Highly efficient and selective phosphorylation of amino acid derivatives and polyols catalysed by 2-aryl-4-(dimethylamino)pyridine-N-oxides-towards kinase-like reactivity

Article abstract of DOI:10.1039/c4cc05388e

The chemoselective phosphorylation of hydroxyl containing amino acid derivatives and polyols by phosphoryl chlorides catalyzed by 2-aryl-4-(dimethylamino)pyridine-N-oxides is described.

Full text of DOI:10.1039/c4cc05388e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Highly efficient and selective phosphorylation of
amino acid derivatives and polyols catalysed by
2-aryl-4-(dimethylamino)pyridine-N-oxides – towards
kinase-like reactivity†
Cite this: Chem. Commun., 2014,
50, 13608
Received 12th July 2014,
Accepted 16th September 2014
James I. Murray,a Rudiger Woscholskiab and Alan C. Spivey*a
DOI: 10.1039/c4cc05388e
www.rsc.org/chemcomm
The chemoselective phosphorylation of hydroxyl containing amino role: mediating activation of the 30-phosphate by an aryl sulfonyl
acid derivatives and polyols by phosphoryl chlorides catalyzed by chloride ‘condensing agent’ and then mediating substitution
2-aryl-4-(dimethylamino)pyridine-N-oxides is described.
of this intermediate by the 50-OH group. Efimov et al.25,26 have
demonstrated that pyridine-N-oxides with electron-donating
Signal transduction cascades in living systems are often controlled 4-substituents provide significant rate enhancements in these
via the post-translational phosphorylation/dephosphorylation of reactions, with 4-dimethylaminopyridine-N-oxide (4-DMAP-N-
proteins and other secondary metabolites.1 In vivo, phosphorylation oxide, catalyst 2d, below) proving optimal.
is catalyzed by protein kinases, which selectively phosphorylate
We envisioned that pyridine-N-oxides would also act as
serine (Ser), threonine (Thr) and tyrosine (Tyr) residues and hence efficient nucleophilic catalysts for the phosphorylation of
play an important role in many disease states, including cancer and OH-containing substrates by phosphoryl chlorides and that
immune system disorders.2 Despite significant interest in drugs by positioning a sterically/electronically tuneable aryl group at
which act on protein kinases,3–7 there are currently no catalytic the 2-position it might be possible to modulate the reactivity
synthetic methods for the site-selective phosphorylation of and selectivity of these catalysts. It was also our expectation that
hydroxyl-containing amino acid residues.2,8
the low basicity of pyridine-N-oxides relative to N-based nucleo-
Miller et al. have reported the use of N-methyl histidine- philic catalysts (e.g. pyridine, 4-DMAP, imidazoles, amidines
containing pentapeptide catalysts for asymmetric phosphorylation etc.) would allow for the phosphorylation of alcohol substrates
of inositol derivatives by phosphoryl chlorides,9–17 e.g. in the which are labile towards subsequent base catalysed elimination
synthesis of D-myo-inositol-1/3-phosphate18 and for site-selective (e.g. Ser - dehydro-Ala).
phosphorylation of teicoplanin.19 They have also reported some
analogous processes via P(III) phosphoramidite transfer catalysed chloride catalysed by simple pyridine-N-oxides 2a–e and by
by tetrazole-containing peptides followed by oxidation.20,21
2-aryl pyridine-N-oxides 2f–l was examined using propylene
Ti(OtBu)4 and Cu(OTf2) catalyzed phosphorylation of alcohols by oxide (PPO) as a non-basic proton scavenger (Table 1).
