Efficient synthesis of unsymmetrical 1,4-disubstituted-2,3-diketopiperazines via tandem reductive amination-cyclization

Article abstract of DOI:10.1016/S0040-4039(00)01565-3

Herein we report the efficient syntheses of 1,4-disubstituted-2,3-diketopiperazines and 1,4,5-trisubstituted-2,3-diketopiperazines, which feature a tandem reductive amination and acylation. Aliphatic and aromatic primary amines serve as viable nucleophiles under the mild reaction conditions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01565-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8735–8739
Efficient synthesis of unsymmetrical
1,4-disubstituted-2,3-diketopiperazines via tandem reductive
amination–cyclization
Douglas C. Beshore* and Christopher J. Dinsmore
Department of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
Received 28 July 2000; revised 31 August 2000; accepted 2 September 2000
Abstract
Herein we report the efficient syntheses of 1,4-disubstituted-2,3-diketopiperazines and 1,4,5-trisubsti-
tuted-2,3-diketopiperazines, which feature a tandem reductive amination and acylation. Aliphatic and
aromatic primary amines serve as viable nucleophiles under the mild reaction conditions. © 2000 Elsevier
Science Ltd. All rights reserved.
Piperazine derivatives are used throughout medicinal chemistry as molecular scaffolds for
biologically active compounds.1 In the course of an ongoing drug discovery project, we required
access to unsymmetrical 1,4-disubstituted-2,3-diketopiperazines. Previous syntheses of diketo-
piperazines have depended upon the prior construction of an ethylenediamine moiety, followed
by bis-acylation with oxalic acid or its ester derivatives.2 We have previously reported an
efficient route to 1,4-disubstituted 2,3-diketopiperazines via reductive amination of a suitably
substituted aldehyde with an N-(2-aminoethyl)oxamate ester scaffold followed by concomitant
cyclization, producing diversely substituted 2,3-diketopiperazines.3 Herein, we report a compli-
mentary method, which introduces N-substitution through a primary amine. Under reductive
amination conditions, 1 is combined with a primary amine 2 to produce 3 via the intermediate
product 4 in a one-pot transformation (Eq. (1)). Because an array of structurally diverse amines
is available, this approach to the construction of unsymmetrical 1,4-disubstituted-2,3-di-
ketopiperazines is convenient and broadly applicable.
(1)
* Corresponding author. E-mail: douglas beshore@merck.com
–
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01565-3

8736
Substrate 1a4a was used to explore the scope of the chemistry. Several amines were combined
with 1a under reductive amination conditions in the presence of sodium triacetoxyborohydride
(Table 1).5 Aliphatic amines produced diketopiperazines 3 in moderate to good yields (entries
1–9). Relatively unhindered amines (entries 1–3) led to moderate isolated yields of 64–75%,
which was the result of a second reductive alkylation of intermediate 4 with aldehyde 1a. The
more branched amines (entries 4–9) were less prone to the aforementioned secondary pathway,
thereby affording increased yields (76–88%). Improved cyclization efficiency may be a result of
increased nucleophilicity of the amine from alkyl group donation and/or decreased bis-alkyla-
tion due to steric hindrance. tert-Butyl amine 2f and quaternary-substituted amine 2i underwent
facile reductive amination, but cyclization was completed only after heating for 90 minutes.
Aromatic amines (entries 10–14) were also successfully transformed; electron withdrawing
(entries 12–13) and sterically hindered (entry 14) anilines required heating for complete
cyclization.
Table 1
Reductive amination of aldehyde 1a with amines 2 to produce diketopiperazines 3 (Eq. (1))
Scheme 1 is a proposed mechanistic pathway, which depicts two possible reaction manifolds
that differ in the order of alkylation and cyclization events. When benzylamine and 1a were
subjected to a variety of reaction conditions,6 we found that the proportion of isolable
by-product 11 decreased with the addition of acetic acid (0, 1, 2, 3 equiv. AcOH; 52%, 18%, 6%,
0% 11). Interestingly, when 11 was resubjected to the reaction conditions described in Table 1,
it failed to produce the desired product 3b from the putative intermediate 12. From this, we infer
that acetic acid affects the manifold by enhancing the conversion 7/8 to 9, diminishing the
conversion of 7/8 to 10/11 and/or enhancing the conversion of 9 to 4. Therefore, reactions
leading to the desired products described in Table 1 proceed via reductive amination, then

8737
cyclization (9“4“3), rather through the N-acyliminium intermediate 12.7 Interestingly, in the
case of phenyl glycinol 2g, 3 equivalents of acetic acid were not sufficient to suppress the
formation of oxazoline by-products cis-5 and trans-6 (Table 1, entry 7).8,9
Scheme 1.
