Traceless solid-phase synthesis of 1,2,4-triazoles using a novel amine resin

Article abstract of DOI:10.1021/ol016578a

Equation presented A solid-phase route to 5-aryl-3-arylthiomethyl-1,2,4-triazoles has been developed that permits the incorporation of two elements of diversity. The heterocycle was constructed upon a novel 4-benzyloxy-2-methoxybenzylamine (BOMBA) resin,

Full text of DOI:10.1021/ol016578a

ORGANIC
LETTERS
2001
Vol. 3, No. 21
3341-3344
Traceless Solid-Phase Synthesis of
1,2,4-Triazoles Using a Novel Amine
Resin
Scott D. Larsen* and Brian A. DiPaolo
Combinatorial and Medicinal Chemistry Research, Pharmacia Corporation,
Kalamazoo, Michigan 49007
scott.d.larsen@pharmacia.com
Received August 14, 2001
ABSTRACT
A solid-phase route to 5-aryl-3-arylthiomethyl-1,2,4-triazoles has been developed that permits the incorporation of two elements of diversity.
The heterocycle was constructed upon a novel 4-benzyloxy-2-methoxybenzylamine (BOMBA) resin, from which traceless cleavage could be
effected with dilute TFA. A library of 96 triazoles was produced with an average yield of 26% (84% yield per step) and an average purity of
80%. In a direct comparison, the new BOMBA resin proved to be markedly superior to Rink Amide resin for this application.
In conjunction with an ongoing therapeutic effort, we
required the development of a library synthesis of 1,2,4-
triazoles that would permit facile variation at the 3 and 5
carbons. 1,2,4-Triazoles have been prepared by a number
of routes, but only a couple have been adapted to solid
phase.1 The most commonly employed approach entails the
direct condensation of thioamides with hydrazides, but this
was not appealing because these starting materials are not
commercially available in sufficient diversity. An alternative
strategy was thus considered that was less concise but more
flexible (Scheme 1). Amidrazones 2 (G ) aryl), readily
derived from acid chlorides 1, have been reported by Thiel
to react with chloroacetyl chloride to afford 3-chloromethyl-
1,2,4-triazoles 3,2 intermediates that seemed ideal for the
subsequent introduction of diversity via nucleophilic addition.
Replacement of the G group on the nitrogen of 2 with a
resin linker offered the prospect for a solid-phase synthesis
that would culminate in the traceless cleavage of triazoles
5.
The success of this approach relied upon the selection of
an appropriate resin linker that would permit the release of
the triazoles under mild conditions. An excellent precedent
was provided by Bilodeau, who used a 2-methoxy-4-
alkoxybenzyl linker for preparing imidazoles on resin, which
were subsequently cleaved with acetic acid.3 Assuming that
Scheme 1
(1) (a) Katritzky, A. R.; Qi, M.; Feng, D.; Zhang, G.; Griffith, M. C.;
Watson, K. Org. Lett. 1999, 1, 1189-91. (b) Wilson, M. W.; Hernandez,
A. S.; Calvet, A. P.; Hodges, J. C. Mol. DiVersity 1998, 3, 95-112.
(2) Thiel, W. Z. Chem. 1990, 30, 365-367.
10.1021/ol016578a CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/22/2001

Scheme 2a
a Reagents and conditions: (a) R1PhCOCl, Et3N, CH2Cl2; (b) Lawesson’s reagent, pyr, 85 °C; (c) MeOTf, CH2Cl2, rt; (d) hydrazine,
2-methoxyethanol, rt; (e) ClCOCH2Cl, CH2Cl2, rt; (f) 1:1 DMF/HOAc, rt; (g) R2PhSH, KOH, 2-methoxyethanol; (h) 10% TFA/CH2Cl2.
triazoles would be released at least as readily as imidazoles,
this linker seemed ideal for our application.