both phosphoryl chlorides22 and N-phosphoryl oxazolidinones,23
Interestingly, pyridine-N-oxide (2a) was a superior catalyst both
Phosphorylation of Ser derivative 1 by diphenylphosphoryl
has been reported by Jones et al., and Sculimbrene et al. have used to 4-DMAP-N-oxide (2d)‡ and the other non-arylated catalysts tested
Ti(OtBu)4 with benzyl pyrophosphate to phosphorylate Ser and Tyr (2b, 2c and 2e). By contrast, 2-aryl-4-DMAP-N-oxides (e.g. 2g and 2k)
derivatives.24 Whilst these methods are moderate to high yielding, tended to be more active than their 2-aryl pyridine-N-oxide analo-
they require the use of metals and often long reaction times and gues (e.g. 2f and 2j) and moreover, although electronically neutral/
have not been used for site-selective phosphorylation of polyols.
rich 2-aryl substituents attenuated catalytic activity (e.g. 2g and 2i),
Pyridine-N-oxides promote coupling of 30-phosphorylated electron-deficient 2-aryl substituents enhanced catalytic activity
nucleotides with 50-OH nucleosides for the preparation of (e.g. 2k and 2l) relative to 4-DMAP-N-oxide itself and to a level
oligonucleotides.25,26 The N-oxides perform a dual catalytic higher than that of pyridine-N-oxide (2a). In particular, 2-[2,4-bis-
(trifluoromethyl)phenyl]-substituted 4-DMAP-N-oxide (2l, 5 mol%)
was a very efficient catalyst giving 97% conversion to phosphoryl-
ated product 3 within 8 h (entry 13). We propose that the high
catalytic activity of this derivative may result from synergistic
enhancement of the nucleophilicity of the N-oxide oxygen by the
a Department of Chemistry, South Kensington Campus, Imperial College London,
SW7 2AZ, UK. E-mail: a.c.spivey@imperial.ac.uk
b Institute of Chemical Biology, Imperial College London, London SW7 2AZ, UK
† Electronic supplementary information (ESI) available: Experimental procedures
4-amino substituent and the enhancement of the nucleofugacity
of the N-oxide by the 2-aryl substituent. These effects would be
and compound characterization data for all new compounds. See DOI: 10.1039/
c4cc05388e
13608 | Chem. Commun., 2014, 50, 13608--13611
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Table 1 Phosphorylation of Ser 1: catalyst evaluation
Table 2 Phosphorylation of Ser 1 (& a-MeSer 5): influence of base
Entry
Catalyst
R
X
Conv.a (%)
Base/H+
scavenger
Yield of
Yield
1
2
3
4
5
6
7
8
None
2a
2b
2c
2d
2e
2f
2g
2h
2i
—
H
H
Cl
H
Cl
—
H
Cl
H
NMe2
NMe2
H
NMe2
H
NMe2
H
NMe2
NMe2
0
Entry
Substrate
Cat.
3 (or 7)/%a
of 4/%a
82 (79)b
50
0
72
78
0
27
47
28
1
2b
3
4
5
6
7
8
1 (5)
1 (5)
1 (5)
1 (5)
1 (5)
1 (5)
1 (5)
1 (5)
2l
2l
6
2l
2l
2l
6
None
PPO
PPO
PSr
PMP
NEt3
NEt3
DBU
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
54 (0)
72 (0)
83 (0)
53c (60)
18 (16)
86c (87)
98c (95)
46 (96)
14 (55)
9 (92)
4-MeOC6H4
4-MeOC6H4
4-MeC6H4
4-MeC6H4
3,5-(CF3)2C6H3
3,5-(CF3)2C6H3
2,4-(CF3)2C6H3
9
10
11
12
13
2l
2j
2k
2l
11
86 (81)b
97 (95)b
a
b
Isolated yield after chromatographic purification. Analogous catalysis
(using 1) was also successful in DMF, MeCN and DMSO, see ESI.
