A novel dealkylation affording 3-aminoimidazo[1,2-a]pyridines: Access to new substitution patterns by solid-phase synthesis

Article abstract of DOI:10.1016/S0040-4039(00)00003-4

The three-component synthesis of 3-aminoimidazo[1,2-a]pyridine derivatives has been extended by a novel acid-induced dealkylation reaction that removes the 1,1,3,3-tetramethylbutyl group derived from the isonitrile input. This reaction was conducted on the solid-phase using the HMBA linker and the resulting products subjected to reductive alkylation using several aldehydes, thereby accessing novel substitution patterns at the 3-position. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00003-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1495–1500
A novel dealkylation affording 3-aminoimidazo[1,2-a]pyridines:
access to new substitution patterns by solid-phase synthesis†
Christopher Blackburn ∗ and Bing Guan
Drug Discovery, Millennium Pharmaceuticals, 75 Sidney Street, Cambridge, MA 02139, USA
Received 2 November 1999; revised 16 December 1999; accepted 20 December 1999
Abstract
The three-component synthesis of 3-aminoimidazo[1,2-a]pyridine derivatives has been extended by a novel acid-
induced dealkyation reaction that removes the 1,1,3,3-tetramethylbutyl group derived from the isonitrile input. This
reaction was conducted on the solid-phase using the HMBA linker and the resulting products subjected to reductive
alkylation using several aldehydes, thereby accessing novel substitution patterns at the 3-position. © 2000 Elsevier
Science Ltd. All rights reserved.
Multiple component condensations (MCCs)1 have received renewed attention as a consequence of their
application to the combinatorial parallel synthesis of large arrays of compounds with diverse substitution
patterns. The number of products accessible in one step is given by the product of the number of
available inputs, and thus very large virtual libraries are possible. There is particular interest in MCCs
involving isonitrile inputs, such as the Passerini1b and Ugi2,3 reactions; post-condensation reactions
leading to constrained scaffolds are particularly notable.4 Recently, we5,6 and others7 described a new
three component condensation (3CC) involving 2-aminopyridine, an aldehyde and an isonitrile, e.g. 1,
leading to drug-like8–11 3-aminoimidazo[1,2-a]pyridines (Scheme 1). Access to a large and diverse series
of compounds based on this scaffold is limited by the commercial availability of the isonitrile component,
and by the low reactivity of the 3-amino group of these compounds when substituted by aryl or hindered
alkyl groups.5,6 We report here a new route to primary amines 2b–6b that improves on established
methods8–11 and is amenable to parallel synthesis. These primary amines proved to be more reactive
towards electrophiles than related secondary amines; in particular, reductive alkylation can be performed
using a variety of aldehydes on the solid-phase to give products that could not be accessed directly by
3CC because the appropriate isonitrile is not readily available.
In principle, substitution of cyanide ion for the isonitrile component in the 3CC reaction would lead
to the desired primary amines directly.12 A condensation reaction between 2-aminopyridine, methyl-4-
formylbenzoate and KCN in MeOH–H2O followed by purification on a cation exchange resin led directly
∗
†
Corresponding author. E-mail: blackburn@mpi.com (C. Blackburn)
Dedicated to the memory of Ray Blackburn, 1933–1999.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00003-4
tetl 16283

1496
