Novel asymmetrically functionalized bis-dipicolylamine metal complexes: Peripheral decoration of a potent anion recognition scaffold

Article abstract of DOI:10.1039/b917692f

We report the design and synthesis of a novel class of asymmetrically functionalized, ditopic bis-dipicolylamine (BDPA) ligands. A key feature of this research involved the controlled, sequential functional group decoration of a potent molecular recognition scaffold. Calorimetric screening identified a BDPA analogue as a highly potent (K_a～ 10⁶ M^-1) and selective sensor for inorganic phosphate.

Full text of DOI:10.1039/b917692f

View Article Online / Journal Homepage / Table of Contents for this issue
www.rsc.org/obc
Volume 7 | Number 24 | 21 December 2009 | Pages 5037–5280
ISSN 1477-0520
FULL PAPER
Joel A. Drewry et al.
Highlightsin
Novel asymmetrically functionalized
bis-dipicolylamine metal complexes:
peripheral decoration of a potent
anion recognition scaffold
ChemicalScience
In this issue...
1477-0520(2009)7:24;1-A

PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Novel asymmetrically functionalized bis-dipicolylamine metal complexes:
peripheral decoration of a potent anion recognition scaffold†
Joel A. Drewry,a,b Steven Fletcher,a,b Haider Hassana and Patrick T. Gunning*a,b
Received 28th August 2009, Accepted 10th September 2009
First published as an Advance Article on the web 12th October 2009
DOI: 10.1039/b917692f
We report the design and synthesis of a novel class of asymmetrically functionalized, ditopic
bis-dipicolylamine (BDPA) ligands. A key feature of this research involved the controlled, sequential
functional group decoration of a potent molecular recognition scaffold. Calorimetric screening
identified a BDPA analogue as a highly potent (Ka ~ 106 M-1) and selective sensor for inorganic
phosphate.
Phosphate-containing anionic species are ubiquitous in biological
systems and play important mediatory roles in signal trans-
duction pathways, as well as in carrying genetic information.
Despite a significant effort to develop potent binding motifs, the
design and synthesis of receptors capable of phosphate anion
recognition under physiological conditions remains a significant
challenge.1 Lewis acidic zinc(II)2 and copper(II)3 homo-dinuclear
bis-dipicolylamine4 (BDPA, Fig. 1) and polyazamacrocyclic
ligand complexes5 have been successfully employed as receptors of
biologically relevant phosphates and phosphotyrosine-containing
peptide sequences.6 Homo-dinuclear BDPA scaffolds have been
used extensively to distinguish between phosphorylated and non-
phosphorylated species,7 and more recently as non-covalent tags
for phosphorylated proteins.8 However, given the lack of bind-
ing functionality to bestow selectivity beyond metal–phosphate
recognition, agent promiscuity for phosphorylated molecules is
routinely reported.9 Significantly, excepting the metal chelating
nitrogen donor groups and the hydrophobic pyridine rings, BDPA
constructs have no binding functionality and offer little to confer
receptor–phosphate substrate selectivity.
Our principal objective was to improve the substrate specificity
of the bis-dipicolylamine motif by developing a facile, modular
synthetic route to multi-functionalized BDPA receptors. In order
to maximize the potential utility of the BDPA scaffold, we then
wished to demonstrate the versatility of our synthetic route by
constructing BDPA analogues containing functionality amenable
to further covalent diversification, e.g. aryl halides that can
participate in Suzuki couplings and amines that can form peptide
bonds. Given the widespread application of BDPA constructs in
molecular recognition, facile synthetic protocols that give access
to customized receptors would be of significant benefit.