498% ee by CSP-HPLC, see ESI. PSr = 1,8-bis(NMe2)-naphthalene,
c
a
1
Conversion to product 3 as determined by H NMR of crude reaction
DBU
= 1,8-diazabicyclo[5.4.0]undec-7-ene, PMP = 1,2,2,6,6-penta-
methylpiperidine, Hyp = (2S,4R)-4-hydroxyproline, Sp5 = 1-amino-1-
b
mixture. Isolated yield after chromatographic purification. PPO =
propylene oxide.
cyclopentane-carboxylic acid.
expected to lower the kinetic barriers both for initial nucleophilic
addition of the N-oxide to the phosphoryl chloride and for Et3N, where imidazole protonation should not be significant,
subsequent substitution at phosphorus by the alcohol to regenerate the intrinsic reactivity is lower than that of catalyst 2l (55% vs.
the N-oxide, respectively during the catalytic cycle.8
96% conversion after 2 h, entries 7 and 6).
Using optimal catalyst 2l, a selection of bases were evaluated
Phosphorylations of Thr 8 and Tyr 10 were investigated next
in place of the PPO for the transformation 1 - 3, but with the with the aim of identifying catalyst/base combinations for
reaction duration restricted to just 2 h (cf. 8 h in Table 1) to selective phosphorylation of these sec-alcohol and phenol-
explore whether an increased rate of phosphorylation could be containing substrates respectively (Table 3).
achieved without promoting subsequent elimination. Analogous
Despite the greater steric demand of sec-alcohol 8 relative to
reactions with a-MeSer 5 were also performed in order to reveal primary alcohol 1, hindered N-oxide 2l again proved to be the
the rates of phosphorylation in the absence of the possibility of optimal catalyst in terms of rate when using PPO as proton
elimination (Table 2).
scavenger (entries 1–4). Interestingly, PMP retarded this reaction
Proton Sponge (PSr) and particularly PMP proved to be (entry 5) whereas PSr provided a significant rate increase,
more effective than PPO in promoting rapid phosphorylation of allowing essentially quantitative phosphorylation within 24 h
Ser derivative 1 without detectable elimination or racemisation (entry 6). For phenol 10, attempted phosphorylation with catalyst
(entries 4 and 5 cf. entry 2). By contrast, NEt3 and particularly
DBU promoted subsequent elimination to dehydro-Ala 4,
Table 3 Phosphorylation of Thr 8 & Tyr 10: influence of catalyst and base
although in the absence of the possibility of elimination these
bases provide comparable phosphorylation rates to PSr and
PMP as evidenced by their efficient promotion of the
phosphorylation of a-MeSer 5 (entries 6 and 8). There does
not appear to be a straightforward correlation between the
Base/H+
scavenger
Entry
Substrate
Cat.
Time (h)
Conv.a (%)
pKa’s of the conjugate acids of the tert-amine auxiliary bases
evaluated here and their effectiveness as co-promoters in these
reactions. This probably reflects the role of the amine in aiding
proton transfer as well as scavenging HCl and promoting
elimination.
1
2
3
4
5
6
7
8
9
10
8
8
8
8
8
8
10
10
10
10
2a
2d
2k
2l
2l
2l
2l
2l
2l
2l
24
24
24
24
24
24
2
2
2
2
PPO
PPO
PPO
PPO
PMP
PSr
PPO
PMP
PSr
NEt3
47
17
38
54
9
Miller’s N-methyl imidazole catalyst 610 provided low levels
of conversion relative to pyridine-N-oxide 2l for phosphoryla-
tion of both Ser 1 (with PPO) and a-MeSer 5 (with PPO and
NEt3) as well as promoting more extensive elimination of Ser 1
to dehydroalanine 4 (with NEt3) (entries 3 and 7). The low
reactivity of catalyst 6 when combined with PPO probably
reflects protonation of the N-methyl imidazole under these
499 (96b)
0 (0c)
499d (79c)
59 (21c)
499d (95b)
Conversion to products 9 or 11 as determined by 1H NMR of crude
reaction mixture. Isolated yield after chromatographic purification.
