Phosphonated chelates for nuclear imaging

Organic &

Biomolecular Chemistry

View Article Online

View Journal | View Issue

PAPER

Phosphonated chelates for nuclear imaging†

a,b

Cite this: Org. Biomol. Chem., 2014, Sabah Abada, Alexandre Lecointre, Câline Christine,*_c

a,b

c

d,e

12, 9601

Laurence Ehret-Sabatier, Falk Saupe, Gertraud Orend, David Brasse,

Ali Ouadi, Thomas Hussenet, Patrice Laquerrière, Mourad Elhabiri* and

Loïc J. Charbonnière*

d,e

c

d,e

f

a,b

A series of bis-, tris- and tetra-phosphonated pyridine ligands is presented. In view of their potential use as

chelates for radiopharmaceutical applications, the physico-chemical properties of the ligands and of their Co(II),

Ni(II), Cu(II), and Zn(II) complexes were studied by means of potentiometry and UV-Vis absorption spectroscopy.

a

The pK values of the ligands and of the complexes, as well as the stability constants for the formation of the

complexes, are presented. The kinetic aspects of the formation of Cu(II) complexes and of their dissociation in

acidic media were studied by means of stopped ﬂow experiments, and the stability of the Cu(II) complex toward

reduction to Cu(I) was investigated by cyclic voltammetry and by titration with diﬀerent reducing agents. The

diﬀerent thermodynamic and kinetic aspects of the polyphosphonated ligands were compared with regard to

the impact of the number of phosphonic acid functions. Considering the very promising properties for com-

99m

plexation, preliminary SPECT/CT imaging experiments were carried out on mice with

Tc using the bis- and

2

1

tetra-phosphonated ligands L and L . Finally, a bifunctional version of chelate L , L*, was used to label MTn12,

a rat monoclonal antibody with both speciﬁcity and relatively high aﬃnity for murine tenascin-C. The labeling

was monitored by MALDI/MS spectrometry and the aﬃnity of the labeled antibody was checked by immuno-

Received 18th July 2014,

Accepted 26th September 2014

DOI: 10.1039/c4ob01514b

99m

staining experiments. After chelation with

Tc, the

Tc-L*-MTn12 antibody was injected into a transgenic

www.rsc.org/obc

mouse with breast cancer and the biodistribution of the labeled antibody was followed by SPECT/CT imaging.

nuclear medicine, particularly in two major imaging modal-

ities, Positron Emission Tomography (PET) and Single Photon

Introduction

The development of radiopharmaceuticals is currently receiv- Emission Computed Tomography (SPECT). Many radioiso-

6

4

ing much interest for diagnostic imaging and therapy in topes are used for these two techniques, among them Cu

+

−

(

5

6

T

1/2 = 12.7 h, β : 17.8%, E

β

+

= 656 keV; β : 38.4%, E

β

−

=

1

–5

99m

73 keV) is particularly promising for PET, and

h, pure γ-emitter, E = 140 keV) is the radionuclide of choice

for SPECT imaging.

Tc (T₁_/₂

a

b

IPHC, Université de Strasbourg, 25 rue Becquerel, 67087 Strasbourg Cedex, France

CNRS, UMR 7178, 67087 Strasbourg Cedex, France. E-mail: l.charbonn@unistra.fr,

γ

2,5,6

caline.christine@unistra.fr

c

Although these radionuclides can be used for their intrinsic

Inserm-Université de Strasbourg UMR_S 1109, MN3 T team, 3 avenue Molière,

99m

− 7

properties, as in the case of thyroid imaging with

^T^c^O4

,

6

7200 Strasbourg, France

d

64

99m

Université de Strasbourg, IPHC, 23 rue du Loess, 67037 Strasbourg, France

CNRS, UMR7178, 67037 Strasbourg, France

the applications of Cu and

ceuticals generally require the use of bifunctional chelates

BFC) designed to selectively bind the radionuclide and to be

covalently linked to a bio-targeting vector such as an antibody,

Tc as targeted radiopharma-

e

f

Laboratoire de Chimie Bioorganique et Médicinale, UMR 7509 CNRS-Université de

Strasbourg, ECPM, 25 rue Becquerel, 67087 Strasbourg Cedex, France.

E-mail: elhabiri@unistra.fr

(

†

Electronic supplementary information (ESI) available: the Experimental section a peptide or a protein. The ideal BFC should form thermo-

for the physico-chemistry investigations; potentiometric plots of the cupric com- dynamically stable and kinetically inert complexes to prevent

1

2

3

4

plexes formed with ligands L , L , L and L together with the corresponding

any ligand exchange reactions or hydrolysis in vivo. The com-

plexation should be eﬃcient and rapid at room temperature

and at low concentration, at a pH suitable for biological

hyperquad analysis; distribution diagrams as a function of pH of the protonated

1

2

3

4

metal complexes formed with ligands L , L , L and L (four figures); absorption

spectral variation induced by copper(II) complexation as a function of pH for

2

4

ligands L to L (four figures); electrochemical analysis of the copper(II) com- applications.

plexes (two Figures); variation of the pM values as a function of pH for ligands

6

4

99m

Among the possible ligands for Cu and

Tc complexa-

2

4

L

to L (three Figures); ESI-MS characterization of the metal complexes (one

tion, hydrophilic phosphonated ligands have proved to be

1

13

31

table); H, C and P NMR spectra of the compounds; stability and protonation

constants of the metallic complexes with Ni(II), Zn(II) and Co(II) formed with

8–16

eﬃcient chelating agents.

With phosphonated ligands, the

1

4

1

13

31

thermodynamic stability constants of the Cu(II) complexes are

ligands L –L (one table); H, C and P NMR spectra of the compounds. See

17

DOI: 10.1039/c4ob01514b

notably higher than those of their acetate analogues.

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9601

View Article Online

Paper

Organic & Biomolecular Chemistry

1

2

3

4

Scheme 1 Tetra-phosphonated ligand L , the activated BFC L*, and the bis and tris-phosphonated ligands L , L and L .

Our eﬀorts to design new chelate systems for subsequent zylamine 5 in two steps (phosphorylation and catalytic hydro-

uses in copper-64 PET imaging led us to the synthesis of the genolysis) and with an overall 81% yield. N-alkylation reaction

1

tetra-phosphonated ligand L (Scheme 1), a very eﬀective and of 2 with 7 resulted in the formation of the bis-phosphonated

1

1,12

2

selective Cu(II) chelating agent (log KCuL = 22.7).

We have compound 4 (24%) which led to ligand L after deprotection of

1

also shown that L can be derivatized into an activated phos- the phosphonate groups using an excess of TMSBr, followed

phonated bifunctional chelate L* (Scheme 1) which can be by methanolysis (83%).

1

8

3

coupled to small peptides or large antibodies.

The synthesis of ligand L was based on a mono N-alkyl-

2

0

In order to fully understand the coordination chemistry of ation of 2 with tetra-ethyl iminobis(methanephosphonate) to

phosphonated chelates, we have designed a series of three new give the bis-diethylester phosphonated intermediate 8 in 58%

ligands bearing two (L ) or three (L , L ) phosphonate func- yield. Subsequent N-alkylation of 8 with diethyl(ethylamino)-

tions on a pyridine framework (Scheme 1). The synthesis of methylphosphonate 7 in acetonitrile aﬀorded the tris-phos-

these ligands is reported here, together with their complexa- phonated precursor 9 (49%). Hydrolysis of phosphonate esters

tion properties towards d elements such as Cu(II), Ni(II), Zn(II) to their corresponding acids in the presence of an excess of

for ligands L , L , L ), and Co(II) ions (for ligand L ).

Tc imaging of 90% yield.

mice will be presented for ligands L and L and for a L*

2

3

4

2

3

4

2

3

(

TMSBr in CH

2

Cl

2

gave, after methanolysis, the ligand L in a

9

9m

Finally the potential of such chelates for

1

2

4

Finally, the formation of ligand L was started by the phos-

phonylation of 2,6-dibromomethylpyridine 2 with triethylphos-

phite to yield intermediate 10 (48%) which, after N-alkylation

labeled monoclonal antibody directed toward breast cancer.

2

0

reaction with tetra-ethyl iminobis(methanephosphonate),

gave the tris-phosphonated derivative 11 in 32% yield. Hydro-

lysis of phosphonate groups using 6 M HCl aﬀorded the ligand

L in 46% yield.

Results and discussion

Synthesis of the ligands

4

Scheme 2 presents the diﬀerent synthetic protocols used for

2

4

2

the synthesis of the ligands L to L . L was obtained according

to two diﬀerent routes. The first one started from bromation of

the commercially available 2,6-dihydroxymethylpyridine 1 with

1

4

Coordination chemistry of ligands L –L

1

4

1

4

Protonated species of ligands L –L . L to L constitute a

PBr

3

at 0 °C, aﬀording 2,6-dibromomethylpyridine 2 (84–95%), homogeneous series of chelators with varying numbers of

2

−

which is then converted into 2,6-bis(ethylaminomethyl)pyri- terminal –PO

dine 3 in 23% yield. Phosphorylation of 3 in the presence of L is closely related to L (both ligands contain three phospho-

diethyl phosphite and 37% aqueous formaldehyde aﬀorded nated –PO

3

phosphonate functions ranging from 2 to 4. If

1

9

4

3

2

−

3

functions), it markedly diﬀers from L with one

the bis-phosphonated compound 4 in 69% yield. Subsequent methylamino-N,N-bis(methanephosphonate) substituent being

2

−

hydrolysis of the phosphonate groups was performed using an replaced by an ethylphosphonate moiety, –CH –PO

excess of TMSBr, followed by methanolysis, aﬀording ligand L

.

2

3

2

Detailed interpretation of physico-chemical data on the

in 83% yield.

metallic complexes (e.g. Cu(II), Ni(II), Zn(II), Co(II)) first requires

However, due to the diﬃculties encountered in the purifi- the complete characterization of the ligand protonation

cation of compound 3, a second approach was elaborated for diagram, at least in the pH range in which these data have

2

the synthesis of L . This protocol started with the preparation been collected and also with respect to the targeted appli-

of diethyl(ethylamino)methanephosphonate 7 from ethylben- cation (i.e. PET and SPECT imaging). We have therefore investi-

9602 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

2

4

Scheme 2 Synthetic protocols for the preparation of ligands L to L . (i) PBr

3

, 0 °C, DMF, 84–95%. (ii) EtNH

2

, MeCN, reﬂux to r.t., 23%. (iii) HPO-

2

3

(

H

7

OEt)

, Pd/C, EtOH, r.t., 98%. (vi) K

0 °C, 49%. (ix) P(OEt)

2

, 0 °C, (CH

2

O)

n

(37% in H

2

O), r.t. to reﬂux for 4, 69%; 0 °C to reﬂux for 6, 83%. (iv) TMSBr, CH Cl , r.t., then MeOH, 83% for L ; 90% for L . (v)

2

CO

3

, MeCN, 50 °C, 24%. (vii) HN(CH

2

PO(OEt)

2

)

2

, K

2

CO

3 2 3

, MeCN, 70 °C, 58% for 9; 32% for 12. (viii) 7, K CO , MeCN,

3

, 135 °C, 48%. (x) 6M HCl, reﬂux, 46%.

1

4

gated the acido-basic properties of ligands L –L in the pH

range 2.0–11.5 by various techniques. First, potentiometric in water

H

n

1

4

Table 1 Protonation constants (logarithmic K

values) of ligands L –L

a

1

4

titrations of ∼2 mM solutions of L –L followed by statistical

first yielded six, four, five and four

protonation constants, respectively (Table 1). Importantly, Equilibrium

H

n

2

1,22

log K

(3σ)

processing of the data

1

b,c

2 b

3 b

4 b

L

H

1

some of the protonation constants were diﬃcult to determine

K

nꢀ

þH^þ!LH

1ꢀm

d

L

11.21(2)

11.1(2)

10.46(2)

11.04(1)

11.8(2)

with a good accuracy in view of their extremely low values

e

H

K

H

23,24

2

1ꢀn

þ!

2ꢀm

d

e

d

(

log K ≪ 2).

A typical example of potentiometric titrations

LH

þH

LH

10.29(2)

10.71(3)

10.07(3)

10.4(1)

8.3(3)_e

2

e

H

K

9.94(3)

10.8(4)

7.9(3)

6.5(5)

with the corresponding speciation diagram obtained after pro-

3

LH²₂

ꢀn

_þ_Hþ

3ꢀm

8.04(4)

8.5(2)

d

6.56(4)

d

6.9(2)

d

!

2

3

cessing the data is reported for L in Fig. 1.

e

H

K

6.8(2)

1

4

3ꢀn

þ!

4ꢀm

d

The most basic protonation constants (K ^aⁿ^d^KLH2 for L –

LH

þH

6.49(6)

4.52(5)

5.7(3)

5.1(6)

3

4

3

4

e

H

K

7.0(3)

5.37(4)

L and K for L ) can be attributed to the tertiary amines, in

LH

5

LH⁴₄ þH^þ!LH

ꢀn

5ꢀm

d

5.53(8)

—

4.5(3)

—

5

agreement with data available in the literature for related poly-

H

K

6

2

5–31

5ꢀn

þ!

6ꢀm

d

aminomethanephosphonate analogues.

