Facile synthesis of N-6 adenosine modified analogue toward S-adenosyl methionine derived probe for protein arginine methyltransferases

Article abstract of DOI:10.1016/j.cclet.2011.09.007

Chemically modified cellular co-factors that provide function, such as immobilization or incorporation of fluorescent dyes, are valuable probes of biological activity. A convenient route to obtain S-adenosyl methionine (AdoMet) analogues modified at N-6 adenosine to feature a linker terminating in azide functionality is described herein. Subsequent decoration of such AdoMet analogues with guanidinium terminated linkers leads to novel potential bisubstrate inhibitors for protein arginine methyltransferases, PRMTs.

Full text of DOI:10.1016/j.cclet.2011.09.007

Available online at www.sciencedirect.com
Chinese Chemical Letters 22 (2011) 1439–1442
www.elsevier.com/locate/cclet
Facile synthesis of N-6 adenosine modified analogue toward
S-adenosyl methionine derived probe for protein arginine
methyltransferases
Wei Hong a, James Dowden b,
*
a School of Chemistry and Chemical Engineering, Beifang University of Nationalities, Yinchuan 750021, China
b School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK
Received 24 May 2011
Available online 10 October 2011
Abstract
Chemically modified cellular co-factors that provide function, such as immobilization or incorporation of fluorescent dyes, are
valuable probes of biological activity. A convenient route to obtain S-adenosyl methionine (AdoMet) analogues modified at N-6
adenosine to feature a linker terminating in azide functionality is described herein. Subsequent decoration of such AdoMet
analogues with guanidinium terminated linkers leads to novel potential bisubstrate inhibitors for protein arginine methyltrans-
ferases, PRMTs.
# 2011 James Dowden. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
Keywords: Protein arginine methyltransferases; S-Adenosyl methionine; Nucleosides
Protein arginine methyltransferases (PRMTs, EC 2.1.1.125) utilize S-adenosyl methionine (AdoMet, Fig. 1) to
catalyze post-translational methylation of arginine residues of a variety of substrate proteins, leading to S-adenosyl
homocysteine (AdoHcy, Fig. 1) and monomethylarginine (mMA). Further methylation often follows, with Type I
PRMTs deliveringasymmetrical dimethylarginine (aDMA) andTypeII PRMTssymmetricaldimethylarginine (sDMA),
respectively [1]. At least nine PRMTs have been identified in mammals and an important roles in governing cellular
processes, such as including gene regulation, RNA splicing and transport, or signal transduction are emerging [2]. There
is a significant interest in developing potential drugs for individual PRMTs, especially for the treatment of hormone
dependent cancer [3]. Functional chemicalprobes maybevaluable for exploringinhibition, andpotential immobilization
of specific PRMTs.
So far, we have reported AdoMet analogue 1 (Fig. 1) as a potent inhibitor of PRMT1 (IC50 ꢀ 6 mmol/L), that
features structure intended to target both AdoMet and arginine subsites of the enzyme [4]. Modification of AdoMet
analogue 1 to facilitate labeling or immobilization of this enzyme for functional studies would be highly desirable. We
therefore designed compound 2 (Fig. 1) to featuring N-6 modification of adenosine to incorporate azide functionality
via a long tether. We speculated that the flexible linker might occupy a channel pointing away from the adenosine
* Corresponding author.
E-mail address: James.Dowden@nottingham.ac.uk (J. Dowden).
1001-8417/$ – see front matter # 2011 James Dowden. Published by Elsevier B.V. on behalf of Chinese Chemical Society. All rights reserved.
doi:10.1016/j.cclet.2011.09.007

