Synthesis of a new linear-cyclic ditopic polyazaligand with a xylenyl linker

Article abstract of DOI:10.1080/00397910802630203

The synthesis of a new linear-cyclic ditopic polyazaligand with a xylenyl linker 1 is described. The sequence is based on the stoichiometric reaction of a dissymmetrical biselectrophilic reagent, p-(chloromethyl)benzyl-p- toluenesulfonate, with phosphoryl

Full text of DOI:10.1080/00397910802630203

Synthetic Communications1, 39: 1992–2000, 2009
Copyright # Taylor & Francis Group, LLC
ISSN: 0039-7911 print=1532-2432 online
DOI: 10.1080/00397910802630203
Synthesis of a New Linear-Cyclic Ditopic
Polyazaligand with a Xylenyl Linker
´ `
Nathalie Le Bris, Helene Bernard, Zein Chamas, and Henri Handel
Department of Chemistry, University of Brest, Brest, France
Abstract: The synthesis of a new linear-cyclic ditopic polyazaligand with a
xylenyl linker 1 is described. The sequence is based on the stoichiometric reaction
of a dissymmetrical biselectrophilic reagent, p-(chloromethyl)benzyl-p-toluenesul-
fonate, with phosphorylcyclam to give rise to an original intermediate possessing
a p-chloroxylenyl pendant arm. The subsequent reaction of this compound with
glyoxal-protected linear tetraamine, followed by deprotection steps, leads to the
desired ditopic ligand with a good overall yield.
Keywords: Bisaminal, cyclam, ditopic ligand, linear tetraamine, phosphoryl
group
INTRODUCTION
Ditopic ligands based on polyazamacrocycles are widely described in the
literature, especially bismacrocycles, which are able to form stable com-
plexes with two metal ions ranging from transition metals to lanthanides
and other heavy metals.[1]
A few ditopic compounds constituted by a cyclic subunit combined
with an acyclic polyamine are also described.[2,3] The combination of
macrocyclic and open-chain polyamines is interesting because it produces
ligands with very different properties, notably in regard to their com-
plexation with metal ions. T. A. Kaden recently showed that the
Received November 4, 2008.
Address correspondence to Henri Handel, Department of Chemistry, UMR
CNRS 6521, University of Brest, Brest, France. E-mail: henri.handel@
univ-brest.fr
1992

Linear-Cyclic Ditopic Polyazaligand
1993
noncyclic moiety, acting as a flexible hook, interacts rapidly with a metal
ion to form a first complex that subsequently rearranges itself: the ion is
translocated from the linear branch inside the macrocyclic core to give
the thermodynamically more stable form. The free acyclic subunit can
coordinate a second metal ion added afterward to give rise selectively
to heterodinuclear metal complexes.[3,4]
However, the syntheses of these dissymmetrical ditopic ligands
require time-consuming protection and deprotection multistep sequences
involving Boc or Ts as protecting groups.[2,3]
Herein, we report the synthesis of a new ditopic ligand with a p-
xylenyl linker 1. This compound is structurally close to the well-known
bismacrocycle AMD3100[5] and the open chain octaamine p-xylenylbis-
1,4,8,11-tetraazaundecane (Scheme 1).[6]
The molecular recognition of anionic species with polyazaligands is
also a very important objective with numerous applications in biological
processes or in environmental areas,[1a,6b] and these latter ligands recently
have proved to be efficient receptors for inorganic phosphates and espe-
cially triphosphate anion in aqueous solution.[8] It was shown that, on
one hand, the ability of the bis-linear tetraamine to form easily poly-
protonated species constitutes undoubtedly the driving force favoring
the formation of ternary species; on the other hand, the rigidity of
AMD3100, because of organizing hydrogen bonds, represents a benefit
for the selectivity of the binding of anionic species. From this point of
view, the linear part of compound 1 should preserve a good ability to
form complexes with anionic species while its macrocyclic moiety should
introduce more selectivity in the recognition process.
Studies concerning the influence of the architecture of these
ditopic polyazaligands on factors governing the recognition of anionic
species and p-stacking interactions[9] with nucleotides should be of great
interest.
Scheme 1. Ditopic ligands.