Scaffold designs focused primarily upon selectively substituting
the 2¢ and 3¢ positions of the flanking pyridine groups, which, due
to their proximities to the metal-chelating pyridine nitrogens, were
considered the most relevant to achieving additional substrate-
selective intermolecular interactions. Two novel sub-families of
bis-dipicolylamine analogues were prepared: a meta-xylyl bridged,
mono-functionalized bis-dipicolylamine (Fig. 1B), and a meta-
xylyl bridged, di-functionalized bis-dipicolylamine (Fig. 1C). The
structures of these novel BDPA-based molecular recognition
architectures are given in Table 1, all of which were prepared via the
Table 1 Library of asymmetric substituted bis-dipicolylamine scaffoldsa
Compound
R1
R2
Fig. 1 Asymmetric functionalization of the BDPA receptor.
1
2
3
4
5
6
7
8
9
H
H
H
H
H
3¢-COOMe
2¢- NH2
2¢-Br
aDepartment of Chemistry, University of Toronto, 3359 Mississauga Road
North, Mississauga, ON, Canada L5L 1C6
2¢-NH2
2¢-NH2
bDepartment of Chemical and Physical Sciences, University of Toronto,
3359 Mississauga Road North, Mississauga, ON, Canada L5L 1C6.
E-mail: patrick.gunning@utoronto.ca; Fax: 905-828-5425; Tel: 905-569-
4588
2¢-NHCOCH3
3¢-COOMe
3¢-COOMe
2¢Br
2¢-NHCOCH3
2¢-NHCOC(CH3)3
3¢-COOMe
2¢-NHCOC(CH3)3
† Electronic supplementary information (ESI) available: 1H and 13C
NMR, HRMS for compounds 1–9, and representative ITC traces. See
DOI: 10.1039/b917692f
a Where the counter anion used was the non-coordinating triflate species.
5074 | Org. Biomol. Chem., 2009, 7, 5074–5077
This journal is
The Royal Society of Chemistry 2009
©

Scheme 1 Reagents and conditions: (a) NBS, BPO, CCl4, 70 ◦C, 12 h, 45–68%; (b) 2-NsNH2, K2CO3, DMF, 60 ◦C, 8 h, 83–95%; (c) thiophenol, K2CO3,
DMF, rt, 3 h, 88–93%; (d) benzene-1,3-dicarboxaldehyde, NaBH(OAc)3, DCE, rt, 2 h, 84–92%; (e) NaBH(OAc)3, DCE, rt, 2 h, 70–91%; (f) Zn(OTf)2,
MeOH, 4 h, 85–95%.
facile and modular synthetic route depicted in Scheme 1. A simple
retrosynthetic analysis of BDPA and its analogues reveals that the
target molecules may be accessed from double N,N-dialkylation
of 1,3-bis(aminomethyl)benzene with 2-(bromomethyl)pyridines.
Indeed, this is the most popular synthetic route to generating
symmetrically functionalized BDPA structures.
However, such an approach is not compatible with the gen-
eration of asymmetrical BDPAs. To this end, we herein present
the controlled, stepwise reductive amination of the alternative
scaffold benzene-1,3-dicarboxaldehyde with synthetic secondary
dipicolylamines (Scheme 1).
appending a similar selection of simple functional groups. As
mentioned in the Introduction, we chose substituents amenable
to further scaffold extension and diversification, including aryl
halides, anilines and esters. To the best of our knowledge, this
work represents the first controlled, sequential decoration of the
BDPA scaffold. Given the excellent overall yields obtained and
functional diversity incorporated, we believe our methodology
lends itself well to rational receptor design and can be employed
to construct numerous novel BDPA ligands.
Isothermal titration calorimetry (ITC) was used to screen
receptors 1–9 for binding affinity against a range of phosphate
species (Table 2). Control binding data was collected for receptor
1 (R1 = R2 = H), which we regarded as a ‘naked’ BDPA complex
and from which all synthetic modifications were compared.
