A "catch-react-release" method for the flow synthesis of 2-aminopyrimidines and preparation of the imatinib base

Article abstract of DOI:10.1021/ol301673q

The development of a monolith-supported synthetic procedure is reported, taking advantage of flow processing and the superior flow characteristics of monolithic reagents over gel-phase beads, to allow facile access to an important family of 2-aminopyrimidine derivatives. The process has been successfully applied to a key precursor on route to Imatinib (Ar = 3-pyridyl, R¹ = 2-methyl-5-nitrobenzyl, R² = H).

Full text of DOI:10.1021/ol301673q

ORGANIC
LETTERS
2012
Vol. 14, No. 15
3920–3923
A “CatchꢀReactꢀRelease” Method for
the Flow Synthesis of 2-Aminopyrimidines
and Preparation of the Imatinib Base
Richard J. Ingham, Elena Riva, Nikzad Nikbin, Ian R. Baxendale, and Steven V. Ley*
Innovative Technology Centre, Department of Chemistry, University of Cambridge,
Lensfield Road, Cambridge, CB2 1EW, U.K.
svl1000@cam.ac.uk
Received June 19, 2012
ABSTRACT
The development of a monolith-supported synthetic procedure is reported, taking advantage of flow processing and the superior flow
characteristics of monolithic reagents over gel-phase beads, to allow facile access to an important family of 2-aminopyrimidine derivatives. The
process has been successfully applied to a key precursor on route to Imatinib (Ar = 3-pyridyl, R1 = 2-methyl-5-nitrobenzyl, R2 = H).
Finding cleaner and more expedient methods to assem-
ble functional molecules is an important driver for organic
synthesis. Recently, advances in enabling tools such as
flow technologies have had a significant impact on the way
in which many of these chemicals are produced;1 this is
especially true in the delivery of heterocyclic building
blocks for medicinal and agrochemical discovery. Accord-
ingly, we and others have been exploring the benefits of
flow based synthesis to expand the diversity and facilitate
the scaleup of several relevant heterocyclic materials;2,3
most recently we have targeted the preparation of biologi-
cally important 2-aminopyrimidines.4 Herein we wish
to report an improved approach to these compounds,
which have traditionally been accessed via cross-coupling
methods,5a by halogen substitution5a,b or by displacement
of alkylthiols.5c,d This thiol displacement, which enables
late-stage diversification, is a particularly synthetically
powerful sequence owing to the ease of preparation of
the starting material.
(1) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan, A. R.;
McQuade, D. T. Chem. Rev. 2007, 107, 2300.
(2) (a) Herath, A.; Dahl, R.; Cosford, N. D. P. Org. Lett. 2010, 12,
412. (b) Opalka, S. M.; Longstreet, A. R.; McQuade, D. T. Beilstein J.
Org. Chem. 2011, 7, 1671. (c) Palde, P. B.; Jamison, T. F. Angew. Chem.,
Int. Ed. 2011, 50, 1.
(3) (a) Zak, J.; Ron, D.; Riva, E.; Harding, H. P.; Cross, B. C. S.;
Baxendale, I. R. Chem.;Eur. J. 2012, 18, early view, doi: 10.1002/
chem.201201039. (b) Baumann, M.; Baxendale, I. R.; Kuratli, C.; Ley,
S. V.; Martin, R. E.; Schneider, J. ACS Comb. Sci. 2011, 13, 405.
(c) Venturoni, F.; Nikbin, N.; Ley, S. V.; Baxendale, I. R. Org. Biomol.
Chem. 2010, 8, 1798.
(4) (a) Hopkin, M. D.; Baxendale, I. R.; Ley, S. V. Chem. Commun.
2010, 46, 2450. (b) Hopkin, M. D.; Baxendale, I. R.; Ley, S. V. Org.
Biomol. Chem. 2012, submitted.
(5) For recent examples, see: (a) El-Deeb, I. M.; Lee, S. H. Bioorg.
Med. Chem. 2010, 18, 3860. (b) Saczewski, J.; Paluchowska, A.; Klenc,
J.; Raux, E.; Barnes, S.; Sullivan, S.; Duzynska, B.; Bojarski, A. J.;
Strekowski, L. J. Heterocycl. Chem. 2009, 46, 1259. (c) Agarwal, A.;
Srivastava, K.; Puri, S. K.; Chauhan, P. M. S. Bioorg. Med. Chem. 2005,
13, 4645. (d) Almario-Garcia, A.; Ando, R.; Aritomo, K.; Frost, J. R.;
Li, A.-T.; Shoda, A.; Uehara, F.; Wanatabe, K. WO 01/70727 A1, 2001.
However, this process is under-represented in the litera-
ture due, at least in part, to the release of sulfur-containing
byproducts which are often malodorous or toxic. We con-
cluded that loading the initial thioether precursor onto a
solid support could circumvent many of these drawbacks. In
this way both the release of sulfurous byproducts during
substitution and the potential for the final product to
become contaminated with toxic thiourea byproduct are
minimized. Furthermore, flow based methods can be har-
nessed to perform reagent loading as well as reaction steps to
afford functionalized materials in an automated fashion.
Macroporous monolithic columns have been identified
as an effective format for solid-phase chemistry.6 The use
of this type of support for flow based catchꢀreactꢀrelease
ꢀ
(6) Peters, E. C.; Svec, F.; Frechet, J. M. J. Adv. Mater. 1999, 1, 1169.
r
10.1021/ol301673q
Published on Web 07/19/2012
2012 American Chemical Society