Using (R)-a-methyl benzylamine as a chiral auxiliary, ketone 1b4b was investigated for
incorporation of asymmetry at C5 in a diketopiperazine (Scheme 2). Standard conditions cleanly
produced two separable diastereomers in a 6:1 ratio favoring (R,R)-13, the predicted major
diastereomer.10 Reduction of 13 with LiAlH4 produced 14, which was debenzylated with
11
Pearlman’s catalyst12 to afford (R)-1-ethyl-3-methylpiperazine 15 in 92% isolated yield over two
steps. The stereochemical assignment was supported by the analysis of Mosher’s amides of
15.13,14
,
Scheme 2. Reagents: (a) (R)-a-methyl benzyl amine (1.0 equiv.), AcOH (3 equiv.), Na(AcO)3BH (1.5 equiv.), 4 A MS,
DCE, 0°C–rt, 15 h. (b) LiAlH4, Et2O, 0°C–rt. (c) H2, Pd(OH)2, HCl, EtOH, 18 h
In summary, we have developed an alternative synthesis of unsymmetrical 1,4-substituted
diketopiperazines via tandem reductive amination–cyclization. The wide availability of aliphatic
and aromatic amines and the mild reaction conditions make this an attractive route for the
construction of unsymmetrical substituted diketopiperazines and piperazines.
Representative procedure: 4-benzyl-1-phenyl-2,3-diketopiperazine (3j). To a vigorously stirring
,
suspension of aniline 2j (77 mL, 0.85 mmol), 4 A powdered molecular sieves (300 mg), sodium
triacetoxyborohydride (270 mg, 1.28 mmol, 1.5 equiv.) and acetic acid (145 mL, 2.55 mmol, 3
equiv.) in 8 mL of 1,2-dichloroethane at 0°C was added aldehyde 1a (260 mg, 1.10 mmol) in 2
mL of 1,2-dichloroethane. After five minutes, the ice bath was removed and the reaction was
warmed to room temperature. After 15 hours, the reaction was slowly poured into saturated
aqueous NaHCO3 solution, which was then extracted three times with CH2Cl2. The combined
organic solutions were washed with brine, dried over Na2SO4, filtered, and concentrated in
vacuo. Purification by flash chromatography (1/9 to 1/1: EtOAc/CH2Cl2) provided diketopiper-
1
azine 3j as a white solid (191 mg, 80% yield); Rf 0.35 (50% EtOAc in CH2Cl2); H NMR (300
MHz, CDCl3) l 7.25–7.45 (m, 10 H), 4.75 (s, 2H), 3.89 (dd, J=7.6, 5.9 Hz, 2H), 3.57 (dd,
J=7.6, 5.9 Hz, 2H); ES HRMS exact mass calculated for C17H17N2O2 (M+H+): 281.1285, found
281.1297.

8738
Acknowledgements
We are grateful to the following: Dr. C. W. Ross III and Dr. H. G. Ramjit for mass spectra;
Dr. M. J. Bogusky and J. S. Murphy for NMR investigations; Dr. B. W. Trotter, Dr. M. A.
Patane and Dr. S. L. Graham for helpful discussions; and Ms. M. A. Guttman for manuscript
assistance.
References
1. (a) Giannis, A.; Kolter, T. Angew. Chem., Int. Ed. Engl. 1993, 32, 1244–1267. (b) Islip, P. J.; Closier, M. D.;
Neville, M. C. J. Med. Chem. 1973, 16, 1030–1034. (c) Hori, T.; Yoshida, C.; Murakami, S.; Kiba, Y.; Takeno,
R.; Nakano, J.; Nitta, J.; Tsuda, H.; Saikawa, I. J. Med. Chem. 1994, 37, 400–407.