Conversion to products 9 or 11 of uncatalysed reaction. Reaction
a
b
c
d
conditions (pKa B 7, cf. pyridine-N-oxide pKa B 1) but with complete within 1 h.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13608--13611 | 13609

View Article Online
ChemComm
Communication
2l and PPO resulted in no reaction (entry 7), but both PMP and
NEt3 allowed essentially quantitative phosphorylation within 1 h
(entries 8 and 10).
As the distinctive reactivity profiles displayed by alcohols 1,
8 and 10 augured well for achieving site-selective phosphoryl-
ation reactions, triol 1227 was subjected to reaction conditions
expected to show selectivity for phosphorylation of the primary
and phenolic OH groups (Table 4). o-Xylenyl phosphoryl chloride
(o-XPCl, 14)§ was employed for the phenol selective reaction as
selective deprotection can be accomplished either by hydro-
genolysis¶ (H2/Pd–C)31 or with acid (e.g. HBr/AcOH).30
Scheme 1 Deprotection of xylenyl phosphate 13b.
Pleasingly, high levels of selectivity for mono-phosphorylation
of the primary OH group (12 - 13a) were achieved by using PPO
with comparable yields for diphenyl phosphoryl chloride and
o-XPCl over an 8 h reaction duration (94% and 87% yields
respectively, entry 1). Good levels of selectivity for mono-
Scheme 2 Selective phosphorylation of chloramphenicolr (16).
Using catalyst 2l and either diphenyl phosphoryl chloride
phosphorylation of the phenol group (12 - 13b) were achieved or o-XPCl (14) in conjunction with PMP, selective mono-
by using PMP over a 1 h reaction duration (81% yield, entry 2), phosphorylation of the primary alcohol group could be
although some of the primary alcohol mono-phosphate 13a was achieved to give products 17 and 18 in 91% and 96% yields
also isolated under these conditions (11% yield). Diphosphorylation respectively. Control reactions using catalysts 2a and 2d
of the primary and phenolic OH groups (12 - 13c) was achieved provided phosphorylated derivative 17 in yields of just 57%
through use of PMP (2.2 eq.) over a 4 h reaction duration (86% yield, and 45% respectively, underscoring the superiority of catalyst
entry 3). No conditions were identified that could selectively mono- 2l. Deprotection of the o-xylenyl phosphate 18, in the presence
phosphorylate the sec-alcohol, but ‘global’ phosphorylation of of the labile gem-dichloro and nitro functions, using HBr in
all OH groups (12 - 13d) was achieved through the use of PSr AcOH/MeOH gave free phosphate 19 quantitatively.
(3.2 eq.) over a 24 h reaction duration (96% yield, entry 4).
Finally, we sought to demonstrate the potential of 2-aryl-4-
The mono-phosphorylated products 13a [P = P(O)(OPh)2 DMAP-N-oxide 2l as a synthetic Tyr kinase mimetic. Tyr kinases
and o-XP] and 13b [P = P(O)(OPh)2] were also individually constitute the largest class of mammalian kinases32 and are
resubjected to the reaction conditions with PPO, PMP and involved in control of cellular processes such as cell metabolism,
PSr for 24 h in the absence of phosphorylating agent; no growth, mobility, survival, proliferation and differentiation.33 The
changes could be detected by 31P NMR in any of these control phosphorylation of Tyr selectively in the presence of Ser and Thr
reactions, confirming that phosphoryl migration does not residues in peptides/proteins therefore represents an appealing
occur under these conditions.
challenge. We used the all-(S) configured synthetic peptide Ac-Ala-
Deprotection of o-xylenyl phosphate 13b was achieved by Tyr-Ala-Ser-Ala-Thr-Ala-OMe (20) as a model substrate (Scheme 3).
hydrogenolysis (H2/Pd–C) in MeOH to give free phosphate 15 in
almost quantitative yield (Scheme 1).