Scheme 1. Reagents and conditions. (i) 2-Aminoazine, ArCHO, Sc(OTf)3, CH2Cl2, MeOH, 25°C, 48 h; (ii) a. 12N HCl:MeOH
(1:1), b. Dowex-500 cation exchange resin, c. 2N NH3, MeOH; (iii) TFA:CH2Cl2 (1:1), 25°C, 10 min
to primary amine 4b but in low yield (25%) and moderate purity (60%). We noted that 3CC products
derived from 1,1,3,3-tetramethylbutylisonitrile (1) were unstable towards acid treatment affording the
corresponding primary amines. To investigate this reaction further, compounds 2a–6a were prepared
as shown in Scheme 1; high yields were obtained with the exception of the condensation involving 2-
aminopyrimidine. Treatment of 2a–6a with 5N HCl in MeOH at ambient temperature for 0.5 h resulted
in clean dealkylation, giving primary amines 2b–6b which were conveniently isolated, in good to high
yields and high purities, by capture on Dowex cation exchange resin, washing with MeOH until acid-free,
and elution from the resin using methanolic ammonia. Presumably, this novel reaction is a reflection of
the ability of the 3-aminoimidazopyridine system to function as a leaving group, and the stability of a
carbocation or its elimination product. Isonitrile 1 thus serves as a cyanide ion equivalent in the 3CC
reaction, affording two-step access to the primary amino compounds. These reaction conditions are more
compatible with library synthesis than hydrogenolysis of N-benzyl compounds5 or the well established
method involving the condensation of α-halocarbonyl compounds with the appropriate 2-aminoazine and
subsequent nitrosation and reduction.8–11
In order to apply this synthesis to the solid-phase, we sought dealkylation conditions more compatible
with the swelling characteristics of polystyrene-based resins. Using the methyl ester 4a to mimic the resin
anchoring point, we compared the kinetics of dealkylation induced by TFA–CH2Cl2 with those of the t-
butyl analogue 4c (Table 1).13 Compound 4a exhibited clean first order kinetics and a half life of 100 s
whereas the t-butyl analogue 4c, which would also be expected to afford a stable carbocation, underwent
much slower dealkylation (t1/2>20 h), during which time trifluoroacetylation of the amino group also
took place. These observations led us to apply the 3CC reaction involving isonitrile 1 to the solid-phase,
employing the base-labile linker 4-hydroxymethylbenzoic acid (HMBA). Esters derived from this resin
are more stable than those obtained from Merrifield resin14 towards the acidic conditions used in the
dealkylation step, and can be cleaved by base under milder conditions.15 4-Carboxybenzaldehyde was
anchored to HMBA-AM resin in the presence of DIC and DIEA, and the resulting resin-bound aldehyde
subjected to a 3CC reaction with 2-aminopyridine and isonitrile 1 (Scheme 2) to give resin 7a. Cleavage
of 7a by treatment with sodium methoxide in MeOH:THF (1:1) afforded 4a in high yield and purity.
Treatment of 7a with TFA:CH2Cl2 (1:1) for 10 min, gave resin 7b which could be cleaved to give primary

1497
amine 4b in good yield and high purity (Table 1, entries 1 and 2). In order to prepare 3-alkyl derivatives
not accessible from commercially available isonitriles, we investigated the reductive alkylation of resin-
bound amine 7b by treatment with aldehydes in the presence of NaBH3CN. After cleavages induced
by treatment with sodium methoxide, monobenzylated derivatives were obtained from aryl aldehydes
in high yields and purities (entries 3–5, and 7), although yields were lowered by steric hindrance at the
ortho position of the aldehyde (entry 6). Cyclopropanecarboxyaldehyde and phenylacetaldehyde afforded
mixtures of mono and dialkylated derivatives (entries 8 and 9).