2. (a) Bischoff, N. Chem. Ber. 1889, 1805. Riebsomer, J. L. J. Org. Chem. 1950, 15, 68–73. (b) Mueller-Westerhoff,
U.; Zhou, M. J. Org. Chem. 1994, 59, 4988–4992.
3. Dinsmore, C. J.; Bergman, J. M. J. Org. Chem. 1998, 63, 4131–4134.
4. (a) Synthesis of 1a: (i) benzylamine, ClCOCO2Me, Et3N, CH2Cl2, 0°C, 1 h. (ii) NaH, allyl bromide, DMF,
0°C–rt, 4 h, 54% over two steps. (iii) OsO4, NaIO4, acetone:H2O (1:1), rt, 2 h, crude product used without further
purification. (b) Synthesis of 1b: (i) N-ethyl-2-methylallylamine, ClCOCO2Me, Et3N, EtOAc, 0°C, 1 h, 100%. (ii)
OsO4, NaIO4, acetone:H2O (1:1), rt, 3 h, 58%.
1
5. Selected data for compounds: 1a: H NMR (400 MHz, CDCl3) 66:34 mixture of amide rotamers: major l 9.35
(s, 1 H), 7.19–7.35 (m, 5 H), 4.50 (s, 2 H), 4.01 (s, 2 H), 3.85 (s, 3 H); minor: l 9.37 (s, 1 H), 7.19–7.35 (m, 5
H), 4.64 (s, 2 H), 4.12 (s, 2 H), 3.78 (s, 3 H); C12H14NO4: m/z (ES+) 236.2 (MH+). 3a: 1H NMR (500 MHz,
CDCl3) l 7.15–7.38 (m, 10H), 4.64 (s, 2H), 3.70 (t, J=7.1 Hz, 2H), 3.21–3.53 (m, 4H), 2.94 (t, J=7.1 Hz, 2H);
HRMS (ES) exact mass calcd for C19H21N2O2 (M+H+): 309.1598; found 309.1585. 3b: 1H NMR (500 MHz,
CDCl3) l 7.26–7.36 (m, 10H), 4.67 (s, 4H), 3.33 (s, 4H); HRMS (ES) exact mass calcd for C18H19N2O2 (M+H+):
1
295.1441; found 295.1431. 3c: H NMR (500 MHz, CDCl3) l 7.27–7.36 (m, 5H), 4.66 (s, 2H), 3.42–3.45 (m, 2H),
3.37–3.40 (m, 2H), 2.85 (tt, J=4.0, 4 Hz, 1H). 0.89 (m, 2H), 0.71 (m, 2H); HRMS (ES) exact mass calcd for
C14H17N2O2 (M+H+): 245.1285; found 245.1269. 3d: 1H NMR (500 MHz, CDCl3) l 7.27–7.37 (m, 5H), 4.84
(sept, J=6.8 Hz, 1H), 4.68 (s, 2H), 3.36–3.39 (m, 2H), 3.32–3.35 (m, 2H), 1.14 (d, J=6.8 Hz, 6H); HRMS (ES)
exact mass calcd for C14H19N2O2 (M+H+): 247.1441; found 247.1425. 3e: 1H NMR (500 MHz, CDCl3) l
7.28–7.38 (m, 5H), 4.68 (s, 2H), 4.43 (tt, J=3.8, 11.9 Hz, 1H), 3.36 (s, 4H), 1.65–1.85 (m, 5H), 1.25–1.46 (m, 4H),
1.09 (dtt, J=13, 13, 3.7 Hz, 1H); HRMS (ES) exact mass calcd for C17H23N2O2 (M+H+): 287.1754; found
287.1744. 3f: 1H NMR (500 MHz, CDCl3) l 7.28–7.36 (m, 5H), 4.66 (s, 2H), 3.46–3.50 (m, 2H), 3.32–35 (m, 2H),
1
1.46 (s, 9H); HRMS (ES) exact mass calcd for C15H21N2O2 (M+H+): 261.1598; found 261.1585. 3g: H NMR
(500 MHz, CDCl3) l 7.19–7.34 (m, 10H), 4.66 (d, J=14.6 Hz, 1H), 4.54 (d, J=14.6 Hz, 1H), 4.44 (m, 1H),
3.84–3.88 (m, 2H), 3.32 (m, 1H), 3.21 (m, 1H), 3.05–3.09 (m, 2H), 3.01 (m, 1H), 2.95 (dd, J=14.1, 6.6 Hz, 1H),