1994
N. Le Bris et al.
RESULTS AND DISCUSSION
In a recent article, we described the synthesis of a new polyazaditopic
ligand constituted by the linear tetraamine 1,4,8,11-tetraazaundecane
and a cyclam moiety linked by a propyl bridge.[10] It consisted, in the first
step, of the functionalization of a phosphoryl-triprotected cyclam by a
3-bromopropyl pendant side chain after reaction with an excess of 1,3-
dibromopropane. The reaction of this stable functionalized intermedi-
ate with a glyoxal-protected linear tetraamine, followed by the sub-
sequent deprotection steps, gave the desired ligand with good overall
yield.
Our attempts to synthesize the analogous ligand bearing a p-xylenyl
linker 1 according to this method failed, certainly because of the high
reactivity of the functionalized phosphorylcyclam that led to polymeric
products and also because of the difficulties removing the excess a,a0-
dibromo-p-xylene. So, we decided to use a dissymmetrical biselectrophilic
reagent able to react selectively toward the free secondary amino group of
phosphorylcyclam 3 in a stoichiometric reaction. We chose the p-(chlor-
omethyl)benzyl-p-toluenesulfonate 2 because the tosylate anion is known
to be a better leaving group than the chloride one. This original product
2 was synthesized in good yield according to an atypical solvent-free
tosylation at solid state, in a mortar,[11] from commercially available
p-(chloromethyl)phenylmethanol (Scheme 2).
As expected, in the usual alkylation conditions, the free secondary
amino function of phosphorylcyclam 3 reacted univocally on the methy-
lene group bearing the sulfonate function to give the functionalized
intermediate p-chloroxylenephosphorylcyclam 4.[10,12] The 13C NMR
spectrum confirmed this selectivity because no signal was observed in
the characteristic area (around 71 ppm) of a methylene group close to a
sulfonate function. One can note that this intermediate 4 bearing a pen-
dant arm possessing a chloride function is stable enough to be isolated
and stocked for later use.
The subsequent selective monoalkylation of one of the secon-
dary nitrogen atoms of glyoxal-protected linear tetraamine 5 with
compound 4 was easily obtained according to our previously described
procedure.[6] The reaction gave the fully protected ligand 6 as a mixture
Scheme 2. Synthesis of biselectrophilic reagent 2.

Linear-Cyclic Ditopic Polyazaligand
1995
of diastereoisomers, a consequence of the chirality of the two protected
subunits.[10] NMR spectra of 6 are complex and difficult to interpret.
The deprotection of compound 6 was reached in two steps. In the
first one, the removal of the bisaminal bridge of glyoxal-protected linear
tetraamine moiety in hydrazine hydrate led to the partially deprotected
compound 7. In the second step, the hydrolysis of the triprotecting phos-
phoryl group of the macrocycle under mild acidic conditions gave the
fully deprotected octaamine 1 as hydrochloride salt.[10] Further treatment
with potassium carbonate in acetonitrile led to the free octaamine 1 with
an overall yield from phosphorylcyclam 3 of 80% (Scheme 3).
With the aim of studying the potentialities of the functionalized inter-
mediate 8 obtained in the reverse sequence, the reaction was undertaken
starting from glyoxal-protected linear tetraamine 5. The general course of
this sequence led to the desired compound with the same overall yield.
Nevertheless, intermediate 8 showed weaker stability than its analog 4
because of the presence of reactive secondary and ternary amino func-
tions; consequently, further alkylation involving phosphorylcyclam 2
was realized in a one-pot procedure (Scheme 4).
In conclusion, the usefulness of the dissymmetrical biselectrophilic
reagent p-(chloromethyl)benzyl-p-toluenesulfonate 2 for the synthesis of
the ditopic ligand possessing a p-xylenyl linker 7 was established. In each
step, the reaction is stoichiometric and univocal and avoids tedious
purifications. Functionalized phosphorylcyclam 4 and to a lesser extent
glyoxal-protected tetraamine derivative 8 represent interesting precursors
in the synthesis of a wide variety of new ditopic polyazaligands. Further
studies to evidence the benefit of such ditopic linear-cyclic structures
Scheme 3. Synthesis of the ditopic ligand 1. Reagents and conditions: (i) K2CO3,
CH3CN, rt, 2 h; (ii) K2CO3, CH3CN, rt, 15 h; (iii) N2H4ꢀH2O, D, 15 h; (iv) HCl,
15 h; and (v) K2CO3, CH3CN, D, 15 h.