Table 1
Solid-phase synthesis of aminoimidazol[1, 2-a]pyridine and pyrimidine derivatives
2-Aminopyrimidine and aminopyrazine afforded resin-bound primary amines 8b and 9b, respectively,
by solid-phase 3CC followed by dealkylation (Scheme 2). Cleavage of these resins by treatment with
sodium methoxide afforded 5b and 6b, respectively, albeit in lower purities than 4b (Table 1 entries 2, 11

1498
Scheme 2. Reagents and conditions. (i) 2-aminoazine, isonitrile 1, Sc(OTf)3, CH2Cl2, MeOH, 25°C, 60 h; (ii) TFA:CH2Cl2
(1:1), 25°C, 10 min; (iii) RCHO, NaBH3CN, DMF–HOAc, 25°C, 50 h; (iv) 0.25 M NaOMe in MeOH:THF (1:1), 25°C, 1 h;
(v) RCOCl, DMF-pyridine (4:1), 25°C, 24 h, or RNCO, DMF, 25°C, 24 h
and 13). Solid-phase reductive alkylations of 8b (entries 14–20) gave products in lower yields and purities
than those derived from resin 7b. In marked contrast, treatment of resin 9b, derived from aminopyrazine,
with the same series of aldehydes under identical reducing conditions, gave rise to mixtures in which
the expected product was a minor component. LC-MS data indicated that reduction of the pyrazine
ring16 was taking place, together with further reductive alkylation reactions. Thus, for example, the
major product from the reductive alkylation of resin 9b with cyclopropanecarboxaldehyde was shown
to be compound 10 (ca. 50% purity by LC-MS diode array trace), in which the pyrazine ring had
been reduced and the secondary amine alkylated. Reductive alkylation of 9b by 4-methoxybenzaldehyde
afforded a reduced and monoalkylated derivative as the major product (ca. 35% purity by LC-MS diode
array trace). To investigate the latter reaction further, 6b was subjected to reductive alkylation with 4-
methoxybenzaldehyde in DMF solution. The major product, isolated in 24% yield after HPLC, was
shown (see experimental) to be compound 11. Reductions of the imidazo[1,2-a]pyrazine heterocycle
on-resin is of interest for accessing a new scaffold for combinatorial synthesis and is under further
investigation.
Primary amines 2b–6b were found to be more reactive towards acid chlorides and isocyanates
than related secondary amines and could be functionalised in solution followed by polymer-supported
quenching.5,17 Derivatives of 4b were more conveniently prepared on the solid-phase by treatment with
excess acid chloride or isocyanate prior to cleavage using sodium methoxide (Table 1 entries 21–24).

1499
1. Experimental
The following procedure is typical for the solution-phase synthesis of primary amines 2b–6b. A 0.5 M
solution of 2-aminopyridine (1.5 mL) in MeOH:CH2Cl2 (1:3) was treated with methyl-4-formylbenzoate
(1 equiv.) and scandium triflate (0.1 equiv.). After 0.5 h, isonitrile 1 (1 equiv.) was added and the solution
was stirred for 48 h at ambient temperature. The solvent was evaporated and the crude product purified
1
by chromatography on neutral alumina eluting with ethyl acetate–hexane to give 4a (85%). H NMR
(CDCl3, 300 MHz) δ 8.22 (1H, d, J=7Hz), 8.12 (2H, d, J=9 Hz), 7.98 (2H, d, J=9 Hz), 7.55 (1H, d, J=7
Hz), 7.16 (1H, dd), 6.78 (1H, dd), 3.94 (3H, s), 1.60 (2H, s), 1.04 (9H, s), 0.95 (6H, s). Dealkylation as
described in the text gave primary amine 4b (75%) 1H NMR (CD3OD, 300 MHz) δ 8.22 (1H, d, J=7Hz),
8.08 (4H, s), 7.44 (1H, d, J=7 Hz), 7.20 (1H, dd), 6.94, (1H, dd), 3.94 (3H, s).
Solid-phase syntheses: HMBA-AM resin 1.15 mmol/g loading (from Novabiochem) (1.6 g,
1.84 mmol) was allowed to swell in DMF (12 mL) and CH2Cl2 (12 mL) and treated with 4-
carboxybenzaldehyde (1.11g, 7.4 mmol), diisopropylcarbodiimide (1.15 mL, 7.4 mmol), and 4-
dimethylaminopyridine (0.225 g, 1.86 mmol). The mixture was shaken at ambient temperature for 24
h then filtered and the resin washed successively with DMF, CH2Cl2, MeOH, CH2Cl2 and dried under
vacuo. The loading of the aldehyde resin was found to be 1.0 mmol/g (quantitative) determined by
cleavage of an aliquot of resin using 0.25 M NaOMe in MeOH:THF (1:1) for 1 h at ambient temperature
and quantification of the released methyl-4-formylbenzoate by LC-MS using authentic material as
calibrant. Theoretical loading capacity18 is given by 1.15 mmol/[1g+(1.15 mmol×MW added to resin)].