1440
W. Hong, J. Dowden / Chinese Chemical Letters 22 (2011) 1439–1442
H
N
N
N
NH
2
NH
NH
2
2
R
O
O
HO
N
HO
N
N
N
X
R
N
N
N
O
O
OH
OH
HO
HO
AdoMet - X = S, R = CH
AdoHcy - X = S, R = H
AzaAdoMet - X = N, R = CH
3
HN
H N
1 - R = H
2 - R =
N
3
3
NH
3
Cl
2
2
O
Fig. 1. Structure of AdoMet and related compounds.
binding site, thus presenting the reactive azide functionality in a remote position available for modification using bio-
orthogonal reactions [5]. Herein, we report convenient synthesis of this N-6 modified AdoMet analogue 2.
AzaAdoMet (Fig. 1) has been reported as a potent but general methyltransferase inhibitor that has generally been
synthesized by various alkylation strategies in moderate overall yields [6]. We have reported that sequential reductive
amination focussed on 50-amino-50-deoxyadenosine [7] offered a flexible alternative route to deliver the target tertiary
amines [4]. Our initial task was to synthesize an equivalent N-6 modified adenosine 9 (Scheme 1) and investigate its
incorporation into this route toward PRMT inhibitors.
Azide-tethered amine 5 [8] was incorporated at the N-6 position of adenosine was attempted using two approaches
shown in Scheme 1. Vorbru¨ggen glycosylation of 6-chloropurine with ribofuranose tetraacetate 3 was achieved using
tin tetrachloride to give the protected 6-chloride adenosine analogue 4 in 78% yield [9]. The purinyl chloride was then
substituted by the azide-tethered amine 5 in the presence of triethylamine to form the desired compound 6 in a
disappointing yield of 43%. Finally, acetate protecting groups were cleaved using sodium methoxide and the crude
product directly reacted with 2, 2-dimethoxylpropane in the presence of toluenesulfonic acid and acetone to give the
N-6 modified adenosine 9 in 51% yield. Superior overall yields could be achieved by subjecting inosine 7 to familiar
conditions to deliver acetonide-protected inosine 8 in 63% yield [10]. The azide-tethered amine 5 was then installed by
in situ activation of using benzotriazole-1-yl-oxytripyrrolidino-phosphonium hexafluorophosphate (PyBOP) in the
presence of N, N0-diisopropylethylamine (DIPEA) and subsequent displacement with amine 5 to furnish the compound
9 in 75% yield [11]. This two-step process was clearly more efficient and could be easily scaled up.
Cl
N
R
N
N
N
N
N
OAc
O
O
O
a
AcO
b
AcO
AcO
N
N
H N
N
3
3
2
AcO
OAc
O
5
AcO
OAc
AcO
OAc
6
4
3
c
R
O
N
N
O
N
N
N
O
O
N
N
O
NH
NH
e
d
HO
R=
HO
HO
N
N
N
H N
N
3
3
2
O
O
O
O
O
OH
HO
5
7
8
9
HN
N
3
3
O
Scheme 1. Reagent and conditions: (a) SnCl4, 6-chloropurine, CH3CN, rt, 78%; (b) 5, Et3N, dry DMF, rt to 60 8C, 43%; (c) NaOCH3, dry MeOH, rt,
and then TsOH, 2, 2-dimethoxypropane, acetone, rt, 51%; (d) TsOH, 2, 2-dimethoxylpropane, acetone, rt, 63%; (e) 5, PyBOP, DIPEA, dry DMF, rt,
75%.

W. Hong, J. Dowden / Chinese Chemical Letters 22 (2011) 1439–1442
1441
R
N
R
N
R
N
N
N
N
O
N
N
N
b
a
O
O
O
HO
N
H N
2
N
N
N
O
O
O
N
O
O
O
O
9
10
11
R
N
N
NHBoc
c
d
O
t-BuO
N
N
H
NHBoc
O
O
BocHN
N
t-BuO
O
O
O
NHBoc
14
12
O
13
R
N
N
N
R
N
NH
NHBoc
2
O
N
HO
O
t-BuO
N
N
N
N
e
N
O
O
OH
HO
2
O
O
HN
H N
N
15
HN
N
3
3
R=
NH
O
2 Cl
BocHN
NHBoc
2
Scheme 2. Reagents and conditions: (a) phthalimide, DIAD, Ph3P, dry THF, rt, 76%; (b) hydrazine hydrate, ethanol, reflux, 82%; (c) 12,
NaBH(OAc)3, DCE, rt, 75%; (d) 14, NaBH(OAc)3, DCE, rt, 99%; (e) 98% TFA, rt, 87%.
We next investigated incorporation of the N-6 modified adenosine analogue 9 into our route to deliver putative
PRMT inhibitors via overall Gabriel synthesis and sequential reductive aminations and deprotection (Scheme 2).
Compound 9 was first converted to the phthalimide 10 using standard Mitsunobu conditions in 76% yield and then
deprotected by refluxing with hydrazine monohydrate in ethanol to deliver the azide-tethered 50-amino-50-deoxy-
adenosine 11 in 82% yield [7]. Condensation of compound 11 with aldehyde 12 (derived from aspartic acid) [12] by
reductive amination with sodium triacetoxyborohydride delivered the amine analogue of AdoHcy 13, a flexible
precursor to a range of potential AdoMet analogues, in 75% yield. Further reductive amination with guanidine bearing
aldehyde 14, gave precursor 15 in good yield (99%), which was purified by silica column chromatography. Finally,
deprotection using aqueous TFA solution removed all protecting groups and the resulting trifluoroacetate salt subject
to anion exchange chromatography using Amberlyst IRA-400 (Clꢁ) to deliver the target N-6 modified AdoMet
analogue 2 as its hydrochloride salts as hygroscopic solid [13].
The in vitro bio-evaluation of N-6 modified AdoMet analogue 2 is in progress. Preliminary evaluation was carried
using human recombinant PRMT1, [S-3H3C]-SAM and substrate recombinant Sam68, and the methyl transfer was
measured by scintillation counting [14]. The preliminary result showed that compound 2 has no significant inhibition
for PRMT1, but compound 1 can inhibit PRMT1 with IC50 ꢀ 6 mmol/L [4]. Through comparing the conformations of
compounds 1 and 2, it was presumed that there is no enough space on PRMT1 binding pocket for the long azide tether
on N-6 position of compound 2, which makes it difficult to bind at there. The further bio-evaluation will be carried out
and molecular modeling will be used to analyze the results.
In summary, we describe facile synthesis of an azide-tethered 50-amino-50-deoxy adenosine derivative 11, a flexible
precursor to functional probes of a range of AdoMet analogues. Decoration of this compound with guanidine
functionally via a propyl linker gave an analogue of our previously described PRMT1 inhibitor.
Acknowledgments
This work was financially supported by the Medical Research Council Grant (No. G0700840) and the University of
Nottingham (Studentship to Wei Hong).

