Nuclear Medicine and Biology 41 (2014) e23–e29

Contents lists available at ScienceDirect

Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio

Radioiodinated and astatinated NHC rhodium complexes: Synthesis Holisoa Rajerison a,⁎, François Guérard a, Marie Mougin-Degraef a, Mickael Bourgeois a, b, Isidro Da Silva c, Michel Chérel a, Jacques Barbet a, b, Alain Faivre-Chauvet a, Jean-François Gestin a,⁎ a b c

Centre de Recherche en Cancérologie Nantes/Angers, 44007 Nantes Cedex 1, France GIP ARRONAX, 44817 Saint-Herblain Cedex, France CEMHTI-CNRS UPR3079, 45071 Orléans Cedex 2, France

a r t i c l e

i n f o

Article history: Received 30 September 2013 Received in revised form 4 December 2013 Accepted 4 December 2013 Keywords: Astatine-211 Iodine-125 Radioimmunotherapy N-heterocyclic carbene (NHC) Rhodium

a b s t r a c t Introduction: The clinical development of radioimmunotherapy with astatine-211 is limited by the lack of a stable radiolabeling method for antibody fragments. An astatinated N-heterocyclic carbene (NHC) Rhodium complex was assessed for the improvement of radiolabeling methodologies with astatine. Methods: Wet harvested astatine-211 in diisopropyl ether was used. Astatine was first reduced with cysteine then was reacted with a chlorinated Rh-NHC precursor to allow the formation of the astatinated analogue. Reaction conditions have been optimized. Astatine and iodine reactivity were also compared. Serum stability of the astatinated complex has been evaluated. Results: Quantitative formation of astatide was observed when cysteine amounts higher than 46.2 nmol/μl of astatine solution were added. Nucleophilic substitution kinetics showed that high radiolabeling yields were obtained within 15 min at 60°C (88%) or within 5 min at 100°C (95%). Chromatographic characteristics of this new astatinated compound have been correlated with the cold iodinated analog ones. The radioiodinated complex was also synthesized from the same precursor (5 min. at 100°C, up to 85%) using [125I]NaI as a radiotracer. In vitro stability of the astatinated complex was controlled after 15 h incubation in human serum at 4°C and 37°C. No degradation was observed, indicating the good chemical and enzymatic stability. Conclusion: The astatinated complex was obtained in good yield and exhibited good chemical and enzymatic stability. These preliminary results demonstrate the interest of this new radiolabeling methodology, and further functionalizations should open new possibilities in astatine chemistry. Advances in knowledge and implications for patient care: Although there are many steps and pitfalls before clinical use for a new prosthetic group from the family of NHC complexes, this work may open a new path for astatine-211 targeting. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Astatine is the heaviest member of the halogen group with the biggest atomic radius (0.45Å). More than twenty astatine isotopes are known, all of them are radioactive. Astatine-211 is regarded as particularly promising for targeted alpha-therapy of micrometastatic diseases [1, 2]. Indeed astatine-211 decays to bismuth-207 by emitting high energy alpha particles (Emean: 6.4MeV, t1/2: 7.2h.), either directly or through its short half-life (0.5 sec) daughter polonium-211. The short path length of the emitted alpha particle with a mean range in human tissues of 65 μm may limit radiotoxicity to neighboring normal tissues if the radioactivity is specifically directed to tumor cells. A monoclonal Abbreviations: NHC, N-heterocyclic carbene; DIPE, Diisopropyl ether; SAB, Succinimidyl astatobenzoate; FAb, Fragment of antibody; THF, Tetrahydrofurane; COD, cyclooctadiene; FD/FI, Field desorption/field ionization; RT, Room temperature; RCY, Radiochemical yield; RCP, Radiochemical purity. ⁎ Corresponding author at: CRCNA, IRS UN, 8 quai Moncousu, BP 70721–44007 - Nantes Cedex 1. Tel.: +33 2 28 08 02 21, +33 2 28 08 02 20; fax: +33 2 28 08 02 04. E-mail addresses: [email protected] (H. Rajerison), [email protected] (J.-F. Gestin). 0969-8051/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.nucmedbio.2013.12.004

antibody fragment can ideally perform this targeting as it combines the recognition of specific or overexpressed receptors on abnormal cells to a biological half-life that matches astatine-211 physical half-life. Astatine-211 is usually produced in a cyclotron by α-irradiation of bismuth targets via the 209Bi (α, 2n) 211At nuclear reaction. Purification and recovery are performed using dry distillation [3] or wet extraction in di-isopropyl ether (DIPE) [4]. Rich chemical reactivity is expected for this element since six different oxidation states (−I, 0, + I, +III, + V, + VII) have been described [5]. However, the absence of stable or long half-life astatine isotope and the small number of α-particle accelerators available for astatine-211 production considerably hamper the development of astatine chemistry. Direct radiolabeling of proteins with astatine was attempted on the basis of the theoretical similarity of astatine and iodine in terms of chemical reactivity. Unfortunately, rapid de-astatination occurred in vitro and in vivo. This instability was attributed to the formation of unstable S-At bonds with free thiols of the proteins, instead of the aromatic electrophilic substitution of tyrosine residue(s) as obtained with iodine [6, 7]. Indirect astatination by radiolabeling a prosthetic group and coupling to the protein in a second step has then been