The resin-bound aldehyde (0.4 g, 0.4 mmol) was treated with a solution of 2-aminopyridine (0.38 g, 4
mmol) and Sc(OTf)3 (200 mg, 0.4 mmol) in CH2Cl2 (5 mL) and MeOH (1.6 mL). After shaking for 30
min at ambient temperature, isonitrile 1 (0.56 g, 4 mmol) was added and the mixture shaken for 60 h at
ambient temperature. Resin 7a was washed successively with CH2Cl2, MeOH, H2O, MeOH, DMF and
CH2Cl2 and dried under vacuo. Dealkylation of resin 7a was effected by treatment with TFA:CH2Cl2
(1:1) for 10 min at ambient temperature (longer reaction times resulted in trifluoroacetylation of the
primary amine). The loadings of resins 7a and 7b were determined by cleavage of an aliquot of resin by
treatment with 0.25 M NaOMe in MeOH:THF (1:1) for 1 h at ambient temperature and quantification
of the released esters by HPLC (see Table 1). Resins 8b and 9b were prepared similarly. Reductive
alkylations of resins 7b–9b were carried out by treatment with the appropriate aldehyde (8 equiv.) and
NaBH3CN (24 equiv.) in DMF:HOAc (7:1) for 50 h; product resins were washed successively with
DMF, MeOH, H2O, MeOH, DMF and CH2Cl2 and cleaved as described above. Yields and purities were
calculated as described in Table 1.
Compound 11 was prepared in the solution phase by treatment of 6b (40 mg, 0.15 mmol) with 4-
methoxybenzaldehyde (2.5 equiv.) in 2.5 ml of DMF:HOAc (15:1) in the presence of NaBH3CN (10
equiv.) for 72 h at ambient temperature. The crude product was subjected to prep. HPLC on a C-18
column eluting with a H2O–MeCN gradient and the major product of mass 392 was isolated (14 mg,
1
24%). H NMR (CDCl3, 600 MHz) δ 7.98 (2H, d, J=9 Hz), 7.84 (2H, d, J=9 Hz), 7.06 (2H, d, J=9
Hz), 6.78 (2H, d, J=9 Hz), 4.20 (1H, d, J=17Hz), 4.05 (1H, d, J=14Hz), 3.98 (1H, d, J=14Hz), 3.96 (1H,
m), 3.74 (3H, s), 3.54 (1H, m), 3.10 (1H, m). The benzylic protons resonating at 4.05 and 3.98 ppm
showed NOEs with the aromatic protons at 7.84 and 7.06 but no correlations with the protons of the
tetrahydropyrazine ring system indicating that the primary amino group was the site of alkylation.

1500
Acknowledgements
We thank Ian Parsons, Izumi Takagi, ZhiSong Ji and Zhi-Dong Jiang for analytical support and Lee
Herman for proof reading the manuscript.
References
1. (a) Armstrong, R. W.; Brown, D. S.; Keating, T. A.; Tempest, P. A. In Combinatorial Chemistry Synthesis and Applications;
Wilson, S. R.; Czarnik, A. W., Eds.; John Wiley & Sons: New York, 1997; p. 153. (b) Armstrong, R. W.; Combs, A. P.;
Tempest, P. A.; Brown, S. D.; Keating, T. A. Acc. Chem. Res. 1996, 29, 123–131. Strocker, A. M.; Keating, T. A.; Tempest,
P. A.; Armstrong, R. W. Tetrahedron Lett. 1996, 37, 1149–1152.
2. (a) Ugi, I. Isonitrile Chemistry; Academic Press: London, 1971. (b) Ugi, I.; Domling, A.; Horl, W. Endeavour 1994, 18,
115–122.
3. (a) Cao, X.; Moran, E. J.; Siev, D.; Lio, A.; Ohasi, C.; Mjalli, A. M. M. Bioorg. Med. Chem. Lett. 1995, 5, 2953–2958. (b)
Short, K. M.; Ching, B. W.; Mjalli, A. M. M. Tetrahedron 1997, 53, 6653–6679.