1996
N. Le Bris et al.
Scheme 4. Alternative synthetic pathway of ligand 1. Reagents and conditions:
(i) K2CO3, CH3CN, rt, 2 h; (ii) phosphorylcyclam 3, K2CO3, CH3CN, rt, 15 h;
(iii) N2H4ꢀH2O, D, 15 h; (iv) HCl, 15 h; and (v) K2CO3, CH3CN, D, 15 h.
toward both metal ions and anions complexation are currently being
investigated.
EXPERIMENTAL
General
Phosphorylcyclam 3[10,12] and bisaminal of linear tetraamine 5[6,10] were
synthesized according to previously described procedures. All reagents
were of commercial quality, and solvents were dried according to stan-
dard procedures. NMR spectra were acquired on Bruker AC 300 or
DX Avance Bruker spectrometers. The chemical shifts are in d values
relative to the internal standard tetramethylsilane (TMS). Elemental
analyses were performed at the Service de Microanalyse of the CNRS,
Gif sur Yvette, France.
Synthesis and Data of p-(Chloromethyl)benzyl-p-toluenesulfonate 2[11]
A mortar was charged with p-(chloromethyl)phenylmethanol (1.57 g,
10 mmol), dry potassium carbonate (6.91 g, 50 mmol), and p-toluene-
sulfonyl chloride (2.86 g, 15 mmol), and the ingredients were ground
vigorously for 5 min. After the completion of tosylation, the remaining
p-toluenesulfonyl chloride (TsCl) was removed by addition of powered
potassium hydroxide (2.81 g, 50 mmol) and vigorously ground for
2 min. Addition of a small quantity of t-butanol (20 mg) accelerates the
disappearance of TsCl. The product was extracted with 3 ꢁ 20 mL of
CH2Cl2 and filtered, and the organic solvent was evaporated. The
product 2 was recrystallized in n-hexane to give a white solid with 90%
yield.
1
F ¼ 58ꢂC. H NMR (300.13 MHz, CDCl3): d ¼ 2.44 (s, 3H), 4.55 (s,
2H), 5.05 (s, 2H), 7.20 (m, 6H), 7.77 (d, 2H, J ¼ 9 Hz). 13C NMR
(75.47 MHz, CDCl3): d ¼ 21.7, 45.6, 71.3, 128.0, 128.9 (2C), 129.9, 133.2,

Linear-Cyclic Ditopic Polyazaligand
1997
133.6, 138.4, 144.9. Anal. calcd. for C15H15O3SCl, 0.5H2O: C, 56.34; H,
5.04; S, 10.03; Cl, 11.09. Found: C, 56.49; H, 4.99; S, 9.61; Cl, 10.81.
Synthesis and Data of p-Chloroxylenephosphorylcyclam 4
Phosphorylcyclam 3 (2.20 g, 9 mmol) and dry potassium carbonate
(6.22 g, 45 mmol) were mixed in acetonitrile (25 mL), and then a solution
of p-(chloromethyl)benzyl-p-toluenesulfonate 2 (2.79 g, 9 mmol) in aceto-
nitrile (25 mL) was added. The mixture was stirred at room temperature
for 2 h. The solids were filtered off, and acetonitrile was evaporated. The
obtained oil was taken up in ether (30 mL). After filtration and solvent
evaporation, compound 4 was obtained as a thick oil with 95% yield.
1H NMR (300.13 MHz, CDCl3): d ¼ 1.64 (m, 4H), 2.26–3.43 (m,
16H), 3.79 (m, 2H), 4.46 (s, 2H), 7.24 (d, 2H, JC-H ¼ 7 Hz), 7.44 (d,
2H, JC-H ¼ 7 Hz). 13C NMR (75.47 MHz, CDCl3): d ¼ 21.8, 26.2, 40.7,
41.7, 42.0, 44.3 (d, JP-C ¼ 11 Hz), 45.5 (d, JP-C ¼ 15 Hz), 46.0, 51.6 (2C),
52.4, 57.9, 128.4, 129.0, 135.7, 139.9. 31P NMR (121.49 MHz, CDCl3):
d ¼ 25.6.
Synthesis and Data of Fully Protected Intermediate 6
A solution of bisaminal 5 (728 mg, 4 mmol) in acetonitrile (20 mL) was
added to a solution of p-chloroxylenephosphorylcyclam 4 (1.53 g,
4 mmol) and potassium carbonate (2.76 g, 20 mmol) in acetonitrile
(20 mL). The mixture was stirred vigorously at room temperature for
15 h. Solids were filtered off, and acetonitrile was evaporated. The
oily residue was composed of a mixture of diastereoisomers of compound
6 with 95% yield.
13C NMR (75.47 MHz, CDCl3): d ¼ 21.4, 23.7, 25.7, 40.2–57.4 (com-
plex zone), 71.3, 74.5, 74.9, 87.0, 125.4, 128.3–128.6 (complex zone),
135.3–142.7 (complex zone). 31P NMR (121.49 MHz, CDCl3): d ¼ 25.5,
25.6.
Synthesis and Data of Phosphoryl-Protected Intermediate 7
The mixture of diastereoisomers of compound 6 was dissolved in hydra-
zine hydrate (30 mL) and refluxed for 15 h. The solution was extracted
with chloroform (3 ꢁ 20 mL). The organic layer was dried with MgSO4,
and solvent was removed under reduced pressure to give a pale oil
composed of the phosphoryl-protected intermediate 7 in 90% yield.
1H NMR (300.13 MHz, CDCl3): d ¼ 1.56 (m, 6H), 2.24–3.48 (m,
28H), 3.61 (s, 2H), 3.77 (m, 2H), 7.13 (d, 2H, JC-H ¼ 7 Hz), 7.32 (d,