2.84 (br s, 1H); HRMS (ES) exact mass calcd for C20H23N2O3 (M+H+): 339.1703; found 339.1703. cis-5: 1H
NMR (500 MHz, CDCl3) l 7.04–7.38 (m, 10H), 4.98 (dd, J=10.2, 3.7 Hz, 1H), 4.74 (d, J=14.5 Hz, 1H), 4.59
(d, J=14.5 Hz, 1H), 4.37 (m. 1H), 4.15 (br d, J=9.3 Hz, 1H), 3.86 (m, 1H), 3.44 (br d, J=13.5 Hz, 1H), 3.40
(dd, J=11.8, 3.8 Hz, 1H), 3.03 (dd, J=11, 11 Hz, 1H), 2.85 (dd, J=13.5, 9.1 Hz, 1H); HRMS (ES) exact mass
1
calcd for C20H21N2O3 (M+H+): 337.1547; found 337.1548. trans-6: H NMR (500 MHz, CDCl3) l 7.18–7.37 (m,
10H), 4.90 (dd, J=9.9, 4.6 Hz, 1H), 4.74 (d, J=14.6 Hz, 1H), 4.58 (d, J=14.6 Hz, 1H), 4.57 (m. 1H), 4.11 (dd,
J=9.2, 7.0 Hz, 1H), 3.80 (dd, J=9.2, 6.8 Hz, 1H), 3.42 (dd, J=12.4, 4.8 Hz, 1H), 3.30–3.36 (m, 2H), 2.88 (dd,
1
J=13.6, 9.2 Hz, 1H); HRMS (ES) exact mass calcd for C20H21N2O3 (M+H+): 337.1547; found 337.1548. 3h: H
NMR (300 MHz, CDCl3): l 7.14–7.40 (m, 10H), 5.35 (dd, J=11.0, 5.6 Hz, 1H), 4.67 (d, J=14.7 Hz, 1H), 4.52
(d, J=14.7 Hz, 1H), 3.74 (s, 3H), 3.42 (dd, J=14.9, 5.6 Hz, 1H), 3.25–3.36 (m, 2H), 3.22 (dd, J=7.3, 4.4 Hz,
1H), 3.05 (dd, J=14.8, 10.9 Hz, 1H), 2.98–3.05 (m, 1H); HRMS (ES) exact mass calcd for C21H23N2O4 (M+H+):
367.1652; found 377.1646. 3i: 1H NMR (500 MHz, CDCl3) l 7.28–7.37 (m, 3H), 7.21–7.27 (m, 5H), 7.10–7.13 (m,
2H), 4.69 (d, J=14.6 Hz, 1H), 4.58 (d, J=14.6 Hz, 1H), 3.82 (d, J=13.7 Hz, 1H), 3.76 (s, 3H), 3.29 (ddd,
J=12.8, 9.0, 3.9 Hz, 1H), 3.10 (ddd, J=12.8, 6.1, 3.9 Hz, 1H), 2.98 (d, J=13.7 Hz, 1H), 2.90 (ddd, J=12.9, 9.0,
3.9 Hz, 1H), 2.50 (ddd, J=12.7, 6.2, 3.9 Hz, 1H), 1.45 (s, 3H); HRMS (ES) exact mass calcd for C22H25N2O4

8739
1
(M+H+): 381.1809; found 381.1804. 3k: H NMR (300 MHz, CDCl3) l 7.32–7.41 (m, 5H), 7.23 (d, J=9.0 Hz,
2H), 6.91 (d, J=9.0 Hz, 2H), 4.75 (s, 2H), 3.81–3.87 (m, 2H), 3.80 (s, 3H), 3.53–3.58 (m, 2H); HRMS (ES) exact
mass calcd for C18H19N2O3 (M+H+): 311.1390; found 311.1381. 3m: 1H NMR (300 MHz, CDCl3) l 7.64 (d,
J=8.5 Hz, 2H), 7.45 (d, J=8.8 Hz, 2H), 7.24–7.34 (m, 5H), 4.69 (s, 2H), 3.88 (m, 2H), 3.53 (m, 2H). HRMS (ES)
1
exact mass calcd for C18H16N3O2 (M+H+): 306.1237; found 306.1222. 3n: H NMR (300 MHz, CDCl3) l 8.21 (d,
J=9.0 Hz, 2H), 7.51 (d, J=9.3 Hz, 2H), 7.25–7.36 (m, 5H), 4.70 (s, 2H), 3.90 (m, 2H), 3.55 (m, 2H). HRMS (ES)
1
exact mass calcd for C17H16N3O4 (M+H+): 326.1133; found 326.1138. 3p: H NMR (500 MHz, CDCl3) l 7.68
(dd, J=8.0, 1.2 Hz, 1H), 7.31–7.41 (m, 6H), 7.30 (dd, J=7.9, 1.6 Hz, 1H), 7.24 (dd, J=7.6, 1.5 Hz, 1H), 4.89