1442
W. Hong, J. Dowden / Chinese Chemical Letters 22 (2011) 1439–1442
References
[1] M.T. Bedford, S. Richard, Mol. Cell 18 (2005) 263.
[2] M.T. Bedford, J. Cell Sci. 120 (2007) 4243.
[3] R.A. Copeland, M.E. Solomon, V.M. Richon, Nat. Rev. Drug Discov. 8 (2009) 724.
[4] J. Dowden, W. Hong, R.V. Parry, et al. Bioorg. Med. Chem. Lett. 20 (2010) 2103.
[5] N.J. Agard, J.M. Baskin, J.A. Prescher, et al. ACS Chem. Biol. 1 (2006) 644.
[6] M.J. Thompson, A. Mekhalfia, D.P. Hornby, et al. J. Org. Chem. 64 (1999) 7467.
[7] A.M. Reeve, C.A. Townsend, Tetrahedron 54 (1998) 15959.
[8] S.S. Iyer, A.S. Anderson, S. Reed, et al. Tetrahedron Lett. 45 (2004) 4285.
[9] M. Hocek, P. Silhar, I.H. Shih, et al. Bioorg. Med. Chem. Lett. 16 (2006) 5290.
[10] R.J. Middleton, S.J. Briddon, Y. Cordeaux, et al. J. Med. Chem. 50 (2007) 782.
[11] Z.K. Wan, E. Binnun, D.P. Wilson, et al. Org. Lett. 7 (2005) 5877.
[12] R.M. Werner, O. Shokek, J.T. Davis, J. Org. Chem. 62 (1997) 8243.
23
[13] Spectral data of 2: slightly yellow hygroscopic solid, ½aꢂD +20.59 (c 1.02, MeOH); y (max/cmꢁ1) 2108 (N3), 1737 (C O), 1670 (C C,
aromatic); 1H NMR (400 MHz, D2O): d 8.35 (s, 2H, ArH), 6.07 (d, 1H, J = 3.2, 10-H), 4.71 (m, 1H, 20-H), 4.39 (m, 2H, 30-H and 40-H), 4.02 (m,
1H, CHCH2), 3.75 (m, 2H, 50-CH2), 3.72 (m, 2H, adenosine-NHCH2), 3.65–3.37 (m, 14H, OCH2 ꢃ 6 and CH2N3), 3.33 (m, 2H, CH2NH), 3.27
(m, 2H, CHCH2CH2N), 3.12 (m, 2H, CH2CH2CH2N), 2.33 (m, 1H, CHCHaHb), 2.21 (m, 1H, CHCHaHb), 1.89 (m, 2H, CH2CH2CH2); 13
C
NMR (100 MHz, D2O): d 170.6 (C), 156.6 (C), 149.0 (C), 146.7 (C), 144.2 (CH), 143.1, (CH), 119.7 (C), 90.1 (CH), 78.0 (CH), 73.0 (CH), 71.5
(CH), 70.4–68.2 (CH2 ꢃ 5), 65.1 (CH2), 54.9 (CH2 ꢃ 2), 51.0 (CH2), 50.4 (CH), 50.1 (CH2), 42.4 (CH2), 38.0 (CH2), 24.3 (CH2), 22.6 (CH2);
ESI-MS 668.4 (M+H+); HRMS calcd. for C26H46N13O8 668.3587, found 668.3611.
+
[14] D.H. Cheng, N. Yadav, R.W. King, et al. J. Biochem. 279 (2004) 23892.