surprise, we could find no literature precedent for the
requisite polystyrene amine resin 6b, despite the fact that
the corresponding benzyl alcohol resin is commercially
available (SASRIN).4 Related commercial amine resins,
including Rink, PAL and Knorr, were rejected on the basis
of excessive steric bulk, extreme acid sensitivity, or the
presence of incompatible functionality (carboxamide).5 Closely
related literature resin BOBA6 (14, Figure 1) was deemed
Chemistry was developed initially in solution, using 2,4-
dimethoxybenzylamine (6a) as a model for the eventual resin
linker (Scheme 2). Anticipating that eventual library produc-
tion would be executed in IRORI MiniKans, conditions were
limited to those compatible with Kan integrity. Acylation
with 4-chlorobenzoyl chloride to make amide 7a (62%) was
straightforward. Conversion to thioamide 8a (90%) was
effected with Lawesson’s reagent in hot pyridine. Direct
conversion of the thioamide to amidrazone 9a by heating
with hydrazine was not successful, so 8a was first converted
to the corresponding thioimidate via alkylation with methyl
triflate, which subsequently reacted smoothly with hydrazine
at room temperature to afford 9a (54% overall). Acylation
with chloroacetyl chloride then produced acylated amidra-
zone 10a (94%), which was isolated as a mixture of regio-
and/or stereoisomers.
Figure 1. Literature benzylamine resins.
The key cyclization of 10a required some development.
Direct heating, as reported by Thiel,2 provided the desired
triazole 11a but was accompanied by significant cleavage
of the dimethoxybenzyl group. However, when the crude
amidrazone 10a, isolated as the HCl salt, was first neutralized
by washing with dilute aqueous NaOH, cyclization in DMF/
HOAc proceeded smoothly (92%) at room temperature with
no detectable debenzylation. Reaction of the resulting
chloromethyltriazole 11a with 4-chlorothiophenol in alkaline
2-methoxyethanol then provided the desired thiomethyltria-
zole 12a (80%). The key model cleavage of the benzyl group
was effected in 90 min at room temperature with 50% TFA
in methylene chloride, affording 13a in 85% yield.
insufficiently reactive to release triazoles under mild acidic
conditions, and the 2-methoxy-4-benzyloxy resin 157 (Figure
1), derived from (HMPB)-MBHA resin, also possessed an
incompatible butyramide linkage.
A number of routes to resin 6b were examined, only one
of which proved reliable upon scale-up (Scheme 3). AMEBA
resin 16,8 prepared from Merrifield resin (IRORI Unisphere
200), was converted to oxime resin 17 with hydroxylamine
(4) Registered trademark of BACHEM.
(5) For an excellent recent review on resin linkers, see: Guillier, F.;
Orain, D.; Bradley, M. Chem. ReV. 2000, 100, 2091-2157.
(6) Kobayashi, S.; Aoki, Y. Tetrahedron Lett. 1998, 39, 7345-7348.
(7) Cho, C. Y.; Youngquist, R. S.; Paikoff, S. J.; Beresini, M. H.; Hebert,
A. R.; Berleau, L. T.; Liu, C. W.; Wemmer, D. E.; Keough, T.; Schultz, P.
G. J. Am. Chem. Soc. 1998, 120, 7706-7718.
With a viable solution-phase route in hand, we turned our
attention to translating the chemistry onto resin. To our
(3) Bilodeau, M. T.; Cunningham, A. M. J. Org. Chem. 1998, 63, 2800-
2801.
(8) Fivush, A. M.; Willson, T. M. Tetrahedron Lett. 1997, 38, 7151-
7154.
3342
Org. Lett., Vol. 3, No. 21, 2001

6b actually afforded nearly double the mass of 13a, indicat-
ing a clear superiority for this particular application.
Scheme 3a
To gauge the generality of the solid-phase synthesis in
Scheme 2, a two-dimensional library of triazoles was
prepared. Eight aroyl chlorides and 12 aryl thiols were
combined to construct a 96-member library, using the IRORI
AccuTag-100 system. Following cleavage, HPLC/MS analy-
sis of the entire library (data not shown) indicated that every
analogue had been successfully prepared in purities ranging
from 20% to 99%. Specific yield and purity data for
representative analogues (at least one example of each
diversity element) are compiled in Table 1. The average yield
a Reagents and conditions: (a) NH2OH‚HCl, pyr, rt, 30 h; (b) 1
M LiAlH4, THF, rt, 24 h.
hydrochloride in pyridine. Reduction to amine resin 6b was
then effected with 1 M LiAlH4 in THF at room temperature.9
The overall yield of 6b from Merrifield resin was 28% (final
loading ) 0.44 mmol/g, determined by FMOC quantitation).