Scheme 1. Proposed Synthetic Route
protocols retains all of the benefits of bead-based chemistry
but offers a number of practical advantages. For example,
the internal structure of a monolith is found to have im-
proved flow characteristics when compared to traditional
polymer beads, allowing more efficient and higher-yielding
syntheses to be conducted.7 When the polymer is contained
within a column, the flow stream is forced to pass through the
monolith, whereas polymer beads are commonly observed to
suffer from solution channelling which reduces their effec-
tiveness. In addition, the rigid cross-linked morphology of
the monolith prevents significant volume changes due to
swelling or shrinking, allowing the use of a wider range of
solvents. Of particular importance from a synthetic point of
view is the possibility to rapidly generate customized func-
tionalized monoliths without recourse to the specialist equip-
ment that is required for the manufacture of bead-type
reagents, and thus they can be prepared at a much lower
cost. Indeed, monolithic reactors have already been prepared
and used to conduct a wide range of chemistries, having
demonstated particular value as reagent sources for flow
chemistry platforms.7ꢀ9
generate suitable monoliths of form M1, possessing high
porosity to reduce back-pressure and high stability to with-
stand prolonged processing under the flow conditions. The
optimized blend required a significant deviation from pre-
vious recipes,7,9 incorporating water and a low molecular
weight alcohol to accommodate the different solubility
characteristics of the hydrochloride monomer salt 3 and
the divinyl benzene cross-linker (Scheme 2). It should also
be noted that polymerization of this material to form gel-
phase beads would be difficult, since compound 3 is mainly
water-soluble and thus the usual biphasic solvent system
would be inappropriate.
Scheme 2. Preparation of the Functionalized Monolith M112
Aminopyrimidines related to 1 have been used previously
as targets for solid-supported synthesis10 due to their sig-
nificant value as building blocks. Here we propose a simple
sequence to prepare structures related to 1 (Scheme 1) which
we anticipated could be readily applied using solid-phase
monolithic reagents and flow chemistry techniques.11
Therefore, an investigation was first conducted to identify
a polymerization mixture that could be used to reproducibly
For the monolith preparation, the monomer, cross-
linker, and porogen mixture were homogenized under
gentle heating (<50 °C).12 The initiator was then added,
and the resulting mixture was transferred into glass
columns.13 The columns were sealed at both ends and
heated at 90 °C for 20 h in a multichannel convection
heater (Figure 1a). After cooling, the monoliths were
washed with ethanol at 60 °C to remove the porogen and
residual monomer, resulting in rigid white monoliths (M1)
that completely filled the glass columns (Figure 1b).
(7) Baumann, M.; Baxendale, I. R.; Ley, S. V.; Nikbin, N.; Smith,
C. D. Org. Biomol. Chem. 2008, 6, 1587.
The monoliths could be prepared in a range of sizes,
allowing for small-scale monolithic reactors (3 mm column
diameter) for use with precious materials, or parallel use of
multiple columns with 15 mm diameter, currently the largest
diameter at which the temperature gradient across the col-
umn allows effective polymerization.7 The monoliths used in
this work had an average dry weight of 1.6 g (column size
7 cm ꢁ10 mm diameter), and elemental analysis indicated a
loading of 2.4 mmol/g isothiouronium chloride which was
(8) (a) Burguete, M. I.; Garcıa-Verdugo, E.; Karbass, N.; Luis, S. V.;
Sans, V.; Sokolova, M. Pure Appl. Chem. 2009, 81, 1991. (b) Burguete,
M. I.; Cornejo, A.; Garcıa-Verdugo, E.; Gil, M. J.; Luis, S. V.; Mayoral,
J. A.; Martınez-Merino, V.; Sokolova, M. J. Org. Chem. 2007, 72, 4344.
(c) Mennecke, K.; Kirschning, A. Beilstein J. Org. Chem. 2009, 5, 21.
(d) Mennecke, K.; Cecilia, R.; Glasnov, T. N.; Gruhl, S.; Vogt, C.;
Feldhoff, A.; Vargas, M. A. L.; Kappe, C. O.; Kunz, U.; Kirschning, A.
Adv. Synth. Catal. 2008, 350, 717.
(9) (a) Smith, C. J.; Smith, C. D.; Nikbin, N.; Ley, S. V.; Baxendale,
I. R. Org. Biomol. Chem. 2011, 9, 1927. (b) Lange, H.; Capener, M.;
Jones, A.; Smith, C.; Nikbin, N.; Baxendale, I.; Ley, S. Synlett 2011,
2011, 869. (c) Nikbin, N.; Ladlow, M.; Ley, S. V. Org. Process Res. Dev.
2007, 11, 458. (d) Roper, K. A.; Lange, H.; Polyzos, A.; Berry, M. B.;
Baxendale, I. R.; Ley, S. V. Beilstein J. Org. Chem. 2011, 7, 1648.
(10) Masquelin, T.; Sprenger, D.; Baer, R.; Gerber, F.; Mercadal, Y.
Helv. Chim. Acta 1998, 81, 646.
(11) All flow reactions and monolith preparations described below
were carried out using Vapourtec R2þ and R4 units, available from
Vapourtec Ltd. Website: http://www.vapourtec.co.uk. Multiple steps
can be telescoped by sending a programmed sequence of commands to
the serial interface of the flow apparatus. For the first use of the
Vapourtec system within our group, see: Baumann, M.; Baxendale,
I. R.; Ley, S. V.; Smith, C. D.; Tranmer, G. K. Org. Lett. 2006, 8, 5231.
(12) The optimized polymerization mixture developed for this work
consisted of the following: 4-(vinylbenzyl)isothiuronium chloride (3;
18.8% w/w); divinyl benzene (DVB cross-linker; 12.5% w/w); 1-propa-
nol (porogen; 57.5% w/w); water (porogen; 10.9% w/w); 1,10-azobis-
(cyclohexane carbonitrile) (ACHC initiator; 1% w/w relative to 3 þ
DVB).
(13) Commercially available Omnifit glass chromatography columns
with fixed end pieces, with 10 mm bore and 100 mm length. Website:
http://www.omnifit.com.
Org. Lett., Vol. 14, No. 15, 2012
3921