1998
N. Le Bris et al.
2H, JC-H ¼ 7 Hz). 13C NMR (75.47 MHz, CDCl3): d ¼ 21.6, 25.9, 30.0,
40.4, 41.3 (2C), 41.6, 43.9, 43.9 (d, JP-C ¼ 11 Hz), 45.1 (d, JP-C ¼ 14 Hz),
45.1, 47.8 (2C), 48.3, 49.0, 51.3, 52.0, 52.2, 53.2, 57.5, 127.5 (2C), 128.4
(2C), 137.6, 138.5. 31P NMR (121.49 MHz, CDCl3): d ¼ 25.5.
Octaamine 1
Synthesis
Compound 7 was dissolved in 3M hydrochloric solution (30 mL) and
stirred at room temperature for 15 h. The solvent was removed under
reduced pressure. The resulting hydrochloride salt was washed with warm
ethanol (3 ꢁ 20 mL). The solid was filtered off and dried to give the ligand
7 as hydrochloride salt.
To obtain the free form of compound 1, the powder was mixed in
acetonitrile (20 mL) with potassium carbonate (10 mmol) and refluxed
for 15 h. Then solids were filtered off, and the solvent was removed to
give a pale oil of octaamine 1 in 90% yield. Overall yield from phosphor-
ylcyclam 3: 80%.
Data of Octaamine 1 as Hydrochloride Salt
1H NMR (300.13 MHz, D2O): d ¼ 2.20 (m, 6H), 3.20–3.78 (m, 28H), 4.38
(s, 2H), 4.57 (s, 2H), 7.61 (m, 4H). 13C NMR (75.47 MHz, D2O): d ¼ 20.7,
21.3, 25.5, 38.2, 39.6, 40.4 (2C), 43.7 (2C), 44.3, 45.6, 46.1, 47.1 (2C), 47.7
(2C), 50.4, 53.9, 61.7, 132.5, 133.8, 134.8, 135.4. Anal. calcd. for
C24H50N8, 7.3HCl, 4H2O: C, 37.49; H, 8.22; N, 13.99; Cl, 32.31. Found:
C, 37.42; H, 8.09; N, 13.96; Cl, 32.34.
Data of Free Octaamine 1
1H NMR (300.13 MHz, CDCl3): d ¼ 1.520 (m, 4H), 1.68 (m, 2H), 2.28–
2.82 (m, 28H), 3.38 (s, 2H), 3.59 (m, 2H), 7.12 (m, 4H). 13C NMR
(75.47 MHz, CDCl3): d ¼ 25.8, 28.2, 30.1, 41.4, 47.0, 47.6, 47.9 (2C),
48.5, 48.6, 48.9, 49.0 (2C), 50.5, 52.3, 53.0, 53.3, 54.1, 57.1, 127.5,
128.9, 137.0, 138.7.
REFERENCES
1. (a) Izatt, R. M.; Pawlak, K.; Bradshaw, J. S. Thermodynamic and kinetic
data for macrocycle interaction with cations, anions, and neutral molecules.
Chem. Rev. 1995, 95, 2529–2586; (b) Caravan, P.; Ellison, J. J.; McMurry, T. J.;