The solution-phase route depicted in Scheme 2 translated
onto solid phase with remarkable ease (6b-13a). The only
variable that consistently needed to be changed was reaction
time. One notable exception was the key cyclization of 10b
to triazole 11b. Whereas neutralization of 10a was critical
for success in solution, we observed no apparent need for
neutralizing resin 10b. Cyclization to 11b occurred with
equal yield and purity regardless of whether 10b was used
directly or first washed with triethylamine. One possible
explanation is that cleavage from the resin is simply not as
facile as the solution-phase debenzylation. Final cleavage
from resin 12b proceeded under mild conditions (10% TFA/
CH2Cl2, rt, 1 h), affording 13a in 65% overall yield (ca. 95%
yield per step) and 78% purity (HPLC at 210 nm).
Rink amide resin (IRORI Unisphere 200) was also
evaluated as a support in this synthesis in a direct side by
side comparison with 6b in which 60 mg of each resin was
placed in separate IRORI MiniKans and subjected to the
eight-step sequence in Scheme 2. In this experiment, resin
6b afforded 7.8 mg of 13 (88% overall yield) in 85% purity,
while Rink amide resin provided only 4.4 mg (23% overall
yield) in <80% purity. Despite its considerably lower loading
(0.44 mmol/g 6b vs 0.94 mmol/g Rink amide), the new resin
Table 1. Yield and Purity Data for Representative Analogues
from Triazole Library
cmpd
R1
R2
puritya (%)
yieldb (%)
13a
13c
13d
13e
13f
13g
13h
13i
4-Cl
4-Cl
4-Cl
4-Cl
4-OMe
4-Cl
83
37
85
83
82
89
85
71
85
86
96
85
58
81
83
87
80
84
83
33
13
24
25
36
31
28
23
23
34
29
29
29
28
27
33
33
36
25
2,4-diOMe
4-OMe
4-Cl
H
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
3,4-diCl
H
3,4-diOMe
pentafluoro
4-AcNH
3,4-diCl
2-naphthylc
4-Me
4-Br
4-Cl
2-CH2OH
4-Cl
4-Cl
13j
13k
13l
13m
13n
13o
13p
13q
13r
13s
13t
4-Cl
2-naphthylc
4-Me
4-CF3
4-Cl
4-Cl
4-NH2
a Determined by HPLC/MS (210 nm). b Overall unpurified yield from
(9) Pietta, P. G.; Cavallo, P. F.; Takahashi, K.; Marshall, G. R. J. Org.
Chem. 1974, 39, 44-48.
resin 6b. c Denotes R1 or R2 being fused phenyl ring.
(10) Synthesis of BOMBA Resin 6b. To a suspension of IRORI
Unisphere 200 Merrifield resin (1.10 g, 1.58 mmol/g) in DMF (25 mL)
was added sodium methoxide (0.293 g, 5.43 mmol), followed by 4-hydroxy-
2-methoxybenzaldehyde (0.828 g, 5.44 mmol). The mixture was heated to
65 °C for 24 h, after which time the resin was filtered and washed
successively with DMF, MeOH, H2O, MeOH, CH2Cl2, and MeOH. The
resin was dried in vacuo; 16 was obtained as an off-white resin (1.05 g).
FTIR indicated aldehyde formation with a strong absorption peak at 1678
cm-1. AMEBA resin 16 (1.05 g) and hydroxylamine hydrochloride (0.70
g, 10.08 mmol) were suspended in pyridine (10 mL). After agitating for 30
h at room temperature, the resin was filtered and washed with 1:1 pyridine/
H2O, H2O, MeOH, CH2Cl2, and MeOH. FTIR analysis indicated complete
disappearance of the carbonyl peak at 1678 cm-1; 17 was obtained as an
off-white resin (1.04 g). To a suspension of oxime resin 17 (0.50 g, 0.64
mmol) in THF (3.8 mL) was added 1 M lithium aluminum hydride in THF
(1.3 mL). The mixture was stirred for 24 h at room temperature, after which
time the reaction mixture was cooled to 0 °C. The excess lithium aluminum
hydride was slowly quenched with EtOH, and the aluminum salt byproducts
were then dissolved with a minimal amount of 10% HCl. Once dissolved,
the resin was filtered and washed with MeOH and CH2Cl2 (alternating,
three cycles), followed by washing with 10% triethylamine in CH2Cl2 to
neutralize the amine. Following a final wash with CH2Cl2, the resin was
dried in vacuo. 6b was obtained as a brownish-yellow resin (0.50 g). Loading
of resin, determined by FMOC quantitation, was 0.44 mmol/g.