The following pro- LH5 þH

LH6

4.19(9)

—

tonation constants are associated with the first protonation

a

b

Solvent: H

2

O; I = 0.1 M (NaClO

4

); T = 25.0 ± 0.2 °C. m = 8, 4 and 6

2

−

2−

1

2

3

4

c

constants of each of the phosphonate –PO

3

units (i.e. RPO

3

respectively for L , for L and for L and L . According to ref 11.

d

e

H

(n−

+

−

Potentiometry. Absorption spectroscopy vs. pH. K

n

= [LH

LH(n−1)]. The uncertainties (3σ) in the log K values correspond to the

added standard deviations (σ) in the cumulative constants. All other

3 3 2 a

RPO H + H ↔ RPO H ) as well as the pK of the pyridine protonation constants have been assumed to be <2.

n

]/[H ]

+

H ↔ RPO H ), while the logarithmic values of the second

3

(n−

H

[

protonation constants of the protonated phosphonate groups

−

+

(

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9603

View Article Online

Paper

Organic & Biomolecular Chemistry

2

3

Fig. 1 (a) Potentiometric titration curve of ligand L and (b) distribution

Fig. 2 (a) Absorption spectrophotometric titration of L as a function of

2

diagrams of the protonated species of the free ligand L . Solvent: water;

pH and (b) variation of the absorbance at 256 nm vs. pH. Solvent: H O;

2

−3

3

−4

I = 0.1 M (NaClO

1.85.

4

); T = 25.0(2) °C; [L ]tot = 2.44 × 10 M; 2.47 < pH <

4 tot

T = 25.0(2) °C; I = 0.1 M (NaClO ); l = 1 cm; [L ] = 2.29 × 10 M.

1

were estimated to be ≪2. If the average basicity of the phos- and the pyridine ionizable site. Generally speaking, the most

1

3

phonate groups does not significantly vary along the L –L

important variations were observed when the secondary

amines are protonated (i.e. a five membered ring can be

H

1

H

series within experimental errors (log Kav = 6.06 for L , log Kav

=

2

H

av

3

4

5

.51 for L and log K = 5.7 for L ), it is increased for L

log Kav = 6.63 for L ) which is lacking one secondary amine. gen bonding with N_p_y_r). The statistical processing

formed upon protonation of the amine and subsequent hydro-

H

4

33–35

(

of the

This feature reflects inter-arm hydrogen bond interactions that absorption and pH data allowed us to substantiate some of the

might exist between the protonated ammonium of one methyl- protonation constants which were previously determined by

amino-N,N-bis(methanephosphonate) moiety and the amine of the potentiometric method (Table 1). These protolytic data

1

4

the adjacent one.

confirm that L –L constitute intricate systems in terms of pro-

To assess the spectrophotometric properties of the polyami- tolysis and strongly suggest the occurrence of hydrogen bond

1

4

nophosphonate ligands L –L and of their protonated species networks between the pyridinyl unit and the side arms of the

and to confirm the pK values determined by the pHmetric ligands.

method, we also performed spectrophotometric titrations

Characterization of the cupric complexes with L –L . The

200 nm < λ < 500 nm) of the free ligands as a function of pH. stoichiometries of the cupric complexes formed with ligands

a

1

4

(

3

1

4

As a typical example, the case of L is presented in Fig. 2. L –L have been assessed by electrospray mass spectrometry

Regardless of the nature of the ligand, the pyridine constitutes (ESI-MS) in the positive mode. ESI-MS mass spectra were

the main chromophore unit and a structured absorption band recorded for aqueous solutions containing equimolar concen-

3

−1

1

centered at ∼260–270 nm (ε ∼ 3 × 10 M cm ), which can be trations of Cu(II) and the ligand (see Fig. 3 for L ). Regardless

3

2

ascribed to its π–π* transitions, was observed. Even though of the nature of the ligand, monocupric monochelates were

the pyridine nitrogen cannot be protonated in the pH range systematically detected which strongly suggests that the

considered, marked spectral variations (hyperchromic shifts) ligands considered in this work display appropriate denticity

were measured which can be explained by non-covalent inter- and pre-organization to fulfill the stereochemical requirements

actions (hydrogen bonding) between the protonated side arms of the Cu(II) ion. The ionization of the cupric complexes was

9604 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

1

Fig. 3 ESI mass spectra of the cupric complexes formed with ligand L .

1

−4

2

Solvent: H O; positive mode. (a) [Cu] = [L ]tot = 4.10 × 10 M.

mainly obtained by the addition of protons and sodium ions.

Similar results were obtained with Zn(II) and Ni(II). The

pseudo-molecular ions of the diﬀerent species observed are

collected in Table S1 (ESI†) and are in good agreement with

the simulated mono-isotopic masses.

1

4

Stability of the Cu(II) complexes with L –L . Similarly to the

free ligands, potentiometric (Fig. S1, ESI†) and absorption vs.

pH (Fig. 4) titrations have been carried out to evaluate the

1

4

stability constants of the cupric complexes with ligands L –L .

Fig. S1 (ESI†) first displays the potentiometric data of aqueous

solutions containing nearly equimolar concentrations of Cu(II)

Fig. 4 Absorption vs. pH titrations of the cupric complexes formed

2

3

2 4

with (a) L and (b) L . Solvent: H O; I = 0.1 M (NaClO ); T = 25.0(2) °C;

and of the ligand. These pHmetric data have been statistically l = 1 cm. (a) [L ]tot = 1.39 × 10 M; [Cu(II)]tot = 1.30 × 10⁻⁴M; (1) pH =

2

−4

2

1

.15; (2) pH = 9.19. (b) [L ]tot = [Cu(II)]tot ₌₁_.₇₈_×₁₀−4 M; (1) pH = 2.49;

3

4

processed and the corresponding stability and protonation

constants of the Cu(II) complexes are shown in Table 2. Simi-

larly to most of the polyaminomethanephosphonated chela-

(2) pH = 11.19.

1

4

tors described so far, ligands L –L basically form monocupric Table 2 Stability and protonation constants of the cupric complexes

1

4 a

formed with ligands L –L

monochelates. This binding feature is confirmed by the mass

1

4

spectrometric data conducted on L –L series (Table S1, ESI†).

log KCuLHx (3σ)

1

4

On the other hand, L –L form Cu(II) complexes whose high

stability constants are comparable, for instance, to EDTPA (log

1

2

3

4

Equilibrium

L

3

6,37

K

CuEDTPA = 23.2)

and are much higher than that of

KCuL

b

L þCu !CuL

22.71(9)

19.50(9)

19.1(3)

19.3(6)

3

8,39

N-methylamino-N,N-bis(methylenephosphonic acid)

constitutes the side arm model of L , L and L . Furthermore,

which

KCuLH

b

1

3

4

CuL þH !CuLH

8.24(8)

8

6.30(1)

7.8(3)_c

6.1(7)_c

c

.20(6)

7.7(5)

8.5(7)

KCuLH2

the substitution of the carboxylate functions of the (2,6-bis[bis-

! _C_u_L_H₂

b

CuLH þH

7.15(9)

4.71(9)

⁵^.⁸⁽³⁾c

⁴^.⁹⁽⁸⁾c

2

5

c

(

1

(carboxymethyl)amino)methyl]pyridine chelator (log K_C_u_L

=

6.9(5)

5.7(3)

4.3(3)

4.9(6)

KCuLH3

b

1

CuLH2 þH ! CuLH3

5.7(1)

—

5.69) by phosphonic units (L ) stabilizes considerably the

corresponding cupric complexes. Interestingly, L forms more CuLH3 þH

stable Cu(II) complexes (log K = 22.71(7)) than its closely

KCuLH4

!

1

b

^C^u^L^H4

4.0(2)

—

_C_u_L1

a

2

Solvent: H O; I = 0.1 M; T = 25.0(2) °C)]. The uncertainties (3σ) in the

2

4

3

4

related analogues L –L . Indeed, the lack of one (L , L ) or two log K_C_u_L_H_xvalues correspond to the added standard deviations (σ) in

2

n

the cumulative constants. KCuLHx = [CuLH ]/[[CuLH(n−1)][H]. Charges

b c

(

L ) phosphonic units is reflected by a destabilization of the

Cu(II) complexes of about 3 orders of magnitude (log K_C_u_L

9.5(9), log K_C_u_L= 19.1(3), log K_C_u_L= 19.3(6)). These data

have been omitted for the sake of clarity. Potentiometry. Absorption

2

=

vs. pH.

1

3

4

clearly demonstrate the importance of the four terminal phos-

phonate units of L for the binding strength with Cu(II).

1

3

mental conditions, while CuL can lead to three protonated

Regarding the protonation properties of the cupric com- species. Only two protonations were measured for the Cu(II)

1

4

1

2

4

plexes with L –L , it is noteworthy that the CuL species is able complexes with both L and L (Table 2). The protonation sites

to be successively protonated four times under our experi- of the cupric complexes most likely correspond to phospho-

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9605

View Article Online

Paper

Organic & Biomolecular Chemistry

3

3–35

nate units, which implies that the secondary amines remain of the phosphonic units. The statistical processing

of the

firmly bound to the Cu(II) cation whatever the pH considered spectral and pHmetric data sets allowed determining the pro-

in this study. The involvement of the pyridinyl nitrogen cannot tonation constants of the Cu(II) complexes, which were found

1

be excluded as well since the stability constant of CuL is by to be in good agreement with those previously measured by

far much higher than that of the Cu(II) complex of a closely the potentiometric method (Table 2).

related analogue for which the pyridine ring has been substi-

tuted by a phenyl one (log KCuL = 15.3). Other linear or cyclic the case of Ni(II), Zn(II) and Co(II). Using potentiometry, it was

Selectivity of Cu(II) complexation toward other metal ions:

4

0

analogues also bearing a pyridine unit displayed the same possible to accurately determine the stability constants

4

1,42

binding characteristic.

(Table 3) of the metallic complexes with the Zn(II), Ni(II) and

1

4

Lastly, if significant diﬀerences can be measured for the Co(II) cations with ligands L –L . Similarly to Cu(II), protonated

phosphonates pK s in the free ligands and in the corres- metal complexes were evidenced and their corresponding pro-

a

ponding complexes, the average phosphonate basicities tonation constants are depicted in the ESI† (Table S2 and

1

3

measured for L –L are almost unaﬀected by the Cu(II) com- Fig. S2 to S5). These data were compared to those available in

1

H

1

H

2

H

1

plexation (L log K = 6.06, CuL log K = 6.27; L log K =

av

5

the literature for reference chelators such as the L analogue

av

2

H

3

H

3

H

1

.51, CuL log Kav = 5.51, L log Kmoy = 5.7, CuL log Kav = 5.96). bearing terminal carboxylates (denoted L

C

in Table 3) and

This feature is in excellent agreement with the observations DO2P

(1,4,7,10-tetraazacyclododecane-1,7-bis(methanephos-

(EDTPA log K = 5.59, phonate)), which constitutes a golden standard and is one of

3

6,37

H

av

made for instance for EDTPA

H

4

CuEDTPA log KCuL = 5.57). L stands in interesting contrast the most eﬀective Cu(II) chelators described so far.

1

4

since it is lacking a secondary amine unit for Cu(II) binding. It

The complexation of Cu(II) by ligands L –L is selective com-

is then proposed that a phosphonate unit is firmly bound to pared to that of Zn(II), Ni(II) or Co(II). This binding selectivity

Cu(II), thus influencing both the number of Cu(II) protonated also exists for other polyaminomethanephosphonate ligands

1

2

complexes and the average protonation constant of the remain- such as DO2P. The sequence of LM stability constants

4

H

av

4

H

av

ing phosphonates (L log K = 6.63, CuL log K = 5.5). The (Table 3) is indeed in agreement with the Irving–Williams

distribution diagrams of the cupric complexes as a function of series which defines that the classical thermodynamic order

pH were calculated and are available in the ESI† (Fig. S2).

for transition-metal complexes is Mn(II) < Fe(II) < Co(II) < Ni(II)

4

8–50

Fig. 4 portrays the absorption spectral variation induced by < Cu(II) > Zn(II).