Thus, ITC was used to accurately assess the relative effects of
functionalization upon agent–substrate specificity and direct our
synthetic modifications. The binding abilities of receptors 1–6
towards various phosphates were measured under identical ex-
perimental conditions (VP-ITC, 25 ◦C, pH 7.2, HEPES buffer
(50 mM) in aqueous solution; 0.1 mM receptor (cell), 2 mM
phosphate (syringe)). The resulting binding isotherms were fitted
to a 1 : 1 non-linear regression model using the Microcal ORIGIN
software package, affording the thermodynamic parameters of
the host–guest interaction. From the data obtained, the most
striking result was the highly potent binding of NaH2PO4 by
the asymmetric mono-amino-functionalized ligand 3 (R1 = 2¢-
NH2, R2 = H (entry 3)) relative to the other phosphorylated
substrates and relative to the other ligands. For example, 3 binds
to NaH2PO4 (Ka = 1.50 ¥ 106 M-1) almost eleven times as potently
as adenosine-5¢-monophosphate (5¢AMP; Ka = 1.37 ¥ 105 M-1)
and about 100 times more strongly than both b-glycerophosphate
(b-GP; Ka = 1.60 ¥ 104 M-1) and p-nitrophenyl phosphate (PNPP;
Ka = 1.32 ¥ 104 M-1). Of particular significance is the observation
that ligand 3 binds NaH2PO4 approximately ten times as strongly
as the control compound 1 (3, Ka = 1.50 ¥ 106 M-1 cf. 1, Ka =
1.93 ¥ 105 M-1, Fig. 2). Initially, it was presumed that the ten-
fold increase in Ka was a direct result of increased hydrogen bond
interactions between the two 2¢-amino groups of 3 and the oxygen
hydrogen bond acceptors of NaH2PO4. However, binding of 3
with NaH2PO4 incurred an unfavourable, two-fold increase in the
enthalpy contribution (3, DH◦ = 4.38 kcal mol-1 cf. 1, DH◦ = 2.18
kcal mol-1) that was more than compensated for by a significant
increase in favourable entropy (3, TDS◦ = 12.8 kcal mol-1 cf.
1, TDS◦ = 9.38 kcal mol-1), suggesting that tighter binding was
entropically driven.
The desired secondary dipicolylamines 13 and 15 were generally
furnished in three steps from commercially available starting
materials employing 2-nitrobenzenesulfonamide (2-NsNH2) as
a latent secondary amino group.10 Briefly, as a representative
example, NBS-mediated aryl-methyl bromination of methyl 6-
methylnicotinate 10 (R1 = CO2Me) with catalytic benzoyl peroxide
(BPO) afforded 11 in 60% yield. Subsequently, N-dialkylation of
2-NsNH2 with two equivalents of bromide 11 gave 12 in excellent
yield. Mild nosyl deprotection was accomplished using thiophenol
and K2CO3 in DMF at room temperature overnight to furnish
the target secondary amine 13. Under high dilution conditions,
benzene-1,3-dicarboxaldehyde was mono-reductively aminated
with the R1-substituted dipicolylamine 13 using NaBH(OAc)3
in situ at room temperature to give 14 in high yield. The remaining
aldehyde group in 14 was then reductively aminated with the R2-
substituted dipicolylamine 15 to give the novel, asymmetrically
functionalized meta-xylene-based scaffold 16, exhibiting two dif-
ferent (R1 and R2) dipicolylamine moieties. Synthetic procedures
and characterization for all derivatives in Table 1 are provided in
the ESI.†
We considered that a BDPA scaffold containing a range of
hydrogen bond donors, acceptors and hydrophobic substituents
at either the 2¢- or 3¢-positions could interact effectively with
a range of target substrates. Indeed, phosphate hydrolysis cat-
alysts based on a 2¢-substituted tetra-amino BDPA scaffold,
prepared by N-dialkylation procedures, proved particularly po-
tent against model RNA substrates.11 Our goal was to exploit
subtle differences in substrate composition with tailored receptor
species.
