; Botta, M. Chem. Med. Chem. 2008, 3, 573–593; (b) Mowbray,
C. E.; Burt, C.; Corbau, R.; Perros, M.; Tran, I.; Stupple, P. A.; Webster, R.; Wood,
A. Bioorg. Med. Chem. Lett. 2009, 19, 5599–5602.
9. (a) Thompson, L. A.; Ellman, J. A. Chem. Rev. 1996, 96, 555–600; (b)Solid-Phase
Synthesis: A Practical Guide; Kate, S. A., Albericio, F., Eds.; Marcel Dekker, 2000;
(c) Nicolaou, K. C.; Hanko, R.; Hartwig, W. Handbook of Combinatorial Chemistry;
Wiley-VCH: Weinheim, 2002.
10. Typical procedure (steps a–c): 2.0 g (1.36 mmol) of Fmoc-protected Rink amide
resin (LL = 0.68 mmol/g), were swelled in 20 mL of DMF. After Fmoc removal by
a standard protocol (20 mL of a 20% solution of piperidine in DMF, 30 min, room
temperature), and washing with DMF (3 Â 20 mL), the resin was suspended
in DMF (20 mL) and reacted at room temperature overnight with 3-
hydroxybenzoic acid (1.1 g, 8.0 mmol), using EDCÁHCl (1.5 g, 8.0 mmol) as
coupling reagent. The resin was washed with DMF (3 Â 20 mL), THF (3 Â 30 mL)
and MeOH (3 Â 20 mL), and then dried under reduced pressure. The yields
calculated after cleavage of the product from a small aliquot of the resin and
subsequent purification by preparative TLC were close to 90%. The support-
bound 3-hydroxy benzoic acid I was placed in a microwave reactor vessel and
swelled in 4 mL of DMF. After filtration, the slurry was suspended in a 10%
solution of HMPA in dry DMF (4 mL) and treated with 0.36 mmol of one of the
six selected 2-bromoacetophenones (Tables 1 and 2) for 30 min at 140 °C under
microwave exposure. The desired intermediate II was obtained in 67–95%
yields as checked after cleavage and purification. The resin was filtered, washed
as above and dried under vacuum. After swelling, support-bound intermediate
II was suspended in a 25% solution of DMFDMA in DMF (5 mL) and heated in a
sealed reactor vessel under microwave exposure for 1 h at 120 °C. The expected
intermediate III was obtained in quantitative yield as checked after cleavage.
The resin was filtered, washed and dried as described above.
Pathway 1: 100 mg (0.064 mmol) of support-bound III, were placed in
a
microwave reactor vessel, swelled and suspended in a 10% solution of HMPA in
dry DMF (8 mL). To the slurry were added BEMP (0.64 mmol, 10 eq.), one of the
three selected guanidines (1.2 mmol, 20 equiv). The reactor was sealed and the
suspension was heated under microwave irradiation at 150 °C for 30 min. After
filtration, washing with DMF (3 Â 4 mL), THF (3 Â 4 mL) and CH2Cl2 (3 Â 4 mL),
the resin was treated for 20 min with a 50% solution of TFA in CH2Cl2 (2 mL). The
cleaved solution was filtered and the resin washed with the same solvent
mixture (3 Â 2 mL). The solvent mixtures were combined and immediately
concentrated with rotary evaporation. Toluene (2 mL) was added twice during
the concentration step in order to remove completely the TFA (42–63% overall
yield).
Pathway 2: 100 mg (0.064 mmol) of support-bound enaminoketone III were
placed in a microwave reactor vessel, swelled and suspended in 5 mL of a 4:1
mixture of DMF/i-PrOH. To the slurry was added H2NOHÁHCl (1.3 mmol, 88 mg)
or one of the three selected hydrazines as hydrochloric salts (0.2 mmol). The
reactor was sealed and the suspension was heated under microwave irradiation
at 90 °C and 100 °C for 30 min to afford the corresponding isoxazole (19–22)
and pyrazole (23–40) derivatives in 30–60% and 25–73% of overall yields,
respectively. The work-up of the reactions followed the same procedure
described above in Path 1.
Unexpectedly, the cyclization reactions with bromo- and meth-
oxy-phenylhydrazines gave very poor yields when the reactions
were performed in large excess of reagents (20 equiv, as done with
hydroxylamine, hydrazine and guanidines). Better results were
obtained with only 3 equiv of substituted hydrazines. Even more
surprisingly, the cyclization with hydroxylamine of enaminoketon-
ic intermediates III bearing biphenyl and the p-pyrrolidin-1-yl-
phenyl R1 substituents failed to give the expected isoxazole
derivatives.
In conclusion, the proposed solid phase synthetic protocol al-
lowed a straightforward preparation of an array of molecules span-
ning a broad range of molecular diversity. The use of microwave
Pellegrino, Giovanni