Cu(II) complexation as a function of the pH of the solution

containing L and L (Fig. S6–S9, ESI†). In addition to the pyri- stabilizes the Cu(II) complexes, while small diﬀerences are

dine-centered π–π* transitions at ∼260 nm, the cupric com- observed for the analogous ligands L –L . L is the most appro-

plexes with polyaminophosphonated ligands display priate system to stabilize Cu(II) in a distorted square planar

informative spectroscopic probes such as intense N→Cu(II) pyramidal geometry. Due to its versatility, the presence of a

For Cu(II), the presence of four phosphonate units clearly

2

3

2

3

1

4

3,44

charge transfer absorptions

as well as weaker Cu(II)-cen- square planar geometry cannot be excluded under acidic con-

4

5,46

tered d–d transitions.

ing several key events such as the Cu(II) complexation by L –L

and the protonation of the cupric complexes and provided under these experimental conditions. By contrast, L and L ,

These absorptions allowed monitor- ditions where the phosphonate units are protonated. Indeed,

1

4

1

L forms stable complexes for pH < 4 and competes with DO2P

3

4

1

3

information on the coordination geometry of the metal. L –L

are characterized by intense N→Cu(II) CT absorptions at about most stable complexes. This would indicate that these two

00–310 nm which clearly indicate that the two secondary ligands better fulfill the stereochemical preferences of Ni(II)

both bearing three phosphonate groups, form with Ni(II) the

3

1

2

1

amines are strongly involved in the complexation of the met- (octahedral geometry) with respect to L or L . L is seemingly

allic ion whatever be the pH considered (2.5–11). Sub-

4

sequently, L , which is lacking one secondary amine, displays

a

1

Table 3 Stability constants of metallic complexes (Cu(II), Zn(II), Ni(II)

a much weaker MLCT absorption. For CuL (4 phosphonates)

1

4

b

and Co(II) formed with ligands L –L and corresponding pM values

3

and CuL (3 phosphonates), the rise in the pH induces a bath-

a

ochromic shift of about 40–50 nm of the N→Cu(II) CT tran- log K_M_L(3σ)

b

1

2

3

4

1

C

pM (pH 7.4)

L

DO2P

sitions as well as the d–d absorption bands, while very weak

2

variations were observed for cupric complexes with L (2 phos-

log KCuL

22.71(7) 19.50(9) 19.1(3) 19.3(6) 15.69 ²⁸^.⁷c

15.46 12.89 12.08 13.06 12.07 17.77

16.50(3) 14.34(6) 18.2(2) 18.1(4)

11.79 9.54 13.08 13.65

17.84(4) 11.99(6) 17.2(2) 12.1(3) 15.84 21.2

4

II

phonates) and L (3 phosphonates, only one secondary N). pCu

log KNiL

—

This shift is probably associated with the weakening of one to

II

pNi

two Cu–N bonds, as observed for the thiophene containing

log KZnL

4

7

II

phosphonate ligand. Above pH 9, no spectral variation was pZn

12.37

16.50(9) 13.9(1)

11.22 8.86

8.93

11.89

nd

8.44

nd

13.68 11.73

log KCoL

—

observed for each of the systems. Regardless of the system con-

sidered, the binding of Cu(II) weakly influences the pyridine

π–π* transitions. The significant spectral variations experi-

II

pCo

KML

a

b

L þM ꢀ !ꢀ ML, KML = [ML]/[M][L]. pM = −log[M]free calculated for

−

6

−5

1

3

tot tot 2

[M] = 10 M and [L] = 10 M at pH 7.4. Solvent: H O; I = 0.1 M; T

c

enced by the tetra- (L ) and tris-phosphonated L ligands there-

=

25,0 °C. pM values calculated without the protonation constants of

fore likely reflect structural and geometrical modifications of

the metal complexes which have not been reported. nd = not

the Cu(II) coordination sphere influenced by the deprotonation determined.

9606 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

not an ideal binding platform for metal ions of strict coordi- Cu(II) above pH 7. Increasing the number of phosphonated

nation geometry such as Ni(II). The main diﬀerences between coordinating groups clearly favors the Cu(II) complexation under

Cu(II) and Ni(II)/Zn(II) can therefore be explained by the stereo- acidic conditions, which is of fundamental importance when

chemical preferences of these two cations.

To further discuss the complexation properties of L –L , we

anticipating PET imaging on cancer cells for instance.

1

4

Electrochemical properties. The electrochemical properties

5

1

64

calculated the pM values for each of these systems (Table 3). of a potential Cu chelate are among the most important

1

Fig. 5a depicts the pM values calculated for L with Cu(II), physico-chemical parameters to be considered for PET

Zn(II), Ni(II) and Co(II) and demonstrates the high selectivity of imaging applications. In vivo reduction of the cupric complex

1

2

4

L

for Cu(II) over a wide range of pH (2–12) (for L –L , see by biological reducing agents might indeed lead to unstable

Fig. S12 to S14, ESI†). The same feature is observed for ligands and extremely labile copper(I) species. Upon reduction, the

2

4

II

64

L –L . On the other hand, Fig. 5b illustrates the pCu values of

Cu(I) cation may undergo dissociation (by solvolysis, acidoly-

1

4

L –L compared to those of DO2P and EDTPA chosen as refer- sis or ligand exchange processes) and the radioactive cation

1

4

ences in this study. Ligands L –L display good aﬃnity for Cu(II) could then be widespread in the body with subsequent loss of

and the pCu values systematically increases with the pH of the the selectivity with respect to the target to be imaged and

II

solution until pH ∼ 7–8 (competition under basic conditions appearance of adverse health eﬀects. It was then of importance

with hydrolysis of the metals). The stability sequence at physio- to establish the electrochemical properties of the Cu(II) chelate

1

4

2

3

1

4

logical pH is as follows: DO2P > L > EDTPA-L > L > L . Inter- with L –L and to assess if they might be subjected to easy

1

estingly, L is among the most powerful Cu(II) chelator under reduction by biological reducing agents (i.e. E1/2 should be

acidic conditions and competes with DO2P which is ineﬃcient ≥−0.4 V per NHE). Cyclic voltamperograms (CV) were recorded

at pH < 4. In the pH range from 6 to 12, DO2P can be con- in water (0.1 M NaClO ) at diﬀerent pH values (3.7 ≤ pH ≤ 11,

4

sidered as the most eﬃcient Cu(II) chelator with respect to Fig. S10 and S11, ESI†). Fig. S10 (ESI†) displays the CVs

1

4

i

L –L . This study demonstrates that the presence of at least measured for ∼mM solutions of CuL (i = 1–4) at pH 3.7.

three phosphonated binding units is required to firmly bind Under these experimental conditions, the less stable cupric

complexes mainly co-exist under negatively multicharged

1

3−

1

2−

2

−

2

0

3

2−

states ([CuL H

3

]

/[CuL H

4

]

; [CuL H] /[CuL H

2

] ; [CuL H

2

]

/

3

−

4

2−

[

CuL H ] ; [CuL H ] ). For the sake of clarity, the CV of a mM

3

2

4 2

solution of Cu(ClO ) has also been recorded. Whatever the

system considered, we systematically observed the presence of

two irreversible cathodic signals as well as a so-called anodic

stripping peak (at ∼0.1 V and caused by redissolution of met-

0

52

allic copper) that is a clear signature of the formation of Cu .

0

The presence of Cu was further assessed by the formation of a

thin orange deposit on the surface of the glassy carbon elec-

3

4

trode. For L and L , an additional cathodic peak at −0.12 V

seemingly corresponds to reduction of free copper(II). The two

2+

i

+ i

cathodic peaks (Table 4) may be ascribed to the Cu L →Cu L

+

i

0

i

and Cu L →Cu + L (i = 1–4) reduction steps, respectively. The

presence of the redissolution peak at ∼0.1 V indeed suggests

that the electrogenerated Cu(I) and Cu(0) complexes derived

1

4

from ligands L –L are unstable and dissociate. This feature is

confirmed by the irreversibility of the two cathodic peaks

whose intensity and potential values are markedly influenced

5

3,54

by the potential scanning rate.

In addition, these electro-

chemical processes are not governed by diﬀusion. Taken

together, these data support a chemical reaction (dissociation

of the cuprous complexes, denoted C) that precedes an elec-

tron transfer (reduction of the cupric complexes, denoted E,

5

EC mechanism).

When the pH was raised (pH > 5.6), the CV pattern is mark-

edly simplified for all systems considered with the formation

of a quasi-reversible system whose redox potential is shifted at

more positive values. The quasi-reversibility is supported by

the i_p_a/i_p_cintensity ratios which are close to 1 and by the ΔE

values (Table 5). The quasi-reversibility persists even after

several scans whatever the potential scanning rate, meaning

that the electrogenerated Cu(I) complexes are stable enough on

II

Fig. 5 (a) Variation of the pM values (M = Cu , Zn , Ni and Co ) as a

1

II

1

function of pH for ligand L and (b) variation of the pCu values for L –

compared to those of DO2P and EDTPA. Solvent: H O; I = 0.1 M; T =

2

5.0(2) °C; pM = −log[M ]free or −log[M ]free, [L]tot ₌₁₀−5 M and [M ]tot

4

L

II

III

II

2

III

−6

or [M ]tot = 10 M.

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9607

View Article Online

Paper

Organic & Biomolecular Chemistry

Table 4 Electrochemical parameters measured for the Cu(II) complexes

1

4

with L –L in water at pH 3.7

0

Cu

st

nd

1

reduction

2

reduction

redissolution

peak

peak (Pc1)

peak (Pc2)

E

pc1

E

pc2

E

re

(V per

(

V per

i

pc1

(V per

i

pc2

I

re

Complex

SCE)

(μA)

SCE)

(μA)

SCE)

1

CuL

−0.51

−0.26

−0.39

−0.55

−66

−1.1

−286

−136

−389

−213

144

121

326

165

0.17

0.12

0.19

0.12

2

CuL

−45

−112

−206

−1.0

3

CuL

−1.13

−1.06

4

CuL

); T = 25.0(2) °C; v = 200 mV s⁻

1

.

Solvent: H

2

O; I = 0.1 M (NaClO

4

Reference = Ag/AgCl.

the CV timescale. These features are in good agreement with

the stronger stabilities of the cuprous complexes which

0

prevent Cu(II) and Cu(I) dissociation and thereby Cu for-

mation (the anodic redissolution peak is suppressed under

basic conditions).

To further assess the electrochemical stability of the cupric

1

4

complexes with L –L , we tested the resistance of these com-

plexes toward biological reductants such as ascorbic acid

0

(

DHA/Asc; E = +0.054 V per NHE at pH 7.0; intracellular con-

centration from 10 000 to 20 000 µM) or NADPH (NAD(P)/

0

NAD(P)H; E = −0.32 V per NHE at pH 7; intracellular concen-

tration from 200 to 500 µM) at pH 7.4 (tris buﬀer). The N–Cu(II)

CT absorptions constituted a valuable probe to evaluate the

reductant-decomposition of the cupric complexes (Fig. 6). As

i

Fig. 6 Spectrophotometric titrations of CuL (i = 1–4) complexes with

(

a) ascorbic acid and (b) NADPH. Solvent: aqueous Tris-HCl (0.05 M) at

pH 7.4, [CuL ]tot = 3 × 10 M; l = 1 cm; T = 25.0(2) °C.

i

−4

i

expected, the CuL complexes (i = 1–4) are more stable with

0

respect to ascorbic acid (E = +0.054 V per NHE) than to

0

2

NAD(P)H (E = −0.32 V per NHE). On the other hand, CuL

1

4

1

formation kinetic studies of the Cu(II) complexes with L –L

were first carried out using stopped flow spectrophotometry at

and to a lesser extent CuL appear to be the most resistant

cupric complexes in the presence of NADPH, in agreement

with their electrochemical properties (Fig. S10 and S11, ESI†).

Formation and dissociation mechanisms of the cupric

1

2−

2

0

3

−

pH ∼ 2. The complexes [CuL H

4

]

, [CuL H

2 3

] , [CuL H ] , and

4

2−

2+

[

CuL H ] are the major species at pH 2.0, while Cu is the

2

1

4

predominant free copper(II) species under these experimental

complexes with L –L

conditions of acidity (KCu(OH)+ = 10⁻

6.29

and KCu(OH)2 =

Formation kinetics. If a high thermodynamic stability of the

−

13.1 56

10

). We have chosen to monitor the formation kinetics

Cu(II) complex with a ligand is a prerequisite for its use in

2

+

6

4

under pseudo-first order conditions ([Cu ]tot > 10 × [L]tot) at

λ > 300 nm (i.e. N→Cu CT absorptions), which corresponds to

the largest spectrophotometric amplitudes between the

absorptions of the reactants (Cu + L –L ) and of the product

CuL (i = 1–4). Whatever the system considered, the absorbance

Cu PET imaging, it may also be associated with very slow

kinetics for the formation of the complex, as observed for

8

some cross bridged cyclen derivatives, and a fast complexa-

2

+

1

4

tion may be preferred to avoid a necessary heating step. The

i

always experienced an exponential growth versus time indicat-

ing a first order kinetics (v = −d[L]/dt = kobs[L]) with respect to

the ligand. The kinetic recordings revealed single rate limiting

steps in the millisecond to second time spans and, impor-

pa1/ipc1 tantly, spectral amplitudes were lost during the mixing time

Table 5 Electrochemical parameters measured for the Cu(II) complexes

with L –L in water at pH > 10

1

4

a

Complex

E

pc1 (V)

E

pa1 (V)

i

(

∼3 ms) of the stopped-flow device. The absorbances at the

1

CuL

−0.5

0.01

−0.26

0.25

0.29

0.83

0.54

1.24

0.94

2

end of the rate limiting steps satisfactorily correspond to the

expected values for the CuL complexes at pH ∼ 2, signifying

the absence of any slowest steps.

CuL

−0.5

i

3

CuL

−0.16

−0.07

4

CuL

a

₎_;_T₌₂₅_.₀₍₂₎_°_C_;_v₌₂₀₀_m_V_s−1

.

Formation kinetics of metal complexes and, in this particu-

Solvent: H

2

O; I = 0.1 M (NaClO

4

4,57,58

Reference = Ag/AgCl.

lar case, of copper(II) complexes

with a wide variety of

9608 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

2

+

ligands showed that these metal complexation processes on [Cu ]tot therefore satisfactorily obeys eqn (3), while a linear

5

9,60

2

usually follow a dissociative Eigen–Wilkins mechanism:

dependence was observed for L . Strikingly, no variation of the

k

2

+

4

obs values with [Cu ]tot was observed for L :

Kos

kex

LðSÞþMðsÞ!MLðsÞþS !MLþs

ð1Þ

2

þ

kCuL K

½Cu ꢂtot

½

CuLꢂ*

kobs ¼

ð3Þ

2

þ

The rate limiting step (kex) corresponds to the desolvation

of the metal in an outer-sphere complex formed in a fast pre-

equilibrium (K_o_s):

1 þK½CuLꢂ* ½Cu ꢂtot

These features can be rationalized by considering the first

step (lost during the mixing time) as a very fast pre-equili-

brium (formation of a kinetic Cu(II) intermediate [CuL]*) fol-

lowed by a rate limiting rearrangement (formation of the final

complex CuL) as described by eqn (4) and (5) (i.e. as described

above, the monomolecular rate constant could not be accu-

rately determined under these experimental conditions):

_k_f_¼_K_o_s_ꢁ_k_e_x_w_i_t_h_k_f_ð_M_ꢀ1

_ꢀ₁_Þ_;_K_o_s_ð_Mꢀ1_Þ_a_n_d_k_e_x_ð_sꢀ1Þð2Þ

s

6

1

Estimating the Kos values for the free ligands at pH ∼ 2

2

+

62–64

8

−1

and Cu and assuming the reported k_e_xvalue

(2 × 10 s

2

+

in water at 25 °C) for the desolvation of the Cu cation, we

f

evaluated the theoretical k values (Table 6).