4. Hulme, C.; Cherrier, M.-P. Tetrahedron Lett. 1999, 40, 5295–5299.
5. Blackburn, C.; Guan, B.; Fleming, P.; Shiosaki, K.; Tsai, S. Tetrahedron Lett. 1998, 39, 3635–3638.
6. Blackburn, C. Tetrahedron Lett. 1998, 39, 5469–5472.
7. (a) Groebke, K.; Weber, L.; Fridolin, M. Synlett 1998, 661–663. (b) Bienayme, H.; Bouzid, K. Angew. Chem., Int. Ed. Engl.
1998, 37, 2234–2237.
8. Starrett, J. E.; Montzka, T. A.; Crosswell, A. R.; Cavanagh, R. L. J. Med. Chem., 1989, 32, 2204–2210.
9. Sabalayrolles, C.; Cros, G. H.; Milhavet, J. C.; Rechenq, E.; Chapat, J. P.; Boucard, M.; Serrano, J. J.; McNeill, J. H. J. Med.
Chem. 1984, 27, 206–212.
10. Rival, Y.; Grassy, G.; Michel, G. Chem. Pharm. Bull. 1992, 40, 1170–1176.
11. (a) Fisher, M. H.; Lusi, A. J. Med. Chem. 1972, 15, 982–985. (b) Trapani, G.; Franco, M.; Ricciardi, L.; Latrofa, A.; Genchi,
G.; Sanna, E.; Tuveri, F.; Cagetti, E.; Biggio, G.; Liso, G. J. Med. Chem., 1997, 40, 3109–3118.
12. For related reactions, see: (a) Bristow, N. W.; Charlton, P. T.; Peak, D. A.; Short, W. F. J. Chem. Soc. 1954, 616–629. (b)
Tadeka, K.; Shudo, K.; Okamoto, T.; Kousuge, T. Chem. Pharm. Bull. 1978, 26, 2924. (c) Knott, E. B. J. Chem. Soc. 1956,
1360. (d) Saint–Ruf, G.; Loukakou, B.; N’Zousi, C. J. Heterocyclic Chem. 1981, 18, 1565–1570.
13. The compounds were dissolved in TFA–CH2Cl2 at a concentration of 50 mM and dealkylations were allowed to proceed
at ambient temperature. Aliquots were removed at time intervals, quenched by treatment with methanolic ammonia and
subjected to HPLC analysis.
14. Syntheses were less successful when Merrifield chloromethyl-PS resin was used in place of HMBA-AM because of partial
cleavages from the resin during the acid-induced dealkylation step.
15. (a) Hutchins, S. M.; Chapman, K. T. Tetrahedron Lett. 1996, 37, 4869–4872. (b) Cheng, Y.; Chapman, K. T. Tetrahedron
Lett. 1997, 38, 1497–1500.
16. For examples of the reduction of related heterocylcles, see: (a) Chavignon, O.; Teulade, J. C.; Madesclaire, M.; Geueiffier,
A.; Blache, Y. J. Heterocyclic Chem. 1992, 29, 691–697. (b) Bonnet, P. A.; Sablayrolles, C.; Chapat, J. P. J. Chem. Res. (M)
1984, 468–480.
17. (a) Kaldor, S. W.; Siegel, M. G.; Fritz, J. E.; Dressman, B. A.; Hahn, P. J. Tetrahedron Lett. 1996, 37, 7193–7196. (b) Booth,
R. J.; Hodge, J. C. J. Am. Chem. Soc. 1997, 119, 4882–4886. (c) Flynn, D. L.; Crich, J. Z.; Devraj, R. V.; Hockerman, S. L.;
Parlow, J. J.; South, M. S.; Woodard, S. J. Am. Chem. Soc. 1997, 119, 4874–4881.
18. Metcalf, C. A.; Vu, C. B.; Sundaramoorthi, R.; Jacobsen, V. A.; Laborde, E. A.; Green, J.; Green, Y.; Macek, K. J.; Merry,
T. J.; Pradeepan, S. G.; Uesugi, M.; Varkhedkar, V. M.; Holt, D. A. Tetrahedron Lett. 1998, 39, 3435–3438.