Linear-Cyclic Ditopic Polyazaligand
1999
Lauffer, R. B. Gadolinium(III) chelates as MRI contrast agents: Structure,
dynamics, and applications. Chem. Rev. 1999, 99, 2293–2352; (c) Kaden, T.
A. Dinuclear metal complexes of bis-macrocycles. Coord. Chem. Rev. 1999,
190–192, 371–389.
2. Alcock, N. W.; Clarke, A. J.; Errington, W.; Josceanu, A. M.; Moore, P.;
Rawle, S. C.; Sheldon, P.; Smith, S. M.; Turonek, M. L. Pyridyl-,
bipyridyl-, and terpyridyl-functionalised azamacrocycles. Supramol. Chem.
1996, 6, 281–291.
3. Siegfried, L.; Honecker, M.; Schlageter, A.; Kaden, T. A. Synthesis and
complexation properties of ditopic ligands built from a macrocyclic and an
open chain chelator moiety. J. Chem. Soc., Dalton Trans. 2003, 3939–3948.
4. (a) Kaden, T. A. Metal complexes of ditopic and polytopic macrocyclic
ligands. J. Incl. Phenom. Macrocycl. Chem. 2004, 49, 3–9; (b) Soibinet, M.;
Gusmeroli, D.; Siegfried, L.; Kaden, T. A.; Palivan, C.; Schweiger, A.
Kinetics and mechanism of the stepwise complex formation of Cu(II) with
tren-centered tris-macrocycles. J. Chem. Soc., Dalton Trans. 2005, 2138–2146.
5. (a) Ronconi, L.; Sadler, P. J. Using coordination chemistry to design new
medicines. Coord. Chem. Rev. 2007, 251, 1633–1648; (b) De Clercq, E. Time-
line: The bicyclam AMD3100 story. Nat. Rev. Drug Discov. 2003, 2, 581–587.
6. (a) Claudon, G.; Le Bris, N.; Bernard, H.; Handel, H. Powerful N-
monoalkylation of linear tetraamines via bisaminal intermediates. Eur.
J. Org. Chem. 2004, 5027–5030; (b) Claudon, G.; Bernard, H.; Le Bris,
N.; Handel, H. Synthesis of three dissymmetrical pentaamines via
bisaminals of linear tetraamine. J. Chem. Res. 2006, 356–358.
~
7. (a) Bianchi, A.; Bowman-James, C.; Garcia-Espana, E. (Eds.). Supramolecu-
lar Chemistry of Anions; Wiley-VCH: New York, 1997; (b) Garcıa-Espana,
~
´
´
E.; Dıaz, P.; Llinares, J. M.; Bianchi, A. Anion coordination chemistry in
aqueous solution of polyammonium receptors. Coord. Chem. Rev. 2006,
250, 2952–2986; (c) Aoki, S.; Kimura, E. Highly selective recognition of
thymidine mono- and diphosphate nucleotides in aqueous solution by ditopic
receptors zinc(II)-bis(cyclen) complexes (cyclen ¼ 1,4,7,10-tetraazacyclodode-
cane). J. Am. Chem. Soc. 2000, 122, 4542–4548.
8. Le Bris, N.; Bernard, H.; Tripier, R.; Handel, H. A potentiometric and NMR
investigestion of triphosphate recognition by linear and bismacrocyclic
octaamines possessing a 1,4-xylenyl linker. Inorg. Chim. Acta 2007, 360,
3026–3032.
´
9. (a) Delepine, A. S.; Tripier, R.; Handel, H. Cyclen-based bismacrocycles for
biological anion recognition. A potentiometric and NMR study of AMP,
ADP, ATP nucleotide complexation. Org. Biomol. Chem. 2008, 6, 1743–
´
~
1750; (b) Aguilar, J. A.; Celda, B.; Fusi, V.; Garcıa-Espana, E.; Luis, S. V.;
Marinez, M. C.; Ramirez, J. A.; Soriano, C.; Tejero, R. Structural character-
ization in solution of multifunctional nucleotide coordination systems.
J. Chem. Soc., Perkin Trans. 2 2000, 1323–1328.
10. Prokhorov, A.; Bernard, H.; Le Bris, N.; Marquet, N.; Handel, H. Synthesis
of a new ditopic ligand possessing linear and cyclic tetraaza subunits. Synth.
Commun. 2008, 38, 1589–1600.

2000
N. Le Bris et al.
11. Kazemi, F.; Massah, A. R.; Javaherian, M. Chemoselective and scalable
preparation of alkyl tosylate under solvent-free conditions. Tetrahedron
2007, 63, 5083–5087.
12. Filali, A.; Yaouanc, J. J.; Handel, H. Stoichiometric mono-N-functionaliza-
tion of tetraazamacrocycles via phosphoryl-protected intermediates. Angew.
Chem., Int. Ed. Engl. 1991, 30, 560–561.