K

; fast

CuLꢂ*

½

The large values of k agree well with the loss of spectral

_L_*_þ_C_u2þ ꢀꢀꢀꢀꢀꢀꢀꢀꢀ½CuLꢂ*

f

ꢀꢀꢀꢀꢀꢀꢀꢀꢀ!

ð4Þ

ð5Þ

amplitude during the mixing time of the stopped flow appar-

atus and suggest that the first steps likely correspond to the

formation of kinetic intermediates. The pseudo-first order rate

kCuL

ꢀ

ꢀꢀ*

½

CuLꢂ* )ꢀꢀꢀCuL

ꢀ1

k

−

1

CuL

constants kobs (s ) of the rate-limiting step were determined

2

+

2

4

for various Cu concentrations (Fig. 7). With the exception of

The kinetic behaviors for L and L correspond to extreme

4

L , the ordinates at the origin were found to be close to zero. situations depicted in eqn (6) and (7) that intimately depend

2

4

Under large excess of Cu(II), the formation of the cupric com- on the very low (L ) or very high (L ) stabilities of the kinetic

plexes is indeed favored with respect to the dissociation intermediate [CuL]*, respectively.

pathway, thus precluding reliable measurement of the dis-

4

ꢁ½Cu²^þ^ꢂtot ^ꢃ¹^t^h^eⁿ^kobs ^¼^k½CuLꢂ

−

1

For L : if K

ð6Þ

ð7Þ

sociation kinetic k CuL parameter. For L (4 phosphonate

½CuLꢂ*

3

units) and L (3 phosphonate units), the dependence of kobs

2

ꢁ½Cu²^þꢂ ꢄ1 then

For L : if K

½

CuLꢂ*

tot

2

þ

kobs ¼k_½_C_u_L_ꢂ

K

_½_C_u_L_ꢂ_*½Cu ꢂ_t_o_t

1

4

Table 6 Estimated Kos, kex and k

f

average values for ligands L –L in

4

water at pH 2

In the case of an intermediate complex such as [CuL ]*,

whose structure is close to the product, the kobs values are

independent of [Cu ] . The ligand L indeed displays three

phosphonated units and only one secondary amine. The stron-

ger implication of the binding units within the kinetic inter-

mediate complex [CuL ]* (i.e. the stability constant of the

1

(M⁻¹s⁻¹

)

Ligand

K

os (M⁻

)

k

f

2

+

4

tot

1

2−

8

L H

6

4

5

4

1.14

0.16

0.43

1.14

2 × 10

2

7

L H

3.2 × 10

8.6 × 10

3

−

L H

4

2−

8

L H

2 × 10

kinetic intermediate is expected to be close to that of the final

product) would explain the diﬀerence with respect to [CuL ]*

1

−1

3

(

=

log K_[

1

= 2.1(2)/k

_[_C_u_L1_]

= 59(9) s ) and [CuL ]* (log K

3

CuL ]*

[CuL ]*

1.5(2)/k[CuL

]

= 220(40) s⁻) for which a saturation kinetics

3

1

has been evidenced (i.e. a much larger diﬀerence between the

kinetic intermediate and the final complex is anticipated). In

2

the case of the L system (bearing 2 phosphonate units), the

2

much lower stabilities of the intermediate complexes [CuL ]*

1

3

compared to that of [CuL ]* (4 phosphonate units) and [CuL ]*

3 phosphonate units) enlighten their peculiar formation kine-

(

tics. Therefore, only the product k[CuL] × K[CuL]* could be evalu-

ated (Table 7). The forward (k_[_C_u_L_]) rate constants as well as the

stability constants of the kinetic intermediates are collected in

2

4

Table 7. The stability constants of [CuL ]* and [CuL ]* were

not determined under our experimental conditions and their

1

.5

−1

2

values have been estimated to be >10

M

for [CuL ]* and

4

∼log K[CuL

4]* for [CuL ]*.

Fig. 7 Variation of the pseudo-ﬁrst order rate constants kobs (s⁻) as a

1

Our kinetic data clearly demonstrated fast formation of

2

+

1

function of [Cu

]

tot for the formation of the cupric complexes with L –

1

4

1

kinetic Cu(II) intermediates for L –L followed by a rate-limit-

ing rearrangement to the final more stable cupric complexes.

According to the Eigen–Wilkins principle (Table 6), the for-

L . Solvent: H

2

O; T = 25.0(2) °C; l = 1 cm; I = 0.1 M (NaClO

4

); [L ]tot

=

−4

2

−4

3

−4

4

3.04 × 10 M, [L ]tot = 1.80 × 10 M, [L ]tot = 2.06 × 10 M and [L ] tot

=

1.63 × 10⁻M.

4

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9609

View Article Online

Paper

Organic & Biomolecular Chemistry

Table 7 Kinetic and thermodynamic constants related to the formation under our experimental conditions. The LMCT absorbance was

1

4

a

of the cupric complexes with ligands L –L at pH 2

monitored versus time for several concentrations of perchloric

+

1

acid in the range 0.02 M < [H ] < 0.5 M for CuL and 0.04 M <

k[CuL] (±σ)

k[CuL] × K[CuL]*

3 −1

+

i

b

−1 −1

(±σ) (×10 M_s−1

)

[H ] < 1 M for CuL (i = 2–4). In all cases, the absorbance vari-

Complexes log K[CuL]

log K[CuL]*

(M

s

)

ation could be fitted with a single exponential function

1

2

3

4

b

c

d

CuL

6.04

2.29

2.46

2.1(2)

≪1.5

59(9)

7.43

(

Fig. 8a) and thereby demonstrates that the rate-limiting dis-

d

c

≫220

3.32(5)

c

d

sociation step is first order with respect to the cupric com-

plexes. The corresponding pseudo first-order rate constants

1.5(2)

220(40)

6.96

b

c

2.6

∼log K[CuL

4

]

4.23(2)

nd

4

a

b

are reported in Fig. 8b. The dissociation kinetics of CuL (i.e.

Solvent: H

2

O; T = 25.0(2) °C; l = 1 cm; I = 0.1 M (NaClO

4

). Apparent

4

3

L lacks a secondary amine with respect to L ) was too fast to

be accurately measured under our experimental conditions.

The pseudo-first-order dissociation rate constants of the

stability constant recalculated from data available in Tables 1 and 2.

Measured values. Estimated or calculated values.

c

d

+

Cu(II) complexes linearly vary with [H ] with a non-zero intercept

mation of the cupric kinetic intermediates is extremely fast

and agrees with the large amplitudes lost during the dead

time of the stopped-flow apparatus (ca. 3 ms). With the excep-

at the origin. These kinetic data can be described by the follow-

−1 −1

ing relationship, with k

lysis rate constant and k

H

(M

s ) being the bimolecular acido-

₍_s−1) the monomolecular solvolysis

rate constant of the complexes with L –L (eqn (9); Table 8):

4

d

tion of L for which the stability constant of the kinetic inter-

1

4

mediate is assumed to be close to that of the final cupric

complex, we clearly demonstrated that the stability of the

kinetic intermediate and the monomolecular rearrangement

rate are intimately dependent on the number of phosphonate

_k_o_b_s_¼_k_H_ꢁ_½_Hþ_ꢂþkd

ð9Þ

tot

Our results show a significant decrease in the N→Cu(II) CT

units. The stepwise increase of the number of phosphonates absorbance that is in agreement with the fast dissociation of

1

3

for L –L stabilizes the kinetic intermediate and slows down

the rearrangement process. The k_[_C_u_L_]× K_[_C_u_L_]_*values are in

i

the same order of magnitude which suggests that the CuL (i =

1

–3) complexes are likely sharing similar rearrangement fea-

tures (the final coordination of the nitrogen binding units).

Dissociation kinetics of the Cu(II) complexes. In addition to

6

4

a high stability, Cu chelators have to also display strong

kinetic inertness with respect to demetallation in the main

biological compartments (e.g. blood plasma at pH 7.35–7.45,

bile at pH 7.4–7.7, etc.). This demetallation reaction can be

even more favored in the more acidic compartments (gastric

juices, pH 1.2–3) or by exchange processes with other specific

endogenous Cu(II) chelators (e.g. Cu(II) transporters in the

blood plasma and liver: metallothionein, ceruloplasmin, trans-

65

cuprein or albumin). For all these reasons, slow dissociation

6

4

or exchange reaction is an essential property for a Cu chela-

tor in order to prevent adverse eﬀects in the body and also

deterioration in image quality. The proton-assisted dis-

1

4

sociation kinetics of the cupric complexes formed with L –L

has therefore been carried out using a stopped-flow absorption

spectrophotometer and monitored at the maximum of the

N→Cu(II) CT transitions.

The global decomplexation reactions, under acidic con-

ditions, can be written as follows:

Â

Ã

Â

Ã

CuL¹

CuL²

CuL³

6ꢀ

2ꢀ

4ꢀ

þ6H^þ

!

L¹H6

2ꢀ

þCu

2þ

Ã

þ

L²H4 þCu

0

2þ

þ4H

Ã

ð8Þ

þ

L³H5 þCu

ꢀ

þ5H

Fig. 8 (a) Variation of the absorbance at 312 nm for the proton-assisted

dissociation of CuL . [CuL ]tot = 7 × 10 M; [H ]tot = 5 × 10 M. Initial

Ã

1

−5

+

−3

_C_u_L4

þ

_L4_H₄

2ꢀ

þ4H

þCu

−

1

pH = 10. (b) Variation of the pseudo-ﬁrst order rate constants kobs (s

)

+

as a function of [H ]tot for the proton-triggered dissociation of the

During the dissociation kinetics of the Cu(II) complexes, we

observed a large amplitude loss of the LMCT absorbance

during the dead time of the stopped flow device (∼3 ms). This

1

3

2

cupric complexes with L –L . Solvent: H O; T = 25.0(2) °C; l = 1 cm; I =

1

−5

2

−5

0

[

.1 M (NaClO

4

); [CuL ]tot = 7 × 10 M, [CuL ]tot = 5.67 × 10 M and

CuL ]tot ₌₆_.₈₄_×₁₀−5 M. The cupric complexes have been initially pre-

3

suggests the existence of very fast steps that are not accessible pared at pH ∼ 10.

9610 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

Table 8 Kinetic constants related to the proton-assisted dissociation of that is subjected to a dissociation rate limiting step. For the

1

3a

the cupric complexes with ligands L –L

solvolysis pathway, charge repulsion between the non-bound

and negatively charged phosphonates as well as structural con-

straints applied to the cupric intermediate complexes may

explain this peculiar kinetic behavior.

2

−1

k

d

(s⁻¹) (± σ)

Complex

k

H

(×10 M s⁻¹) (± σ)

1

CuL

3.0(2)

1.11(1)

2.8(3)

81(4)

0.3(2)

45(14)

2

CuL

3

CuL

Labeling and characterization of MTn12 antibody with L*

a

Solvent: H

2

4

O; T = 25.0(2) °C; I = 0.1 M (NaClO ).

Tenascin-C (TNC) is a glycoprotein highly expressed during

embryogenesis but largely restricted or absent in healthy adult

6

tissues. Its expression is regained in several pathological

the Cu–phosphonate bonds and suggests a rate-limiting step

related to the protonation of the bound secondary amines and

to the subsequent dissociation of the N–Cu coordination

bonds. The large signal amplitude lost during the dead time

of the stopped-flow apparatus indeed corresponds to succes-

sive fast protonation equilibria prior to the limiting dis-

6

7

situations including many human cancer types. Many anti-

bodies against TNC have been identified and developed, and

some of them are currently used in preclinical or clinical

6

8

trials. The MTn12 antibody is a rat monoclonal antibody of

the IgG1 type with both specificity and relatively high aﬃnity

6

9

1

for murine tenascin-C. Therefore this antibody was a good

candidate to test novel labeling strategies in preclinical murine

models of cancer, including breast cancer or insulinoma

sociation step. As an example of the L system, the protonated

1

2−

species [CuL H

4

]

(i.e. the 4 phosphonate units are proto-

nated) might correspond to the kinetic intermediate originat-

ing from fast protonation steps. The absorption electronic

7

0

models, in which tenascin-C is known to be highly expressed

in vivo.

1

4−

1

3−

1

2−

spectra of the [CuL H

2

]

, [CuL H

3

]

and [CuL H

4

]

species

Prior to labeling, a MALDI/MS analysis was carried out on

MTn12 which was found to have a molecular mass of

are similar and indicate that the involved protonations do not

alter the chromophore (i.e. LMCT). As an example, the

suggested mechanism is shown schematically in Fig. 9 for

CuL . The limiting step corresponds to the protonation of an

amino function and to the concomitant cleavage of the corres-

148 100 Da. The L* labeled MTn12 (L*-MTn12) was also ana-

1

lyzed by MALDI/MS (Fig. 11). The mass spectrum shows a shift

+

of the monocharged peak [M + H] to the higher masses due

to the addition of the BFC. At the maximum, this shift is of

about 1000 Da which would roughly correspond to two ligands

per antibody.

ponding N–Cu bond to aﬀord the unstable intermediate I

which then dissociates very rapidly by the attack of another

1

proton. It is noteworthy that the I kinetic intermediate has a

1

putative structure similar to that which can be anticipated for

the CuL system structure in excellent agreement with its very

4

Immunofluorescence staining

fast dissociation under very acidic conditions.