for the entire library was 26%, with an average purity of
80%. Because the library analogue yields were lower than
expected from the development work, a second cleavage was
performed, resulting in an additional average of 5.0 mg (50%
yield) for each analogue that was of slightly lower average
purity (71%). Thus the actual average yield for the library
was closer to 75%, although optimum purity was achieved
by performing a single lower-yielding cleavage operation.
To expand the scope of our triazole synthesis, two
alkanethiols (3-phenyl-1-propanethiol and benzyl mercaptan)
were successfully reacted with resin 11b, affording the
corresponding triazoles 13 in 71% and 74% purities,
respectively, after TFA cleavage. Attempts to incorporate
an alkyl acid chloride into Scheme 2 were less promising,
however. At the acyl amidrazone formation stage, the
Org. Lett., Vol. 3, No. 21, 2001
3343

solution-phase chemistry became very complex, discouraging
us from expanding the solid-phase chemistry in this direction.
In conclusion, we have developed a novel benzylamine
resin (4-benzyloxy-2-methoxybenzylamine, BOMBA)10 on
which a de novo solid-phase synthesis11 of disubstituted
1,2,4-triazoles was successfully demonstrated. To the best
of our knowledge, this is the first reported synthesis of 1,2,4-
triazoles entailing tethering of the nascent heterocycle to the
resin through a ring nitrogen, thereby permitting a “traceless”
synthesis. The construction of heterocycles on solid phase
via linkage directly to the ring has proven in general to be
a valuable strategy because it permits facile variation of
substituents.12 The overall yields and purities realized in the
production of a 96-member library suggest that the chemistry
is sufficiently general to accommodate a variety of substit-
uents, although clearly diversity would be expected to be
tolerated to the greatest degree in the nucleophilic addition
step. It is worth noting that the success of this synthetic route
was greatly facilitated by the IRORI AccuTag technology,13
which is able to conveniently accommodate the introduction
of diversity early in a synthetic sequence.
(11) Procedures for the Solid-Phase Synthesis of 5-(4-Chlorophenyl)-
3-{[(4-chlorophenyl)thio]methyl}-1H-1,2,4-triazole (13a) from Resin 6b.
Six IRORI MiniKans, filled with BOMBA resin 6b (60 mg in each), were
placed in a 100-mL round-bottom flask, to which were added CH2Cl2 (30
mL) and triethylamine (2.09 mL, 15.0 mmol). The MiniKans were degassed.
After cooling to 0 °C, 4-chlorobenzoyl chloride (1.90 mL, 15.0 mmol) was
added and the MiniKans were stirred at 0 °C for an additional 30 min.