characteristic changes in the IR spectra, notably new absorp-
tions in the regions of 1250 and 1400 cmꢀ1
.
The thioether unit of M2 was subsequently activated by
oxidation prior to its displacement with an amine nucleo-
phile (Scheme 4). m-CPBA was chosen as the oxidant, due
to its high solubility and existing precedent for use in this
type of process.10
Scheme 4. Oxidative Activation of Monolith M2
Figure 1. (a) A batch of monoliths undergoing the curing process
in a convection heater. (b) Prepared monolithic reagents, re-
moved from and within the glass column reactor.
highly reproducible over multiple monoliths. The monoliths
were typically prepared in batches of four or more and could
be stored as sealed columns for periods of several weeks at
room temperature without any noticeable degradation.
The monoliths were then reacted with solutions of a
An excess of m-CPBA was used for this transformation,
to encourage complete oxidation of all of the sulfur atoms
on the monolith. Despite this, elemental analysis suggested
that statistically there was a mixture of sulfone and some
sulfoxide present; however, both were expected to undergo
the desireddisplacement reaction. Consequently, release of
the heterocycles was achieved by substitution of M3 using
various amine nucleophiles (Scheme 5).
substituted enaminone 214 in the presence of Hunig’s base
€
to construct the pyrimidine core (M2, Scheme 3). Taking
advantage of the solid-supported regime, more than one
stoichiometric equivalent of the enaminone could be used
in order to drive the reaction to completion, without
sacrificing the purity of the immobilized intermediate.
The surplus enaminone could be readily recovered from
the output stream and recycled.
Scheme 5. Amine Displacement of 2-Aminopyrimidine 1
Scheme 3. Cyclization with Enaminones 2 (the “Catch” Step)a
As the SNAr reaction was found to be slow for certain
reagent combinations, the yield was generally improved by
increasing the quantity of amine used, and employing a
stopꢀflow reaction technique. Under computer control,
the reagents were added over several injections. The two
main input streams were pumped at 0.1 mL minꢀ1 for
50 min, which was sufficient time to fill the monolith (pore
volume ∼50% of the column volume). The pumps were
then stopped, and the reagent mixture was held within
the monolith for 1 ꢀ 2 h at an elevated temperature to
facilitate the reaction (90 °C, or at higher temperature with
aniline nucleophiles). The pyrimidines 1 were then eluted as
a mixture with the excess amine and could be isolated by
aqueous extraction and flash column chromatography.
As with any catch and release procedure, this results in
a segmented flow process. Nevertheless, it has many
of the advantages of a continuous flow procedure, such
as easy automation and simplified purification of
products.
a The two reagents were pumped continuously in separate streams
from stock solutions, mixing in a T-piece immediately before entering
the monolith to minimize undesired reactions. The combined flow rate
gave a residence time of approximately 15 min.
Following a washing regime of the monolith to remove
any unbound reactants, elemental analysis of the polymer
postreaction indicated a loading of 2.0 mmol/g, based on N
analysis. Further evidence for the formation of the required
pyrimidine M2 was provided by the disappearance of the 13C
resonance of the thiouronium center at 171 ppm detected
using Magic Angle Spinning NMR spectroscopy, as well as
(14) Abdulla, R. F.; Fuhr, K. H. J. Org. Chem. 1978, 43, 4248–4250.
3922
Org. Lett., Vol. 14, No. 15, 2012