First, we successfully prepared a small set of asymmetrically
mono-functionalized BDPA analogues of the parent 1 (R1
=
R2 = H), incorporating a range of substituents (compounds 2–
4, Table 1). Subsequently, we prepared both symmetrically and
asymmetrically di-functionalized BDPAcompounds 5–9 (Table 1),
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 5074–5077 | 5075
©

Table 2 Thermodynamic binding data (ITC) for various substrates to a range of receptors in 50 mM HEPES buffer, pH 7.2, 25 ◦
C
Entry
Agent
na
Substrate
K ¥ 104 M-1
DG◦/kcal mol -1b
DH◦/kcal mol-1
TDS◦/kcal mol-1
1
2
3
4
5
6
7
8
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
1
1
1
1
1
1
1
1
Na[H2PO4]
Na[H2PO4]f
Na[H2PO4]
Na[H2PO4]f
Na[H2PO4]
Na[H2PO4]
5¢-AMPc
5¢-AMPc,f
5¢-AMPc
5¢-AMPc,f
5¢-AMPc,f
5¢-AMPc,f
b-GPd
19.3 1.23
15.5 3.94
150 6.82
10.6 0.52
16.3 0.42
1.70 0.11
16.3 0.71
12.4 0.73
13.7 0.59
ndg
-7.20
-7.08
-8.42
-6.82
-7.12
-5.75
-7.09
-6.94
-6.99
ndg
2.18 0.02
-0.17 0.02
4.38 0.02
3.58 0.07
5.98 0.04
4.31 0.37
3.31 0.27
2.83 0.05
5.01 0.05
ndg
9.38
6.91
12.8
10.4
13.6
10.1
10.4
9.77
12.0
ndg
9
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
ndg
ndg
1
ndg
ndg
ndg
ndg
7.17 0.61
2.35 0.09
2.93 0.16
1.60 0.04
ndg
1.00 0.01
1.26 0.01
1.47 0.03
1.34 0.06
1.32 0.02
4.47 0.61
4.53 1.74
1.57 0.01
-6.62
-5.87
-6.08
-5.66
ndg
-5.48
-5.60
-5.66
-5.08
-5.55
-6.35
-6.34
-5.70
0.98 0.02
6.03 0.17
6.12 0.30
7.14 0.22
ndg
0.89 0.05
1.76 0.09
7.54 0.20
9.02 1.11
7.25 0.11
0.62 0.02
0.25 0.02
2.64 0.12
7.59
11.9
12.2
12.8
ndg
6.37
7.36
13.2
14.6
12.8
6.97
6.58
8.34
1
1
b-GPd,f
1
b-GPd
ndg
1
b-GPd,f
b-GPd
1
1
1
1
1
1
1
b-GPd
PNPPe
PNPPe,f
PNPPe
PNPPe,f
PNPPe,f
PNPPe,f
a Binding stoichiometry. b Calculated from DG◦ = DH◦–TDS◦. c Adenosine-5¢-monophosphate. d b-Glycerophosphate; e p-Nitrophenyl phosphate. f Buffer
included 5% DMSO (v/v). g Not determined: thermodynamic parameters were not obtained due to insufficient enthalpy of binding to accurately fit the
data to a 1 : 1 binding model.
Fig. 2 Representative ITC binding isotherms for 1 (A) and 3 (B) binding to NaH2PO4 (30 ¥ 10 mL injections), 50 mM HEPES buffer, pH 7.4, 25 ◦C.
In order to rationalize this thermodynamic data, we postulated
that the increased enthalpic penalty was required to break the
solvation sphere of the hydrophilic amino groups of 3, but in
which this cost was more than repaid by the associated liberation
of several ordered water molecules to afford the observed positive
entropic term. Our data does not preclude the fact that increased
intermolecular hydrogen bonding interactions are taking place
between 3 and NaH2PO4, only that overall there is an enthalpic
penalty component involved in their formation—specifically,
desolvation—for which the new bonds do not fully compensate. In
turn, this would suggest that BDPA receptors equipped with more
hydrogen bond donors may better compensate the apparent loss of
enthalpy via increased host–guest interactions. Thus, receptor 5,
equipped with four aniline hydrogen bond donors (R1 = 2¢-NH2,
R2 = 2¢-NH2) was prepared; however, this incurred a ten-fold
decrease in binding potency with NaH2PO4 relative to the binding
5076 | Org. Biomol. Chem., 2009, 7, 5074–5077
This journal is
The Royal Society of Chemistry 2009
©

of 3 with NaH2PO4, respectively (5, Ka = 16.3 ¥ 104 M-1 cf. 3, Ka =
150 ¥ 104 M-1 (entries 3 and 5)). Unsurprisingly, binding 5 with
NaH2PO4 resulted in a favourable increase in the entropy relative
to 3 (5, TDS◦ = 13.6 kcal mol-1 cf. 3, TDS◦ = 12.8 kcal mol-1),
presumably due to the desolvation of the additional aniline groups.