In order to verify that the labeling protocol did not perturb the

As far as acidolysis or solvolysis is concerned, the most immunoaﬃnity of the MTn12 antibody towards tenascin-C,

inert system unexpectedly corresponds to the cupric complexes immunofluorescence staining experiments were performed in

2

with L that possesses fewer phosphonate units (Fig. 10). For which MTn12 and L*-MTn12 were incubated on insulinoma

7

0,71

the acidolysis pathway, increasing the number of phosphonate tissue sections from RIP1-Tag2 mice.

groups may assist the proton transfer to the nitrogen atoms MTn12 antibody was then revealed by staining with an anti-rat

relay mechanism) and hence increase the dissociation rate k_H. secondary antibody labeled with a Cyanine3 dye (Cy3).

The number of phosphonate units (Fig. 10) indeed strongly A control experiment was performed in which no MTn12 was

The presence of the

(

influences the kinetic inertness of the I

1

intermediate complex added prior to the addition of the secondary antibody. As can

1

Fig. 9 Suggested proton-promoted dissociation mechanism of CuL .

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9611

View Article Online

Paper

Organic & Biomolecular Chemistry

be seen in Fig. 12, both MTn12 and its L* labeled version

could be detected in tumors of the pancreata sections, con-

firming that the labeling with L* did not aﬀect the immu-

noaﬃnity of MTn12.

99m

SPECT/CT imaging studies with

L , L and L*-MTn12

Tc complexes of ligands

1

2

1

2

99m

The chelating aﬃnity of ligands L and L towards

Tc was

first checked using radiochromatography experiments consist-

9

9m

ing in the comparison of the retention factors of

Tc in the

presence and absence of the ligands. The radiochemical purity

1

2

for L and L was above 95%. The biodistribution of the com-

plexes formed with the two ligands was then analyzed by

SPECT imaging of mice injected with the radioactive com-

H d

Fig. 10 Variation of the acidolysis (k ) and solvolysis (k ) rate constants

1

2

3

measured for CuL , CuL and CuL complexes as a function of the plexes. As can be seen in Fig. 13, the biodistribution profile

number of phosphonates.

was very diﬀerent in each case.

Fig. 11 (A) MALDI-TOF MS spectra of MTn12 (upper) and L*-MTn12 (lower). (B) Zoom of the m/z region from 142 000 to 155 000 showing the 1000

units shift in mass due to the added chelate.

9612 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

1

5,72

related with the phosphonate contents of the chelate.

In

2

contrast, 7 minutes after injection of the L complex, a large

part of the complex could be found in the kidneys, and it was

transferred to the bladder after 1 h 10 min. After 4 h, radio-

2

labeled L could still be found in the kidneys and the spleen,

but also at the joints (knees and shoulders). It is surmised that

2

the rapid clearance of the complex of L may be related to the

presence of the two ethyl chains replacing two phosphonate

moieties and giving the complex a more lipophilic character.

9

9m

Finally, the targeting eﬃciency of

Tc-L*-MTn12 was

questioned by injecting it into a transgenic mouse with breast

7

3

cancer (a transgenic MMTV-NeuNT breast cancer mouse). As

can be seen in Fig. 14, the radiolabeling was very rapidly

found to accumulate in the liver followed by its excretion

through the bladder. The same distribution could be observed

after 5 h 07 min and 22 h 50 min.

99m

In living mice, the rapid accumulation of the

Tc-L*-

MTn12 antibody in the liver may be explained by the fact that

IgG containing a Fc (constant domain) domain is known to be

primarily endocytosed and degraded by liver parenchymal,

endothelial and Kupﬀer cells thanks to the specific expression

7

4

of the cognate receptors by these liver cell types.

Conclusion

We have synthesized and characterized a new family of ligands

based on a pyridine scaﬀold functionalized with varying

numbers of phosphonated functions. The physico-chemical

properties of the ligands and their complexes have been

studied. Whatever the ligands, monometallic species were

observed by mass spectrometry for equimolar solutions of

metal and ligands. All the ligands showed a conventional

trend for complexation, following the Irving–Williams order of

4

8–50

1

2

aﬃnity.

Nevertheless, ligands L and L displayed a very

II

good selectivity for Cu(II) (Δlog KML > 5) with pCu values com-

patible with their potential use for Cu PET imaging. Interest-

ingly, the loss of a single phosphonate function compared to

6

4

1

the tetra-phosphonated ligand L immediately resulted in a

drastic loss of aﬃnity towards Cu(II) of three orders of magni-

tude. More surprising, the loss of a second phosphonate func-

2

tion to lead to L did not have such an impact on the aﬃnity.

These results remain poorly understood in the absence of a

clear definition of the coordination polyhedra oﬀered by the

ligands. Unfortunately, the presence of numerous protonated

forms of the Cu(II) complexes in water highly complicates the

Fig. 12 Immunoﬂuorescence staining on RIP1-Tag2 insulinoma tissue

sections using L*-MTn12 (a), MTn12 (b) or no secondary antibody as a

control (c). Antibody binding was revealed with a Cy3-coupled anti-rat

secondary antibody (red) and cell nuclei were stained with DAPI (blue).

1

2,47

interpretations of possible EPR measurements,

and was

not informative in our hands. Full electrochemical characteri-

1

In the case of L , the complex rapidly accumulates in the zation of the cupric complexes was performed, here again

1

2

kidneys after 5 minutes, and an important part of it can be showing the availability of the complexes of L and L to resist

found in the bladder after 2 h. But it could be noticed that a biological reduction, a necessary prerequisite for their use

large part of the complex accumulated in bone tissues of the in vivo. A thorough investigation of the formation and dis-

knees, shoulders, the nose and the spine, as can be observed sociation kinetics of the cupric complexes evidenced very rapid

after 4 h (Fig. 13). This behavior has already been observed for formations, while the proton assisted dissociations were pro-

9

9m

other phosphonate based

Tc complexes and is clearly portional to the number of phosphonate units.

This journal is © The Royal Society of Chemistry 2014

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9613

View Article Online

Paper

Organic & Biomolecular Chemistry

9

9m

1

2

Fig. 13 Top (left) and 100° (right) SPECT/CT images of mice injected with

Tc complexes of L and L , after 4 h 16 and 4 h 30 respectively.

Considering their potential interest for in vivo imaging, the recorded on an Avance 300 spectrometer at 300 MHz for

1

2

99m

31

ability of ligands L and L to coordinate

Tc were studied proton frequency and 75 MHz for carbon frequency. P NMR

by radiochromatography and by SPECT/CT imaging on healthy spectra (161.9 MHz) were recorded on Avance 400 apparatus

9

9m

1

mice, showing a preferred fixation of the

TcL complex on with H-decoupling. Chemical shifts are given in parts per

2

75

bone tissues, while the complex of L was rapidly excreted million, relative to the residual protic solvent peak. IR

through the kidneys and bladder. Finally, a bifunctional spectra were recorded on a Nicolet 380 FT-IR spectrometer

1

version of L was used to label MTn12, a rat monoclonal anti- (Thermo Scientific). Mass spectra (MS) and elemental analyses

body directed against murine tenascin-C. Although immuno- were obtained by the Service Commun d’Analyse de l’UdS.

fluorescence staining experiments revealed unperturbed

Solvents were used as received. Commercially available

9

9m

selectivity in tumors of the pancreata, the injection of Tc reagents were used without further purification.

radiolabeled antibody into a transgenic mouse bearing a

19

2,6-Bis(ethylaminomethyl)pyridine 3 and tetra-ethyl imino-

2

0

breast primary cancer showed the antibody to be readily accu- bis(methanephosphonate) were synthesized according to the

mulated in the liver, followed by excretion, with no specific literature procedures.

signal in the tumor. A possible explanation is that the constant

2,6-Dibromomethylpyridine 2. To a solution of 2,6-dihydroxy-

fragments of this IgG antibody were primarily endocytosed by methylpyridine (2.80 g, 20.1 mmol) in DMF (15 mL) was added

7

4

liver cells and our current eﬀorts are directed towards the dropwise, at 0 °C, PBr

3

(4.35 mL, 46.3 mmol). The mixture was

development of F(ab)′₂fragments of MTn12 (that lacks the stirred at room temperature for 4 h and water was added

constant fragments) for their labeling and in vivo imaging.

(20 mL). After extraction with Et

2

O (3 × 100 mL), the organic

layer was dried with Na SO and the solvent was removed

2

4

under reduced pressure. 2,6-Dibromomethylpyridine 2 was

obtained as a pure white solid (4.81 g, 95%). All analyses corre-

spond to those reported in the literature.

Tetraethyl(((pyridine-2,6-diyl-bis(methylene))-bis(ethylazane-

Experimental section

Synthesis of the compounds

7

6

General methods. Column chromatography and flash diyl))-bis(methylene))-bis (phosphonate) 4. Procedure 1: A

1

9

column chromatography were performed respectively on silica mixture of 2,6-bis(ethylaminomethyl)pyridine 3 (170 mg,

0.063–0.200 mm, Merck) and silica gel (40–63 μm, Merck). 0.88 mmol) and diethyl phosphite (243 mg, 1.76 mmol) was

Solvent mixtures used for TLC and flash column chromato- cooled for 10 minutes in an ice-bath and 220 µL of form-

(

1

13

graphy are reported in v/v ratios. H and C NMR spectra were aldehyde (37% in water) were added slowly maintaining the

9614 | Org. Biomol. Chem., 2014, 12, 9601–9620

This journal is © The Royal Society of Chemistry 2014

View Article Online

Organic & Biomolecular Chemistry

Paper

9

9m

Fig. 14 SPECT/CT images of a transgenic mouse with breast primary cancer, 5 minutes (left) and 25 minutes (right) after injection of

Tc-L*-

MTn12.

temperature below 10 °C. The ice bath was removed and the removed in vacuo and the crude oil was purified by flash

mixture was stirred at room temperature for 30 minutes and chromatography (CH Cl /MeOH). Compound 4 was obtained

2

refluxed for 3 h. The solution was concentrated under vacuum as an oily light yellow product (878 mg, 24%).

and the crude product was purified by flash chromatography

(((Pyridine-2,6-diyl-bis(methylene))-bis(ethylazanediyl))-bis-

2

(

CH Cl /MeOH 100/0 to 92/8) to give compound 4 as an oil (methylene))diphosphonic acid L . To a solution of com-

2

1

3

302 mg, 69%). H NMR (CDCl

3

, 300 MHz): δ 7.59 (t, JH–H

= 7.6 Hz, 2H, H_p_y_r), 4.09 (qd, (1.6 mL, 12.0 mmol) was added and the resulting mixture was

=

2 2

pound 4 (295 mg, 0.60 mmol) in CH Cl (3 mL), TMSBr

3

7

.7 Hz, 1H, H_p_y_r), 7.37 (d, J

H–H

3

J_H_–_H= 7.4 Hz, J

= 7.4 Hz, 8H, CH O), 3.85 (s, 4H, NCH - stirred at room temperature for 24 h. TMSBr (1.6 mL,

H–P

2

Pyr), 2.93 (d,

JH–P = 10.5 Hz, 4H, (EtO)

2

OPCH

2

N), 2.70 (q, 12.0 mmol) was added and the mixture was stirred at room

3

J_H_–_H= 7.1 Hz, 4H, CH CH N), 1.28 (t, J

= 7.1 Hz, 12H, temperature for 24 h. The solvent was removed under reduced

3

2

H–H

3

13

CH CH O), 1.03 (t, J

= 7.1 Hz, 6H, CH CH N). C NMR pressure, the resulting residue was taken up with methanol

3 2

3

2

H–H

(

4

8

CDCl

3

, 75 MHz): δ 158.8 (Cpyr), 136.7 (Cpyr), 121.2 (Cpyr), 61.9 (5 mL) and the mixture was stirred at room temperature for

2

3

d, J

= 6.8 Hz, CH O), 61.2 (d, J

= 8.8 Hz, NCH -Pyr), 2 h. The evaporation–dissolution procedure was repeated with

C–P

2

C–P

2

1

3

9.4 (d, J_C_–_P= 163.1 Hz, (EtO) OPCH N), 49.5 (d, J_C_–_P

.8 Hz, CH CH N), 16.6 (d, JC–P = 6.1 Hz, CH CH O), 12.0 ture for 24 h. The solvent was evaporated under reduced

3 2 3 2

=

methanol (5 mL) and the solution was stirred at room tempera-

2

3

1

+

(

CH CH N). P NMR (CDCl , 161.9 MHz): δ 25.6 . ESI /MS pressure and the residue was dissolved in a minimum of

3

2

3

+

MeOH): m/z

=

493.2 ([M

+

H] , 100%). Calc. for methanol. Addition of diethyl ether resulted in the formation

C

21

H

41

N

3

O

6

P

2

·0.5 H O: C, 50.19; H, 8.42; N, 8.36; Found: C, of a precipitate, which was collected by centrifugation and

2

1

5

0.11; H, 8.45; N, 8.52.

dried under vacuum to aﬀord ligand L in 83% (240 mg). H

3

Procedure 2: To a mixture of 2,6-dibromomethylpyridine 2 NMR (D O, 200 MHz): δ 7.90 (t, J

= 7.7 Hz, 1H, H_p_y_r), 7.44

H–H

2

3

(

2 g, 7.55 mmol) and amine 7 (3.1 g, 15.85 mmol) in MeCN (d, JH–H = 8.0 Hz, 2H, Hpyr), 4.63 (s, 4H, NCH

22 mL), under a nitrogen atmosphere, freshly flame-dried

2

-Pyr), 3.43 (q,

3

2

J_H_–_H= 7.3 Hz, 4H, CH CH N), 3.34 (d, J

= 12.4 Hz, 4H,

3

2

H–P

3

13

K CO (4 g, 30.2 mmol) was added. The reaction mixture was HO PCH N), 1.30 (t, J

2

= 7.1 Hz, 6H, CH CH N). C NMR

3 2

3

2

H–H

heated at 50 °C overnight. After filtration, the solvent was (D

2

O, 75 MHz): δ 149.8 (Cpyr), 139.9 (Cpyr), 124.2 (Cpyr), 57.6 (d,

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9615

View Article Online

Paper

Organic & Biomolecular Chemistry

3

J

C–P = 4.5 Hz, NCH

2

-Pyr), 51.6 (d, JC–P = 3.8 Hz, CH

3

CH

2

N), 155.9 (Cpyr), 137.4 (Cpyr), 122.9 (Cpyr), 121.9 (Cpyr), 62.0 (m,

1

3

4

9.4 (d, J_C_–_P= 136.5 Hz, HO PCH N), 8.2 (CH CH N). NCH -Pyr, CH O), 50.3 (dd, J

= 157.9 Hz, J

= 8.2 Hz,

3

2

3

2

C–P

3

1

−

31

P NMR (D

80.1 ([M – H] , 100%). Calc. for C13

2.51; H, 5.88; N, 8.75; Found: C, 31.92; H, 5.35; N, 8.34.