Scheme 6. Access to the Imatinib Base 515,16
Table 1. Collection of 2-Aminopyrimidine Derivatives
when more nucleophilic amines were employed in the
displacement.
As a further benchmark for proving the applicability of
this process, we wished to apply this flow sequence to the
preparation of a known pharmaceutical target. We chose
toprepare pyrimidine 4, a known precursor inthe synthesis
of the tyrosine kinase inhibitor Imatinib 5 (Scheme 6).15
Following the previouslydescribedroute, compound4 was
obtained in 48% overall yield (Table 1, entry 6). Advanta-
geously this method circumvents problems encountered
with the low solubility of the intermediates, which have
hampered previous flow chemistry approaches.4
In conclusion, using a new supported thiouronium salt
M1, we were able to rapidly generate 10 substituted
aminopyrimidines 1 from readily accessible enaminones
and amines. The supported variant of this reagent offers
many benefits over solution-phase synthesis, including
simplified isolation of the product, the potential for auto-
mation, and containment of toxic and malodorous byprod-
ucts. The new monolithic reagent offers benefits over
traditional polymer bead reagents, in particular, a more
suitable morphology for effective use in a continuous flow
regime and a low-cost and accessible preparation procedure.
Only one purification operation is required over three steps,
and the compounds are retrieved in comparable yields to
batch routes to similar targets. The procedure was also
attractively deployed in the preparation of the key amino-
pyrimidine unit 4 found in the Imatinib base (Gleevec).
Acknowledgment. We gratefully acknowledge funding
from the EPSRC (R.J.I. and N.N.), Novartis (R.J.I.), the
BP 1702 endowment (S.V.L.), and the Royal Society (I.R.
B.) and Dr. David Reid (Department of Chemistry, Uni-
versity of Cambridge) for performing MAS NMR
experiments.
a Isolated yield over all three steps, after purification by flash column
chromatography, based on an assumed 4.2 mmol loading of monolith M1.
To explore the scope and utility of this process as a
general method to access 2-aminopyrimidines, we used a
number of aromatic enaminones as starting materials
in conjuction with several aromatic and aliphatic nitro-
gen nucleophiles to effect the final release (Table 1).
We were pleased to obtain a collection of products in
moderate to good yields over the three contiguous steps. In
general, higher recovery of the final product was achieved
Supporting Information Available. Experimental pro-
cedures and spectral data. This material is available free
of charge via the Internet at http://pubs.acs.org.
(16) Kalesh, K. A.; Sim, D. S. B.; Wang, J.; Liu, K.; Lin, Q.; Yao,
S. Q. Chem. Commun. 2010, 46, 1118.
(15) Liu, Y.-F.; Wang, C.-L.; Bai, Y.-J.; Han, N.; Jiao, J.-P.; Qi,
X.-L. Org. Process Res. Dev. 2008, 12, 490.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 15, 2012
3923