However, the improved entropic component (D(TDS◦) of 0.8 kcal
mol-1) was negated with a concomitant unfavourable increase in
with increased diversity of substituents to probe further substrate
specificity.
In conclusion, we report the development of a highly efficient,
facile and modular synthetic route to orthogonally functionalized
BDPA receptors. Our initial synthetic efforts have identified
BDPA derivative 3 as an extremely potent and relatively selective
receptor for NaH2PO4 (Ka = 1.50 ¥ 10-6 M-1). Given the modular
nature of the synthetic protocols outlined here, there is great
scope to develop novel multi-functionalized BDPA recognition
architectures.
We gratefully acknowledge the University of Toronto, Canadian
Foundation for Innovation, Ontario Research Fund and Con-
naught Foundation for financial support of this work. We would
also like thank Vijay Shahani for his computational assistance.
enthalpy contribution (5, DH◦ = 5.98 kcal mol-1, cf. 3, DH◦
=
4.38 kcal mol-1), a likely result of the increased energy required
to desolvate the additional aniline functional groups. Our study
illustrates the delicate structural balance required to furnish potent
anion receptors operating in an aqueous environment. As noted
by Berger and Schmidtchen,12 simply increasing the number of
complementary binding groups on the receptor molecule does not
necessarily confer increased binding affinity and ignores enthalpy–
entropy compensation effects.13 One must consider whether the
favourable binding enthalpy achieved through a more extensive
binding interface is sufficient to compensate for the resultant
unfavourable enthalpic cost of increased receptor desolvation.
Our modular synthetic approach facilitates rapid and iterative
functional group modifications to the BDPA receptor–allowing
facile access to ‘solvation design’ receptors.
Notes and references
1 S. Tamaru and I. Hamachi, Recognition of Anions, 2008, 95–125; M.
Kruppa and B. Konig, Chem. Rev., 2006, 106, 3520–3560; P. D. Beer
and P. A. Gale, Angew. Chem., Int. Ed., 2001, 40, 486; E. A. Katayev,
P. Melfi and J. L. Sessler, Modern Supramolecular Chemistry., 2008,
315–347.
2 P. T. Gunning, R. D. Peacock and A. C. Benniston, Chem. Commun.,
2004, 2226–2227; A. C. Benniston, P. T. Gunning and R. D. Peacock,
J. Org. Chem., 2005, 35, 116–123; J. C. Mareque-Rivas, R. Prabaharan
and R. Torres Mart´ın de Rosales, Chem. Commun., 2004, 76–77; J. C.
Mareque-Rivas, R. Torres Mart´ın de Rosales and S. Parsons, Chem.
Commun., 2004, 610–611.
3 S. L. Tobey and E. V. Anslyn, J. Am. Chem. Soc., 2003, 125, 10963–
10970.
4 H. Tamamura, A. Ojida, T. Ogawa, H. Tsutsumi, H. Masuno, H.
Nakashima, N. Yamamoto, I. Hamachi and N. Fugii, J. Med. Chem.,
2006, 49, 3412–3415; M. S. Han and D. H. Kim, Angew. Chem., Int.
Ed., 2002, 41, 3809–3811.