Diethyl-((benzyl(ethyl)amino)methyl)phosphonate 6. To a 100%), 525.1 ([M + Na] , 100%).

mixture of N-ethylbenzylamine (5 mL, 33.6 mmol) and diethyl- Hexaethyl-(((pyridine-2,6-diyl-bis(methylene))(ethylazanediyl))-

2

O, 161.9 MHz): δ 8.4. ESI /MS (MeOH/H

2

O): m/z = (EtO)

O·HBr: C, (CDCl

([M + H] , 14%), 504.8 ([M + H] , 14%), 523.1 ([M + Na] ,

2

OPCH

2

N), 34.0 (CH

2

Br), 16.5 (m, CH

3

CH

2

O). P NMR

−

+

3

H

25

N

3

O

6

P

2

·H

2

3

, 161.9 MHz): δ 24.4. ESI /MS (CH

2

Cl ): m/z = 502.8

+

phosphite (5.7 mL, 43.6 mmol) at 0 °C, 6 mL (67.2 mmol) of tris(methylene))-tris(phosphonate) 9. To a mixture of com-

aqueous 37% formaldehyde was added. The resulting mixture pound 8 (2.81 g, 5.6 mmol) and amine 7 (1.31 g, 6.7 mmol) in

was warmed to room temperature and stirring was continued MeCN under a nitrogen atmosphere, freshly flame-dried

at room temperature for 3 h. The reaction mixture was then K CO (1.55 g, 11.2 mmol) was added. The reaction mixture

2

3

heated under reflux for 3 h. The solvents were removed under was heated at 80 °C for 5 h. After filtration, the solvent was

reduced pressure and the resulting residue was purified by removed in vacuo and the crude oil was purified by flash

flash chromatography (CH Cl /MeOH 97/3) to give compound chromatography (CH Cl /MeOH 96/4). Compound 9 was

2

1

6

3

as a colorless oil (7.95 g; 83%). H NMR (CDCl , 300 MHz): obtained as an oily yellow product (1.81 g, 49%). H NMR

δ 7.24–7.38 (m, 5H, HArom), 4.13 (qd, JH–H = 6.8 Hz, JH–P = 6.8 (CDCl

Hz, 4H, CH O), 3.77 (s, 2H, CH -Ph), 2.90 (d, J

N), 2.71 (q,

.33 (t, JH–H = 6.9 Hz, 6H, CH

H, CH CH N). C NMR (CDCl , 75 MHz): δ 138.3 (C_A_r_o_m), 7.0 Hz, 2H, CH CH N), 1.20 (t, J

28.3 (CArom), 127.5 (CArom), 126.4 (CArom), 61.1 (d,

3

, 300 MHz): δ 7.54 (t, JH–H = 7.5 Hz, 1H, Hpyr), 7.34 (d,

J_H_–_H= 8.2 Hz, 2H, H_p_y_r), 4.08–3.97 (m, 12H, CH O), 3.79 (s,

N), 4H, NCH -Pyr), 3.13 (d, JH–P = 9.2 Hz, 2H, (EtO) OPCH NEt),

2 2 2 2

O), 1.08 (t, JH–H = 7.0 Hz, 2.87 (d, JH–P = 10.8 Hz, 4H, [(EtO)

2

3

= 10.6 Hz,

2

H–P

2

3

2

1

3

1

6

8

1

H, (EtO)

2

OPCH

2

J

H–H = 6.8 Hz, 2H, CH

3

CH

3

2

3

CH

2

OPCH

2

]

2

N), 2.62 (q, JH–H

= 6.7 Hz, 18H, CH CH O),

=

1

3

2

3

2

H–H

3

2

3

13

J

C–P

=

3 2 3

0.96 (t, JH–H = 7.3 Hz, 3H, CH CH N). C NMR (CDCl ,

75 MHz): δ 158.9 (Cpyr), 157.7 (Cpyr), 136.9 (Cpyr), 121.8 (Cpyr),

3

.5 Hz, CH

2

O), 58.6 (d, JC–P = 8.8 Hz, CH

.1 Hz, CH CH N), 48.0 (d, J

2

-Ph), 48.1 (d, JC–P

1

3

2

= 161.9 Hz, (EtO) OPCH N), 121.6 (C_p_y_r), 62.6 (t, J

= 8.0 Hz, NCH -Pyr), 62.0 (d, J

=

3

2

C–P

2

C–P

2

C–P

3

31

2

3

5.9 (d, JC–P = 5.6 Hz, CH

, 161.9 MHz): δ 25.8. ESI /MS: m/z = 285.2 ([M] , 20%). 8.6 Hz, NCH

Calc. for C H NO P·H O: C, 55.43; H, 8.64; N, 4.62; Found: [(EtO) OPCH ] N), 49.4 (d, J

3

CH

2

+

O), 11.2 (CH

3

CH

2

N). P NMR 6.5 Hz, CH

2

O), 62.0 (d, JC–P = 8.6 Hz, CH

2

-Pyr), 50.2 (dd, JC–P = 158.4 Hz, JC–P = 8.7 Hz,

2

O), 61.0 (d, JC–P

+•

1

3

(CDCl

3

= 8.9 Hz, CH CH N), 49.3 (d,

NEt), 16.5 (m, CH CH O), 11.9

2 3 2

1

4

24

3

2

2 2

C–P

3

2

1

C, 54.74; H, 8.35; N, 4.91.

J

C–P = 163.3 Hz, (EtO)

2

OPCH

3

1

+

Diethyl-((ethylamino)methyl)phosphonate 7. A solution of 6 (CH

9 g, 31.5 mmol) in ethanol (100 mL) was hydrogenated in the MS (CH Cl ): m/z = 616.3 ([M + H] , 100%). Calc. for

3

CH

2

N). P NMR (CDCl , 161.9 MHz): δ 24.4, 25.4. ESI /

3

+

(

2

presence of Pd/C catalyst (0.9 g, 10% Pd). After 8 h at reflux,

C

24

H

48

N

3

O

9

P

3

·2H

2

O: C, 44.24; H, 8.04; N, 6.45; Found: C,

the reaction was complete. The catalyst was filtered oﬀ and the 43.88; H, 7.79; N, 6.62.

filtrate was evaporated to give compound 7 as a pale yellow oil (((Pyridine-2,6-diyl-bis(methylene))(ethylazanediyl))-tris-

1

3

(

6 g, 98%). H NMR (CDCl

3

, 300 MHz): δ 4.13 (m, 4H, CH

2

O), (methylene))triphosphonic acid L . To a solution of com-

N), 2.73 (q, JH–H pound 9 (1.81 g, 2.93 mmol) in CH Cl (10 mL), TMSBr

= 6.8 Hz, 6H, CH CH O), (11.62 mL, 88 mmol) was added at 0 °C, and the resulting

2

3

2

6

1

7

.99 (d, JH–P = 12.1 Hz, 2H, (EtO)

.8 Hz, 2H, CH CH N), 1.31 (t, J

H–H = 7.0 Hz, 3H, CH

5 MHz): δ 61.9 (d, JC–P = 7.2 Hz, CH

2

OPCH

2

=

2

3

2

H–H

3

2

3

13

.10 (t,

J

3

CH

2

N). C NMR (CDCl

3

,

mixture was stirred at room temperature for 24 h. TMSBr

O), 45.3 (d, JC–P = 16.3 (11.6 mL, 88 mmol) was added and the mixture was stirred at

= 153.8 Hz, (EtO) OPCH N), 16.3 room temperature for 24 h. The solvent was removed, the

2

3

2

1

Hz, CH CH N), 43.6 (d, J

3

2

C–P

2

31

3

(

d,

CDCl

[M + H] , 42%), 413.2 ([2M + Na] , 100%).

J

C–P = 5.4 Hz, CH

3

CH

2

O), 14.6 (CH

3

CH

2

N). P NMR resulting residue was taken up with methanol (10 mL) and the

): m/z = 196.1 mixture was stirred at room temperature for 2 h. The evapor-

ation–dissolution procedure was repeated with methanol

+

3

, 161.9 MHz): δ 25.9. ESI /MS (CH

2

Cl

2

+

Tetraethyl-((((6-(bromomethyl)pyridin-2yl)methyl)azanediyl)- (10 mL) and the solution was stirred at room temperature for

bis(methylene))-bis(phosphonate) 8. To a mixture of 2,6- 24 h. The solvent was evaporated under reduced pressure and

dibromomethylpyridine 2 (3.2 g, 12.1 mmol) and tetra-ethyl the residue was dissolved in a minimum of methanol.

2

0

iminobis(methanephosphonate)

(3.07 g, 9.66 mmol) in Addition of diethyl ether resulted in the formation of a precipi-

MeCN under a nitrogen atmosphere, freshly flame-dried tate, which was collected by centrifugation and dried under

3

1

K CO (2.51 g, 18.15 mmol) was added. The reaction mixture vacuum to aﬀord ligand L (1.71 g, 90%). H NMR (D O +

2

3

2

3

was heated at 50 °C overnight. After filtration, the solvent was NaOD, 200 MHz): δ 7.68 (t, JH–H = 7.4 Hz, 1H, Hpyr), 7.44 (d,

3

removed in vacuo and the crude oil was purified by flash

JH–H = 7.4 Hz, 1H, Hpyr), 7.29 (d, JH–H = 6.7 Hz, 1H, Hpyr), 3.90

chromatography (CH Cl /MeOH 100/0 to 92/8). Compound 8 (s, 2H, NCH -Pyr), 3.84 (s, 2H, NCH -Pyr), 2.54 (m, 8H,

2

1

3

was obtained as an oily light yellow product (2.81 g, 58%). H

O

3

PCH

2

N, CH

C NMR (D O, 75 MHz): δ 159.0 (Cpyr), 158.2 (Cpyr), 138.0

= 8.1 Hz, 1H, (C_p_y_r), 123.0 (C_p_y_r), 122.6 (C_p_y_r), 62.5 (m, NCH -Pyr), 60.2 (d,

3 2 3 2

CH N), 0.85 (t, JH–H = 7.4 Hz, 3H, CH CH N).

3

13

NMR (CDCl

3

, 300 MHz): δ 7.66 (t, JH–H = 7.8 Hz, 1H, Hpyr),

= 8.1 Hz, 1H, H_p_y_r), 7.31 (d, J

H–H H–H

2

3

7

.51 (d, J

2

3

1

3

H

(

pyr), 4.51 (s, 2H, CH

2

Br), 4.11 (m, 10H, NCH

2

-Pyr, CH

2

O), 3.23

2

JC–P = 6.0 Hz, NCH -Pyr), 55.4 (dd, JC–P = 145.9 Hz, JC–P = 8.8

2

3

1

d, JH–P = 9.9 Hz, 4H, (EtO)

2H, CH CH O). C NMR (CDCl , 75 MHz): δ 158.7 (C_p_y_r), 48.1 (d, J

2

OPCH N), 1.29 (t, JH–H = 7.0 Hz, Hz, [HO

2

3

PCH

2 2 3 2

] N), 52.6 (d, JC–P = 138.9 Hz, HO PCH NEt),

1

3

31

1

= 8.0 Hz, CH CH N), 10.5 (CH CH N). P NMR

3

2

3

C–P

3

2

3

2

9616 | Org. Biomol. Chem., 2014, 12, 9601–9620

View Article Online

Organic & Biomolecular Chemistry

Paper

−

2

(

2 2 2

D O, 161.9 MHz): δ 17.1, 16.5. ESI /MS (D O): m/z = 446.1 1H, Hpyr), 4.71 (s, 2H, NCH -Pyr), 3.51 (d, JH–P = 21.2 Hz, 2H,

−

2

13

[M – H] , 100%). Calc. for C H N O P ·2H O·2HBr: C, 22.34; HO PCH ), 3.35 (d, J_H_–_P= 12.0 Hz, 4H, [HO PCH ] N).