Following treatment of peptide 20 with o-XPCl (14), PMP and
catalyst 21 in CH2Cl2 for 24 h, LC-MS analysis revealed B49%
To further exemplify the utility of this organocatalytic conversion to product 21 (structure confirmed by MS2) along
site-selective phosphorylation/deprotection sequence, selective with B21% recovered substrate; other minor peaks (o10%
mono-phosphorylation of the primary alcohol in the anti- each) also eluted but these could not be identified and did not
microbial agent chloramphenicolr (16) was investigated appear to correspond to phosphorylation at the Ser or Thr
(Scheme 2).
positions or elimination products thereof (see ESI†).
Table 4 Site-selective phosphorylation of triol 12
Product yielda (%)
Entry
Base/H+ scavenger (eq.)
Reaction time (h)
Phosphorylating agent (eq.)
13a
94 (87b,c
11
o5
—
13b
13c
13d
1
2
3
4
PPO (2.0)
PMP (1.0)
PMP (2.2)
PSr (3.2)
8
1
4
ClP(O)(OPh)2 (1.2)
XPCl (1.0)
ClP(O)(OPh)2 (2.2)
ClP(O)(OPh)2 (3.2)
)
o5
81
—
—
86
o5
—
—
—
96
o5
24
—
a
b
c
Isolated product yield after chromatographic purification. Isolated product yield of analogous reaction using XPCl (14, 1.2 eq.). Reaction
conducted in CH2Cl2 : DMF (9 : 1, 0.2 M) due to low solubility of reagents. ‘—‘ Indicates that this product was not detected by TLC analysis.
13610 | Chem. Commun., 2014, 50, 13608--13611
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
5 M. D. Hack, D. N. Rassokhin, C. Buyck, M. Seierstad, A. Skalkin,
P. ten Holte, T. K. Jones, T. Mirzadegan and D. K. Agrafiotis, J. Chem.
Inf. Model., 2011, 51, 3275–3286.
6 A. C. Dar and K. M. Shokat, Annu. Rev. Biochem., 2011, 80, 769–795.
7 Z. O’Brien and M. Fallah Moghaddam, Expert Opin. Drug Metab.
Toxicol., 2013, 9, 1597–1612.
8 J. I. Murray, A. C. Spivey and R. Woscholski, J. Chem. Biol., 2013, 6,
175–184.
9 B. R. Sculimbrene and S. J. Miller, J. Am. Chem. Soc., 2001, 123,
10125–10126.
10 B. R. Sculimbrene, A. J. Morgan and S. J. Miller, J. Am. Chem. Soc.,
2002, 124, 11653–11656.
11 B. R. Sculimbrene, Y. Xu and S. J. Miller, J. Am. Chem. Soc., 2004, 126,
13182–13183.
Scheme 3 Tyr kinase-like selectivity in the mono-phosphorylation of
heptapeptide 20 using catalyst 2l.
12 A. J. Morgan, S. Komiya, Y. Xu and S. J. Miller, J. Org. Chem., 2006,
71, 6923–6931.
13 Y. K. Wang, A. Morgan, K. Stieglitz, B. Stec, B. Thompson, S. J. Miller
and M. F. Roberts, Biochemistry, 2006, 45, 3307–3314.
14 Y. Xu, B. R. Sculimbrene and S. J. Miller, J. Org. Chem., 2006, 71,
4919–4928.
15 K. J. Kayser-Bricker, P. A. Jordan and S. J. Miller, Tetrahedron, 2008,
64, 7015–7020.
16 Y. K. Wang, W. Chen, D. Blair, M. Pu, Y. Xu, S. J. Miller,
A. G. Redfield, T. C. Chiles and M. F. Roberts, J. Am. Chem. Soc.,
2008, 130, 7746–7755.
17 C. M. Longo, Y. Wei, M. F. Roberts and S. J. Miller, Angew. Chem., Int.
Ed., 2009, 48, 4158–4161.
18 B. Sculimbrene, A. Morgan and S. Miller, Chem. Commun., 2003,
1781–1785.