5 E. Kimura, S. Aoki, E. Kikuta and T. Koike, Proc. Natl. Acad. Sci.
U. S. A., 2003, 100, 3731–3736.
6 A. Ojida, Y. Mito-oka, K. Sada and I. Hamachi, J. Am. Chem. Soc.,
2004, 126, 2454–2463.
7 T. Anai, E. Nakata, Y. Koshi, A. Ojida and I. Hamachi, J. Am. Chem.
Soc., 2007, 129, 6232–6239; A. Ojida, Y. Mito-oka, M. Inoue and I.
Hamachi, J. Am. Chem. Soc., 2002, 124, 6256–6258.
8 T. Kohira, K. Honda, A. Ojida and I. Hamachi, ChemBioChem, 2008,
9, 698–701.
More generally, the binding of all four phosphate substrates
including NaH2PO4 to our host molecules 1–6 were driven by
large positive entropic terms and were, with the exception of
the binding of 2 to NaH2PO4 (entry 2), endothermic (+DH◦).
Several groups have reported similar entropy-driven endothermic
binding interactions between organometallic complexes and anion
guests.6,14 In water, thermodynamically favourable increases in
entropy are routinely observed in both artificial15 and biological
host–guest systems.16 Better known as the hydrophobic effect,
the gain in entropy is obtained by effective displacement of
ordered water from the hydrophobic surfaces of the host–guest
binding interfaces.17 Therefore, desolvation of the predominantly
hydrophobic DPA units via anion binding to the Lewis acidic zinc
ion would explain the favourable entropic term.18 Unsurprisingly,
binding to the most hydrophobic substrate PNPP, with presum-
ably the most ordered and largest solvation sphere of the four
phosphates examined, resulted in the biggest entropic gain (where
1–3, TDS◦ > 12.8 kcal mol-1 (entries 19–21, respectively)).
More generally, we noted the high affinities of receptors 1–6
for NaH2PO4 and 5¢-AMP relative to b-GP and PNPP. The
binding affinities for b-GP and PNPP were approximately ten-
fold lower in potency in all four host compounds, e.g. 1 binds
preferentially to 5¢-AMP (Ka = 1.63 ¥ 105 M-1) over PNPP (Ka =
1.47 ¥ 104 M-1). However, receptor specificity for the organic
phosphate substrates (5¢-AMP, b-GP and PNPP) remained elusive,
with the differences in binding affinities, within error, generally
negligible. We attributed the lack of substrate specificity to a
lack of complementary binding functionality present on the
BDPA scaffold. We are currently developing BDPA scaffolds
9 T. Sakamoto, A. Ojida and I. Hamachi, Chem. Commun., 2009, 141–
152.
10 T. Fukuyama, M. Cheung, C. K. Jow, Y. Hidai and T. Kan, Tetrahedron
Lett., 1997, 38, 5831–5834.
11 G. Feng, D. Natale, R. Prabaharan, J. C. Mareque Rivas and N. H.
Williams, Angew. Chem., Int. Ed., 2006, 45, 7056–7059.
12 M. Berger and F. P. Schmidtchen, Angew. Chem., Int. Ed., 1998, 37,
2694–2696.
13 D. H. Williams and M. S. Westwell, Chem. Soc. Rev., 1998, 27, 57–63.
14 (a) M. Rekharsky, Y. Inoue, S. Tobey, A. Metzger and E. Anslyn,
J. Am. Chem. Soc., 2002, 124, 14959–14967; (b) S. Kubik, R. Kirchner,
D. Nolting and J. Seidel, J. Am. Chem. Soc., 2002, 124, 12752–12760.
15 P. T. Gunning, Org. Biomol. Chem., 2005, 3, 3877–3879.
16 D. Chandler, Nature, 2005, 437, 640–647.
17 C. Tanford, The Hydrophobic Effect, 2nd edn, Wiley, New York, 1980.
18 D. A. Stauffer, R. E. Barrans, Jr. and D. A. Dougherty, J. Org. Chem.,
1990, 55, 2762–2767.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 5074–5077 | 5077
©