C

1

2

24

3

9

3

2

3

2

3

2 2

H, 4.69; N, 6.51; Found: C, 22.80; H, 4.81; N, 6.24.

2

NMR (D O, 100 MHz): δ 151.9 (Cpyr), 148.5 (Cpyr), 146.2 (Cpyr),

Diethyl-(((6-bromomethyl)pyridine-2-yl)methyl)phosphonate 127.7 (d, JC–P = 4.3 Hz, Cpyr), 125.7 (Cpyr), 56.6 (m, NCH

2

-Pyr),

1

3

1

0. Compound 2 (480 mg, 1.81 mmol) was dissolved in 52.6 (dd, J

= 144.7 Hz, J

C–P C–P

= 5.8 Hz, [HO PCH ] N), 34.8

3

2 2

1

31

triethyl phosphite (315 μL, 1.81 mmol) under a nitrogen (d, JC–P = 123.4 Hz, HO

3

PCH

2

). P NMR (D

2

O, 161.9 MHz): δ

−

atmosphere. The mixture was heated at 50 °C, protected from 13.7. ESI /MS (D

2

O): m/z = 389.0 ([M − H] , 100%). Calc. for

light, for 4 days. After cooling at room temperature, excess C H N O P ·HCl: C, 25.34; H, 4.25; N, 6.56; Found: C, 25.07;

9

17 2 9 3

triethyl phosphite was removed under high vacuum. The crude H, 4.32; N, 6.40.

product was purified by flash chromatography (CH Cl /MeOH

4/6). Compound 10 was isolated as an orange oil (281 mg,

2

MALDI-MS analyses

9

4

1

8%). Compound 10 is not stable and is rapidly degraded. H Prior to analysis, the MTn12 antibodies, either uncoupled or

3

NMR (CDCl

3

, 300 MHz): δ 7.55 (t, JH–H = 7.8 Hz, 1H, Hpyr), coupled to the BFC chelate, were desalted on a C18 ZipTip

= 7.8 Hz, 2H, H ), 4.42 (s, 2H, CH Br), 3.98 (qd, (Millipore). Ten µg of antibody were loaded and then eluted in

O), 3.30 (d, JH–P = 21.6 2 μL MeCN 80%/TFA 0.1%. The whole eluate was spotted on

), 1.20 (t, JH–H = 7.0 Hz, 6H, CH CH O). the target plate with 0.6 μL of sinapinic acid (saturated solu-

C NMR (CDCl , 75 MHz): δ 155.7 (d, J_C_–_P= 2.7 Hz, C_p_y_r), tion diluted 3 times in MeCN 50%/TFA 0.1%). The mass

3

7

J

.21 (d, J

H–H

pyr

2

3

2

H–H = 7.1 Hz, JH–P = 7.1 Hz, 4H, CH

2

3

Hz, 2H, (EtO)

2

OPCH

2

3

2

1

3

1

52.0 (d, JC–P = 8.3 Hz, Cpyr), 136.7 (d, JC–P = 2.2 Hz, Cpyr), 122.8 spectrometer (Autoflex, Bruker Daltonics) was operated in posi-

(

d, JC–P = 5.0 Hz, Cpyr), 120.9 (d, JC–P = 2.6 Hz, Cpyr), 61.4 (d, tive linear mode, with BSA as a calibrant.

2

1

J_C_–_P= 6.5 Hz, CH O), 35.7 (d, J

= 134.3 Hz, (EtO) OPCH ),

Production and purification of MTn12 antibody. The

2

C–P

2

3

31

69

3

3.2 (CH

2

Br), 15.6 (d,

JC–P = 5.7 Hz, CH

3

CH

2

O). P NMR MTn12 anti-tenascin-C mouse antibody was purified from

(CDCl

3

, 161.9 MHz): δ 24.4.

hybridoma culture supernatants by aﬃnity chromatography

Tetraethyl-((((6-(diethylmethylenephosphonate)pyridin-2yl)- using a HiTrap protein G sepharose column and an AktaPrime

methyl)azanediyl)-bis(methylene))-bis(phosphonate) 11. To a purification system (GE Healthcare).

mixture of compound 10 (800 mg, 2.5 mmol) and amine 7 The labeling experiment was carried out on 500 μg (500 μL

790 mg, 2.5 mmol) in MeCN under a nitrogen atmosphere, of a 1 mg mL⁻solution in PBS) of the antibody MTn12 to

freshly flame-dried K CO (684 mg, 4.95 mmol) was added. which 2 mg of solid BFC L* (2.7 μmol) was added, corres-

1

(

2

3

The reaction mixture was heated at 80 °C, protected from light, ponding to 800 equivalents of the ligand for the antibody. The

for 48 h. After filtration, the solvent was removed in vacuo and reaction medium was adjusted to a pH of 7.4 with 200 mM

the crude product was purified by flash chromatography NH

CH Cl /MeOH 100/0 to 96/4). Compound 11 was obtained as reaction medium was purified repeatedly to eliminate the

an oily product (476 mg, 32%). H NMR (CDCl , 300 MHz): δ excess of ligand with PBS using Vivaspin500 (Sartorius)

4 3

HCO and was stirred for 2 h at room temperature. The

(

2

1

3

7

H

.60 (t, JH–H = 8.4 Hz, 1H, Hpyr), 7.45 (d, JH–H = 7.9 Hz, 1H, modules with a 30 kDa cut-oﬀ (500 μL of PBS per cycle; 20

3

pyr), 7.25 (d, JH–H = 7.5 Hz, 1H, Hpyr), 4.07 (m, 14H, CH

2

O, cycles).

= 22.0 Hz, 2H, (EtO) OPCH ), 3.20 (d, Immunostaining experiment. Immunofluorescence analysis

2 2 2

H–P = 10.4 Hz, 4H, [(EtO) OPCH ] N), 1.28 (t, JH–H = 7.1 Hz, of murine insulinoma tumors was performed by staining pan-

2

NCH -Pyr), 3.36 (d, J

2

H–P

2

3

J

3

13

70

1

2H, CH

3

CH

2

O), 1.24 (t, JH–H = 7.1 Hz, 6H, CH

3

CH

2

O).

C

creata sections from RIP1-Tag2 mice as previously described.

NMR (CDCl , 75 MHz): δ 158.4 (d, J

= 2.1 Hz, C_p_y_r), 151.9 (d, The MTn12 antibody, either coupled to the BFC chelate or

3

C–P

J

4

C–P = 8.2 Hz, Cpyr), 137.0 (d, JC–P = 2.4 Hz, Cpyr), 122.8 (d, JC–P

.8 Hz, Cpyr), 121.8 (d, JC–P = 3.3 Hz, Cpyr), 62.5 (t, JC–P = 7.9 (2 µg mL ), and a Cy3-coupled anti-rat secondary antibody

=

uncoupled, was incubated overnight at 4 °C on tissue sections

3

−1

2

Hz, NCH -Pyr), 62.2 (d, J

= 6.6 Hz, CH O), 62.1 (m, CH O), (1/1600 for 1 h at room temperature; Interchim) was used for

2

C–P

2

1

3

5

3

0.4 (dd, JC–P = 159.2 Hz, JC–P = 8.3 Hz, [(EtO)

6.6 (d, JC–P = 135.1 Hz, (EtO)

P NMR (CDCl , 161.9 MHz): δ 25.1, 24.4. ESI /MS (CH Cl ):

2

OPCH

2

]

N), detection. Cell nuclei were stained with DAPI. Negative con-

O). trols were performed by omitting the primary antibody.

1

2

OPCH

2

), 16.5 (m, CH CH

3

1

+

SPECT animal imaging. All work performed on animals was

3

2

+

m/z = 559.2 ([M + H] , 100%). Calc. for C21

4

H

41

N

2

O

9

P

3

·2H

2

O: C, in accordance with the local ethics committee following the

rules (agreement number AL/05/12/02/13 of the regional

2.43; H, 7.63; N, 4.71; Found: C, 42.90; H, 7.03; N, 5.06.

9

9m

(

(((6-(Methylene)pyridin-2yl)methyl)azanediyl)-bis(methyl- ethical committee). Biodistribution of the

Tc complexes of

ene))triphosphonic acid L . Compound 11 (190 mg, ligands L and L was obtained by tail vein injection of the

.34 mmol) was dissolved in 6 M HCl (6 mL) and the mixture complexes. Transgenic breast cancer mice (MMTV-NeuNT TG.

4

1

2

0

7

3

was heated at reflux overnight. The solvent was evaporated NK line) in the FVB background were obtained by breeding

under reduced pressure and the residue was dissolved in a MMTV-NeuNT male with wild-type FVB female mice. Genotyp-

minimum of methanol. Addition of diethyl ether resulted in ing was performed on DNA extracted from tail as previously

the formation of a precipitate, which was collected by centrifu- described, using two primer pairs specific to the NeuNT

gation and dried under vacuum to give ligand L (60 mg, (forward1:GGAAGTACCCGGATGAGGAGGGCATATG,

7

0

4

and

1

3

4

6%). H NMR (D

2

O, 200 MHz): δ 8.39 (t, JH–H = 7.9 Hz, 1H, reverse1: CCGGGCAGCCAGGTCCCTGTGTACAAGCCG; forward2:

3

H_p_y_r), 7.90 (d, J

= 7.9 Hz, 1H, H_p_y_r), 7.84 (d, J_H_–_H= 7.9 Hz, TCCTGTGTGGATCTGGATGA, and reverse2: GTCAGCGGCTC-

H–H

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9617

View Article Online

Paper

Organic & Biomolecular Chemistry

CACTAACTC). Female mice with macroscopically visible primary labeled MTn12 antibody (250 μL at 1.8 mg mL⁻in PBS, pH

1

tumors were used three months after the palpation of the first 7.4). The labeling reaction was completed after 30 min with

9

9m

primary tumor for intravenous injection of the labeled

Tc- slow stirring at room temperature.

L*-MTn12 antibody in the lateral tail vein and in vivo imaging

The mixture was purified by centrifugation at 13 000 RPM

by SPECT.

over 30 000 MW cutoﬀ filtering devices (Amicon Ultra 0.5 mL

In vivo CT/SPECT analysis. In order to recognize the centrifugal filters, Millipore).

diﬀerent organs, prior to SPECT analysis, a CT image was The radiochemical purity of

recorded on the AMISSA platform, a home-made multimodal- to be more than 95% by instant thin layer chromatography

9

9m

Tc-L*-MTn12 was estimated

9

9m

ity imaging system for small animals combining X-ray, SPECT (ITLC).

Tc-L*-MTn12 remains immobile, and other species

9

9m

99m

−

and PET devices. The µCT delivered a 3D reconstructed

(

Tc-glucoheptonate and

rated saline was used as a developing solvent.

4

TcO ) are mobile when satu-

7

volume of the animal in real time.

9

9m

Targeting of

Tc-labeled molecules was monitored via the

µSPECT imaging technique. Mice were anesthetized by intra-

−

1

peritoneal injection of 10 µl g of a solution made with keta-

mine hydrochloride 10% (Imalgene, Centravet, Velaine en

Acknowledgements

Haye, France) and Xylazine 5% (Rompun, Centravet, Velaine This work was supported by the Centre National de la

en Haye, France) and placed in prone position in the animal Recherche Scientifique and the Université de Strasbourg (UMR

holder after injection in the tail vein. µSPECT imaging was per- 7509 and UMR 7178 CNRS-Université de Strasbourg). The

formed 4 minutes post-intravenous injection of 124 MBq, French Institute for Research on Cancer (INCa) is gratefully

9

9m

1

18 MBq and 59 MBq in 220 µL, 200 µL and 100 µL of

Tc acknowledged for financial support (G.O., L.C., P.L.). Matthieu

1

2

molecules for L , L and L*-MTn12, respectively, injected in Chessé (UMR 7509) is gratefully acknowledged for ESI-MS

the tail vein. The µSPECT system consists of a four head detec- measurements. Dr Jean-Michel Chezal is also gratefully

tion gamma camera. Each head comprises five separated acknowledged for fruitful discussions.

detection modules arranged along a circle of 58 mm with the

pinhole as the center. A detection module consists of a

3

YAP : Ce matrix of 8 × 8 scintillating crystals 2.3 × 2.3 × 28 mm

References

each coupled to a multi-anode photomultiplier 8 × 8 (Hama-

matsu H 8804). The distance from the pinhole to the axis of

rotation is 28 mm and the distance between the pinhole and

the crystal is 58 mm which results in a magnification factor of

1 B. M. Zeglis and J. S. Lewis, Dalton Trans., 2011, 40, 6168.

2 S. Bhattacharyya and M. Dixit, Dalton Trans., 2011, 40,

6112.

3 T. J. Wadas, E. H. Wong, G. R. Weisman and

C. J. Anderson, Chem. Rev., 2010, 110, 2858.

4 M. Shokeen and C. J. Anderson, Acc. Chem. Res., 2009, 42,

832.

7

8

2

.07. Images were reconstructed using the OSEM (Ordered

Subset Expectation Maximisation) iterative algorithm adapted

for pinhole imaging. Images were viewed and quantified using

the Anatomist freeware (http://brainvisa.info/index_f.html).