(d, J=14.7 Hz, 1H), 4.65 (d, J=14.6 Hz, 1H), 3.83 (m, 1H), 3.76 (m, 2H), 3.50 (m, 1H); HRMS (ES) exact mass
calcd for C17H16BrN2O2 (M+H+): 359.0390; found 359.0412.
6. Analogous results were obtained when aniline was used rather than benzylamine.
7. (a) While the generation of N-acyliminium ions (e.g. 12) from intermediates comparable to 10 and 11 using
Brønsted or Lewis acid conditions is well precedented (Ref. 7b), it is unlikely that 12 is formed under the
comparatively mild reaction conditions described herein. (b) Hiemstra, H.; Speckamp, W. N. In Comprehensive
Organic Synthesis; Trost, B. M., Fleming, I. Eds.; Pergamon Press: New York, 1991; Vol. 2, pp. 1047–1082.
8. Structures of cis-5 and trans-6, which were isolated from the reaction in 12 and 14%, respectively:
9. As a further control experiment, trans-6 was resubjected to the reaction conditions described in Table 1. After 6
hours, no detectable amount of 3g was formed. This supports our assertion that the generation of an
N-acyliminium ion (e.g. 12) from an N,O-acetal (e.g. 10 or trans-6) is unlikely to be a major contributor to the
formation a diketopiperazine under these conditions.
10. (a) Nichols, D. E.; Lloyd, D. H.; Johnson, M. P.; Hoffman, A. J. J. Med. Chem. 1988, 31, 1406–1412. (b) David,
D. M.; Pyne, S. G.; Kanemaguire, L. A. P. J. Chem. Soc., Chem. Commun. 1990, 888–889. (c) Fuller, J. C.;
Belisle, C. M.; Goralski, C. T.; Singaram, B. Tetrahedron Lett. 1994, 35, 5389–5392.
11. Van Der Burg, W. J.; Delobelle, J.; Ramon, C.; Vargaftig, B. J. Med. Chem. 1970, 13, 35.
12. (a) Chui, C. K.-F. Synth. Commun. 1996, 26, 577–584. (b) Gillard, J.; Abraham, A.; Anderson, P. C.; Beaulieu,
P. L.; Bogri, T.; Bousquet, Y.; Grenier, L.; Guse, I.; Lavallee, P. J. Org. Chem. 1996, 61, 2226–2231.
13. The absolute configuration was determined using the NMR-based method of Hoye and Renner for cyclic
secondary amines (Ref. 14c). Thus, compound 15 was converted to the corresponding (S)- and (R)-MTPA
amides i and ii (1.2 equiv. MTPA-Cl, 3.3 equiv. Et3N, CH2Cl2). Dominant conformations are illustrated below
1
for i (amide rotamer ratio 72:28) and ii (77:23). Selected H NMR spectral data (CDCl3) indicate chemical shift
differences are consistent with (R)-stereochemistry at C5 of 13.
14. (a) Dale, J. A.; Dull, D. L.; Mosher, H. S. J. Org. Chem. 1969, 34, 2543–2549. (b) Ohtani, I.; Kusumi, T.;
Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113, 4092–4096. (c) Hoye, T. R.; Renner, M. K. J. Org.
Chem. 1996, 61, 2056–2064.
.