19 S. Han and S. J. Miller, J. Am. Chem. Soc., 2013, 135, 12414–12421.
20 P. A. Jordan, K. J. Kayser-Bricker and S. J. Miller, Proc. Natl. Acad. Sci.
U. S. A., 2010, 107, 20620–20624.
21 P. A. Jordan and S. J. Miller, Angew. Chem., Int. Ed., 2012, 51,
2907–2911.
22 S. Jones, D. Selitsianos, K. J. Thompson and S. M. Toms, J. Org.
Chem., 2003, 68, 5211–5216.
In summary, 2-aryl-4-DMAP-N-oxides such as 2l are highly
efficient organocatalysts for the site-selective phosphorylation
of alcohols. Their use is particularly advantageous for sub-
strates prone to elimination and when selectivity for primary
(over sec and phenolic) or for phenolic (over primary and sec)
OH groups in polyols is required. Preliminary studies have also
demonstrated that catalyst 2l displays Tyr kinase-like selectivity
for the mono-phosphorylation of a heptapeptide containing
Tyr, Ser and Thr residues. These catalysts may therefore hold
promise for the preparation of phospho-peptides for biomedical
applications although clearly achieving selectivity in the presence
of alternative nucleophiles and more polar solvents will prove
challenging; further studies towards this end are ongoing in our
laboratory.
We thank the EPSRC, the ICB (Imperial College London) and
the SCI (Messel Scholarship, JIM) for funding and Dr Lisa
D. Haigh (Imperial College London) for assistance with LC-MS2
analysis.
23 S. Jones and C. Smanmoo, Org. Lett., 2005, 7, 3271–3274.
24 O. S. Fenton, E. E. Allen, K. P. Pedretty, S. D. Till, J. E. Todaro and
B. R. Sculimbrene, Tetrahedron, 2012, 68, 9023–9028.
25 V. A. Efimov, O. G. Chakhmakhcheva and Y. A. Ovchinnikov, Nucleic
Acids Res., 1985, 13, 3651–3666.
26 V. A. Efimov, A. A. Buryakova, I. Y. Dubey, N. N. Polushin,
O. G. Chakhmakhcheva and Ovchinnikov YuA, Nucleic Acids Res.,
1986, 14, 6525–6540.
27 I. Manet, S. Monti, P. Bortolus, M. Fagnoni and A. Albini, Chem. – Eur. J.,
2005, 11, 4274–4282.
28 K. Jarowicki and P. Kocienski, J. Chem. Soc., Perkin Trans. 1, 1998,
4005.
29 J. S. McMurray, D. R. Coleman, W. Wang and M. L. Campbell,
Biopolymers, 2001, 60, 3–31.
Notes and references
‡ By contrast, Efimov25,26 found that 4-DMAP-N-oxide (2d) outper-
formed pyridine-N-oxide (2a) for oligonucleotide coupling, presumably
due to competing catalyst demands in the dual catalytic cycle.
§ o-XPCl has been described for phosphorylation previously31 but its
use synthetically has not been pursued.
¶ H2/PtO2 as used for Ph phosphate deprotection reduces phenols.28–30
1 G. Manning, D. B. Whyte, R. Martinez, T. Hunter and S. Sudarsanam, 30 G. Toth, Z. Kele and G. Varadi, Curr. Org. Chem., 2007, 11, 409–426.
Science, 2002, 298, 1912–1934. 31 S. Yamago, US Pat., 2008177049A1, 2008.
2 H. J. Jessen, N. Ahmed and A. Hofer, Org. Biomol. Chem., 2014, 12, 32 G. Manning, D. B. Whyte, R. Martinez, T. Hunter and
3526–3530.
S. Sudarsanam, Science, 2002, 298, 1912–1934.
33 J. Dengjel, I. Kratchmarova and B. Blagoev, Mol. BioSyst., 2009, 5,
1112–1121.
3 Z. A. Knight and K. M. Shokat, Chem. Biol., 2005, 12, 621–637.
4 S. K. Grant, Cell. Mol. Life Sci., 2009, 66, 1163–1177.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13608--13611 | 13611