5

6

S. Liu, Adv. Drug Delivery Rev., 2008, 60, 1347.

S. Liu and D. S. Edwards, Chem. Rev., 1999, 99, 2235.

General protocol for radiolabeling

1

2

Ligand L (5 mg) or ligand L (4.5 mg) were dissolved in water

7 P. Valsamaki, A. Gotzamani-Psarrakou, S. Tsiouris,

E. Molyvda-Athanasopoulou, K. Psarrakos, V. Papantoniou,

S. Gerali and C. Zerva, Int. J. Cancer, 2006, 119, 968.

8 R. Ferdani, D. J. Stigers, A. L. Fiamengo, L. Wei, B. T. Y. Li,

J. A. Golen, A. L. Rheingold, G. R. Weisman, E. H. Wong

and C. J. Anderson, Dalton Trans., 2012, 41, 1938.

9 Y. Guo, R. Ferdani and C. J. Anderson, Bioconjugate Chem.,

2012, 23, 1470.

−

1

(

1 mL). Sodium ascorbate (150 μL, 5 mg mL ) and stannous

−

1

chloride (75 μL, 2 mg mL ) in 0.1 M hydrochloric acid were

then added to the above solution. The pH of the resulting solu-

tion was adjusted to 6.5–7.0 with 1 M NaOH. Freshly eluted

9

9m

−

−1

TcO

resulting mixture was then stirred at room temperature for

0 min for optimum labelling yield.

4

(1 mL, 600–700 MBq mL ) was added, and the

3

The labeling eﬃciency was estimated chromatographically 10 X. Sun, M. Wuest, Z. Kovacs, A. D. Sherry, R. Motekaitis,

using ITLC-SG strips as the stationary phase and acetone as

the mobile phase. After developing the TLC, it was cut into two

Z. Wang, A. E. Martell, M. J. Welch and C. J. Anderson,

J. Biol. Inorg. Chem., 2003, 8, 217.

equal halves and radioactivity of each segment was determined 11 S. Abada, A. Lecointre, M. Elhabiri and L. J. Charbonnière,

9

9m

−

in a well-type γ-ray counter. The free

TcO

4

moved with the

Dalton Trans., 2010, 39, 9055.

solvent, while the radiolabeled formulation remained at the 12 S. Abada, A. Lecointre, I. Déchamps-Olivier, C. Platas-Igle-

point of application.

sias, C. Christine, M. Elhabiri and L. Charbonnière, Radio-

chim. Acta, 2011, 99, 663.

9

9m

99m

L* was labeled with

Tc using

Tc-glucoheptonate as

9

9m

−

an intermediate. Briefly,

TcO₄(1.4 GBq in 400 μL) was 13 J. Tanwar, A. Datta, A. K. Tiwari, M. Thirumal, K. Chuttani

and A. K. Mishra, Bioconjugate Chem., 2011, 22, 244.

(2 mg mL in 0.1 M HCl). 14 P. Panwar, S. Singh, N. Kumar, H. Rawat and A. K. Mishra,

This solution is then added to the vial containing the L* Bioorg. Med. Chem., 2007, 1138.

added to a solution containing 50 μL sodium glucoheptonate

10 mg ml⁻) and 150 μL SnCl

1

−1

(

2

9618 | Org. Biomol. Chem., 2014, 12, 9601–9620

View Article Online

Organic & Biomolecular Chemistry

Paper

1

5 R. Torres Martin de Rosales, C. Finucane, S. J. Matherb and 40 Y. M. Polikarpov, F. I. Belskii, S. V. Matveev, T. Y. Medved

P. J. Blower, Chem. Commun., 2009, 4847.

6 K. Ogawa and H. Saji, Int. J. Mol. Imaging, 2011, 537687.

and M. I. Kabachnik, Izv. Akad. Nauk. SSSR, Ser. Khim.,

1982, 3, 710.

1

7 I. Lukeš, J. Kotek, P. Vojtíšek and P. Hermann, Coord. 41 K. P. Guerra, R. Delgado, L. M. P. Lima, M. G. B. Drew and

Chem. Rev., 2001, 216–217, 287. V. Félix, Dalton Trans., 2004, 1812.

8 C. Christine, M. Koubemba, S. Shakir, S. Clavier, L. Ehret- 42 D. K. Cao, J. Xiao, Y. Z. Li, J. M. Clemente-Juan, E. Coronado

1

Sabatier, F. Saupe, G. Orend and L. J. Charbonnière, Org.

Biomol. Chem., 2012, 10, 9183.

and L. M. Zheng, Eur. J. Inorg. Chem., 2006, 1830.

43 B. Kurzak, A. Kamecka, K. Kurzak, J. Jezierska and

P. Kafarski, Polyhedron, 1998, 17, 4403.

44 A. Sornosa Ten, N. Humbert, B. Verdejo, J. M. Llinares,

M. Elhabiri, J. Jezierska, C. Soriano, H. Kozlowski,

A. M. Albrecht-Gary and E. Garcia-Espana, Inorg. Chem.,

2009, 48, 8985.

1

9 T. Kobayashi, T. Yaita, Y. Sugo, H. Suda, S. Suzuki, Y. Fujii

and Y. Nakano, J. Heterocycl. Chem., 2006, 43, 549.

0 S. Aime, M. Botta, E. Garino, S. Geninatti Crich,

G. Giovenzana, R. Pagliarin, G. Palmisano and M. Sisti,

Chem. – Eur. J., 2000, 6, 2609.

2

1 (a) P. Gans, A. Sabatini and A. Vacca, HYPERQUAD2000, 45 T. Gajda, A. Jansco, S. Mikkola, H. Lonnberg and H. Sirges,

Leeds, U.K., and Florence, Italy, 2000; (b) P. Gans,

A. Sabatini and A. Vacca, Talanta, 1996, 43, 1739.

2 P. Gans, Data Fitting in the Chemical Sciences, John Wiley &

Sons, Chichester, 1992.

3 (a) L. Blaha, I. Lukeš, J. Rohovec and P. Hermann,

J. Chem. Soc., Dalton Trans., 1997, 2621; (b) M. Wozniak

J. Chem. Soc., Dalton Trans., 2002, 1757.

46 A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier,

Amsterdam, 2nd edn, 1984, 553.

47 J. Brandel, A. Lecointre, J. Kollek, S. Michel, V. Hubscher-

Bruder, I. Déchamps-Olivier, C. Platas-Iglesias and

L. J. Charbonnière, Dalton Trans., 2014, 43, 9070.

2

and G. Nowogrocki, Talanta, 1979, 26, 1135; 48 H. Irving and R. J. P. Williams, J. Chem. Soc., 1953, 3192.

c) M. Wozniak and G. Nowogrocki, Bull. Soc. Chim. Fr., 49 R. J. Motekaitis, A. E. Martell and R. A. Hancock, Coord.

978, I-153. Chem. Rev., 1994, 133, 39.

(

1

2

4 (a) K. P. Guerra, R. Delgado, L. M. P. Lima, M. G. B. Drew 50 H. J. Schneider, Angew. Chem., Int. Ed., 2009, 48, 3924.

and V. Félix, J. Chem. Soc., Dalton Trans., 2004, 1812; 51 W. R. Harris, C. J. Carrano and K. N. Raymond, J. Am.

(b) Y. Couturier and C. Petitfaux, Bull. Soc. Chim. Fr., 1978,

Chem. Soc., 1979, 101, 2213.

I-121.

52 D. Grujicic and B. Pesic, Electrochim. Acta, 2002, 47, 2901.

5 L. Pellegatti, J. Zhang, B. Drahos, S. Villette, F. Suzenet, 53 C. Bucher, E. Duval, J. M. Barbe, J. N. Verpeaux, C. Amatore

G. Guillaumet, S. Petoud and E. Tóth, Chem. Commun,

008, 6591.

6 I. Lukeš, J. Kotek, P. Vojtíšek and P. Hermann, Coord.

and R. Guilard, Acad. Sci. Paris Ser. II c, 2000, 3, 211.

54 C. Amatore, J. M. Barbe, C. Bucher, E. Duval, C. Amatore,

R. Guilard and J. N. Verpeaux, Inorg. Chim. Acta, 2003, 356,

267.

2

Chem. Rev., 2001, 287, 216.

7 K. Popov, H. Ronkkomaki and L. H. J. Lajunen, Pure Appl. 55 (a) R. S. Nicholson and I. Shain, Anal. Chem., 1964, 36, 706;

Chem., 2001, 73, 1641.

8 T. Ichikawa and K. Sawada, Bull. Chem. Soc. Jpn., 1997, 70,

(b) A. J. Bard and L. R. Faulkner, Electrochemical Methods:

Fundamentals and Applications, Wiley & Sons, New York,

2ème édition, 2001.

8

29.

9 T. Ichikawa and K. Sawada, Bull. Chem. Soc. Jpn., 1997, 70, 56 R. N. Patel, R. P. Shrivastava, N. Singh, S. Kumar and

2

111.

K. B. Pandeya, Indian J. Chem., Sect. A: Inorg., Bio-inorg.,

Phys., Theor. Anal. Chem., 2001, 40, 361.

57 E. A. Ambundo, M. V. Deydier, L. A. Ochrymowycz and

D. B. Rorabacher, Inorg. Chem., 2000, 39, 1171.

58 T. S. Roche and R. G. Wilkins, J. Am. Chem. Soc., 1974, 96,

5082.

3

0 A. Zakrzewski and J. Geisler, Chem. Anal., 1987, 32, 151.

1 A. Zakrzewski and J. Geisler, Chem. Anal., 1984, 29, 631.

2 A. Mohamadou and C. Gerard, J. Chem. Soc., Dalton Trans.,

2

001, 3320.

3

3 H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbühler,

Talanta, 1985, 32, 95.

59 M. Eigen and R. G. Wilkins, Adv. Chem. Ser., 1965, 49, 55.

4 H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbühler, 60 M. Eigen, Pure Appl. Chem., 1963, 6, 97.

Talanta, 1985, 32, 251. 61 R. M. Fuoss, J. Am. Chem. Soc., 1958, 80, 5059.

5 H. Gampp, M. Maeder, C. J. Meyer and A. D. Zuberbühler, 62 L. Helm and A. E. Merbach, Chem. Rev., 2005, 105, 1923.

Talanta, 1985, 32, 1133. 63 D. T. Richens, Chem. Rev., 2005, 105, 1961.

6 W. Duan, K. Satoh and K. Sawada, Bull. Chem. Soc. Jpn., 64 H. Diebler, M. Eigen, G. Ilgenfritz, G. Maab and

001, 74, 487. R. Winkler, Pure Appl. Chem., 1969, 20, 93.

7 R. J. Motekaitis, I. Murase and A. E. Martell, Inorg. Nucl. 65 T. B. Bartnikas and J. D. Gitlin, J. Biol. Chem., 2003, 278,

Chem. Lett., 1971, 7, 1103. 33602.

8 K. Sawada, T. Kanda, Y. Naganuma and T. J. Suzuki, 66 R. Chiquet-Ehrismann, G. Orend, M. Chiquet, R. P. Tucker

J. Chem. Soc., Dalton Trans., 1993, 2557. and K. S. Midwood, Matrix Biol., 2014, 37C, 112.

9 B. Kurzak, A. Kamecka, K. Kurzak, J. Jezierska and 67 G. Orend and R. Chiquet-Ehrismann, Cancer Lett., 2006,

P. Kafarski, Polyhedron, 1998, 17, 4403. 244, 143.

2

Org. Biomol. Chem., 2014, 12, 9601–9620 | 9619

View Article Online

Paper

Organic & Biomolecular Chemistry

6

8 K. S. Midwood, T. Hussenet, B. Langlois and G. Orend, 73 W. J. Muller, E. Sinn, P. K. Pattengale, R. Wallace and

Cell. Mol. Life Sci., 2011, 68, 3175. P. Leder, Cell, 1988, 54, 105.

9 E. Aufderheide and P. Ekblom, J. Cell Biol., 1988, 107, 2341. 74 T. Løvdal, E. Andersen, A. Brech and T. Berg, J. Cell Sci.,

0 F. Saupe, A. Schwenzer, Y. Jia, I. Gasser, C. Spenlé, 2000, 113, 3255.

B. Langlois, M. Kammerer, O. Lefebvre, R. Hlushchuk, 75 H. E. Gottlieb, V. Kotlyar and A. Nudelman, J. Org. Chem.,

T. Rupp, M. Marko, M. van der Heyden, G. Cremel, 1997, 62, 7512.

C. Arnold, A. Klein, P. Simon-Assmann, V. Djonov, 76 F. Dioury, C. Ferroud, A. Guy and M. Port, Tetrahedron,

A. Neuville-Méchine, I. Esposito, J. Slotta-Huspenina, 2009, 65, 7573 and references therein.

K.-P. Janssen, O. de Wever, G. Christofori, T. Hussenet and 77 D. Brasse, B. Humbert, C. Mathelin, M. C. Rio and

6

7

G. Orend, Cell Rep., 2013, 5, 482.

1 D. Hanahan, Nature, 1985, 315, 115.

2 R. Torres Martin de Rosales, R. Tavaré, A. Glaria, G. Varma,

A. Protti and P. J. Blower, Bioconjugate Chem., 2011, 22, 455.

J. L. Guyonnet, Phys. Med. Biol., 2005, 50, 5799.

78 Z. El Bitar, L. Barabara, V. Bekaert, D. Huss and D. Brasse,

Conf. Proc. of 2008 IEEE Nuclear Science Symposium and

Medical Imaging Conf., 2008, 3801.

7

9620 | Org. Biomol. Chem., 2014, 12, 9601–9620

Article Doi

Phosphonated chelates for nuclear imaging

DOI: 10.1039/c4ob01514b

Source and publish data:

Authors:

Article abstract of DOI:10.1039/c4ob01514b

Full text of DOI:10.1039/c4ob01514b

Products guided by the article

R&D Labs maybe for 7703-74-4

Relevant to this article

Hot Product