Once warmed to room temperature, the MiniKans were stirred for another
6 h. The reaction solution was then decanted off, and the MiniKans were
washed with CH2Cl2 and MeOH (alternating, three wash cycles). Following
a final wash with pyridine, the MiniKans containing resin 7b were dried in
vacuo. To a solution of Lawesson’s reagent (1.60 g, 3.96 mmol) in pyridine
(25 mL) was added 7b. The MiniKans were degassed, and the reaction
mixture was heated to 85 °C. After heating for 24 h, the reaction solution
was decanted, and the MiniKans were washed with CH2Cl2 and MeOH
(three alternating cycles). After a final wash with pyridine, the MiniKans
were resubjected to the reaction conditions for an additional 24 h. Once
again the reaction solution was decanted, and the MiniKans were washed
with 1:1 pyridine/H2O, followed by CH2Cl2 and MeOH (three alternating
cycles). After a final CH2Cl2 wash, the MiniKans were dried in vacuo. To
the six IRORI MiniKans, now containing thioamide resin 8b, in a 100 mL
round-bottom flask was added CH2Cl2 (20 mL). The mixture was degassed
before the addition of methyl triflate (0.907 mL, 8.00 mmol). After agitating
overnight at room temperature, the reaction solution was decanted, and the
MiniKans were washed with CH2Cl2 and acetonitrile (alternating, three
cycles). After a final wash with 2-methoxyethanol in preparation for the
next step, the MiniKans were dried in vacuo. The MiniKans containing
thioimidate resin were submersed in 2-methoxyethanol (25 mL), degassed,
and mechanically agitated for 10 min before hydrazine monohydrate (1.21
mL, 25.0 mmol) was added. After agitating for 24 h at room temperature,
the reaction solution was decanted, and the MiniKans were washed with
CH2Cl2 and acetonitrile (three alternating wash cycles). After a final wash
with 2-methoxyethanol, the MiniKans were resubjected to the reaction
conditions for an additional 24 h. The reaction solution was again decanted,
and the MiniKans were washed with CH2Cl2 and acetonitrile (alternating,
three cycles). After the final wash with CH2Cl2, the MiniKans were dried
in vacuo. To the MiniKans containing amidrazone resin 9b was added
CH2Cl2 (25 mL). The mixture was degassed before the addition of
chloroacetyl chloride (2.00 mL, 25.0 mmol). After the reaction was capped
and mechanically shaken at room temperature for 24 h, the reaction solution
was decanted, and the MiniKans were washed with CH2Cl2 and MeOH
(three alternating cycles). Finally, the MiniKans were washed with DMF
and dried in vacuo. Acylated resin 10b, in IRORI MiniKans, was submersed
in 1:1 DMF/AcOH (30 mL), degassed, and mechanically agitated at room
temperature for 20 h. The reaction solution was decanted, and the MiniKans
were washed with CH2Cl2 and MeOH (alternating, three wash cycles). The
MiniKans were dried in vacuo. A small sample of the resin was then cleaved
with 10% TFA in CH2Cl2. HPLC/MS analysis at this step indicated a purity
Supporting Information Available: Full experimental
details for the solution-phase synthesis of 13a from 6a,
including analytical and spectroscopic data for each inter-
mediate, and procedures for the synthesis of the 96-member
triazole library with the IRORI AccuTag-100 system. This
material is available free of charge via the Internet at http://
pubs.acs.org. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL016578A
of 96% at 254 nm and 87% at 210 nm. In a 50-mL round-bottom flask,
potassium hydroxide (1.68 g, 30 mmol) was dissolved in 1:1 THF/2-
methoxyethanol (30 mL). After the solution was degassed, 4-chlorothiophe-
nol (4.34 g, 30 mmol) was added, and the solution was stirred for 10 min.
Triazole resin 11b, in IRORI MiniKans, was added, and the reaction mixture
was once again degassed. After stirring at room temperature for 24 h, the
reaction solution was decanted, and the MiniKans were washed with CH2Cl2
and MeOH (three alternating wash cycles). After drying in vacuo, one of
the MiniKans, originally loaded with 60 mg of resin 6b, was cleaved with
10% TFA/CH2Cl2 (rt, 1 h), affording 5.7 mg of 13a (65% overall yield
from 6b). Purity of the final product was 89% at 254 nm, and 78% at 210
1
nm. H NMR (300 MHz, CDCl3) δ 4.28 (s, 2 H), 7.26 (d, J ) 8.1 Hz, 2
H), 7.31 (d, J ) 8.5 Hz, 2 H), 7.43 (d, J ) 8.3 Hz, 2 H), 7.95 (d, J ) 8.4
Hz, 2 H). 13C NMR (100 MHz, DMSO-d6) δ 28.87, 126.58, 127.94, 128.75,
129.28, 129.36, 129.54, 130.56, 131.27, 134.48, 134.77. MS (ES+) m/z
336 (100), 338 (72).
(12) (a) Backes, B. J.; Ellman, J. A. Curr. Opin. Chem. Biol. 1997, 1,
86-93. (b) For a recent review on solid-phase heterocycle syntheses, see:
Franzen, R. G. J. Combi. Chem. 2000, 2, 195-214.
(13) Mendonca, A. J. Am. Lab. 1998, 30, 46-47.
3344
Org. Lett., Vol. 3, No. 21, 2001