e24

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

considered. Indirect radiolabeling mainly uses the electrophilic substitution of leaving groups attached to vinylic [8] or aromatic backbones [9–11] with At(+ I). Accordingly, succinimidyl astatobenzoate (SAB), which is currently the most frequently used for proteins labeling with astatine-211, is synthesized from the halodemetalation of a stannylated aromatic precursor [12]. Radiolabeling with this prosthetic group shows some lack of stability in vivo, especially for small and rapidly metabolized antibody fragments. Radioactivity accumulation in lungs, spleen, stomach and thyroid was observed in biodistribution studies highlighting the in vivo release of astatine [13]. This instability can be correlated to the relative weakness of the aryl–astatine bond compared to the aryl–iodine one (~ 49 kcal/mol vs. 62 kcal/mol) [14]. A contrario, the B-At bond is stronger, which led Wilbur and coworkers to develop boron cages as pendant groups for proteins' radiolabeling [15]. The boronic prosthetic group can be attached to the protein and then be radiolabeled under mild conditions. No de-astatination occurs in vivo, but some protein pharmacokinetic modifications have been observed. The chemistry of astatine oxidation states other than At(+I) has been clearly less explored. To the best of our knowledge, only one study reports prosthetic groups with At(+III) in their structures [16], and no example of prosthetic groups synthesized directly from higher oxidation state (N + I) species has been yet described. In addition, previously mentioned species have not been clearly identified [17]. Nevertheless, a few investigations have focused on astatide chemistry. Astatination of aromatic compounds by halide substitution [18] or from the corresponding aromatic diazonium salts [19] to form SAB type compounds has been described. Afterward, on the basis that astatide should interact as a soft ligand according to the Pearson's theory and form strong complexes with soft metal cations (such as cationic rhodium, iridium, platinum, palladium, silver, gold…), Pruszynski and coworkers conducted studies about the stability of astatide–mercury complexes [20]. They demonstrated that the astatide complex is much more stable than the iodinated analogue. More recently new rhodium (+III) and iridium (+III)-astatide complexes stabilized by the thioether ligand 16S4-diol have been developed, but no in vitro or in vivo stability data have yet been published [21]. Over the last decade, the N-heterocyclic carbene (NHC) ligand family has attracted considerable attention as valuable ligands in coordination chemistry and homogeneous catalysis, notably after the discovery of persistent bulky free NHC stable in the solid state [22–24]. These strongly nucleophilic ligands were found to be interesting substitutes for phosphine ligands with stronger σ-donor and weaker π-acceptor ability [25–27]. They can bind firmly to several transition metal ions under various oxidation states, and lead to complexes with low sensitivity to air, heat and moisture, highly stable especially when bulky ligands are used. In addition to electronic and steric beneficial properties, robust synthetic procedures give access to NHCs complexes bearing various tags [28, 29] including peptide and protein coupling moieties [30]. More recently, the interest of NHC complexes as pharmaceutical compounds highlighted their intrinsic promising properties as antimicrobial [31] and antitumor agents [32]. The unique stabilization properties of NHC ligands and the possibility of tailored structural changes led us to consider the introduction of astatine in an NHC complex structure. Herein, as a proof of this original concept, we choose to attach astatide to rhodium(I) by ligand exchange; this metallic nucleus is stabilized by an NHC ligand. Ultimately, this complex could be subsequently conjugated to biomolecules with a proper selection of ligand functionalization. This choice has been made according to three distinct ideas: first, we have hypothesized that rhodium(I) soft cation and astatide soft ligand should form strong bonds. In a second time, we have considered that the chloride anion of a rhodium-halide structure should be easily displaced by astatide having regards to their respective nucleophilic characters. Finally, we have assumed that electronic and steric stabilizing properties of bulky NHC should increase the stability of the radiolabeled complexes

in vitro and in vivo. In this paper, we thus describe the synthesis and radiolabeling of a new prosthetic group, the (1-benzyl-3nitrophenylimidazolidene)rhodium(I) chloride complex with iodide and astatide. The in vitro stability of astatinated complexes under physiological conditions has also been evaluated.

2. Materials and methods 2.1. General 2.1.1. Reagents Chemicals purchased from commercial sources were reagent grade or better and were used without further purification. All solvents were obtained as HPLC grade or better. Solvents for HPLC analysis were degassed before use.

2.1.2. Spectral analysis 1 H and 13C{H} NMR spectra were obtained on a Bruker AC (400MHz). Proton chemical shifts are expressed as parts per million (ppm) using solvent peak as internal reference. High resolution mass spectra measurements were recorded on Waters-Micromass GCT Premier spectrometers.

2.1.3. Radioactive materials All radioactive materials were handled according to the approved protocols at the Centre de Recherche en Cancérologie de Nantes-Angers. Astatine-211 was produced at the CEMHTI (Orléans, France) using the 209Bi(α, 2n) 211At reaction by bombarding a 240 μm thick natural bismuth layer on copper target with a 1.95–2.15 μA beam of 28MeV α-particle during 2 h. Wet extraction of astatine in DIPE was previously described [33]. The harvested activity was determined in an ACAD 2000 ionization chamber (Lemer Pax, Carquefou, France). Iodine-125 was purchased (Perkin Elmer, Courtaboeuf, France) as [ 125I]NaI in 0.048 M NaOH. For radioactive experimentation, each reaction condition was duplicated and performed using at least two different astatine or iodine batches.

2.1.4. Chromatography Non radioactive compounds: Silica gel chromatography was conducted with 2–25 μm 60 Α silica gel (Carlo Erba SDS, Val de Reuil, France). TLC was carried out on precoated silica gel 60F254 TLC plastic sheets (Merck, Darmstadt, Germany), and reaction products were detected by UV light (254 nm). HPLC analysis was performed using a Waters HPLC System equipped with a Waters 486 Tuneable Absorbance. A Waters Prep NovaPack HR Silica column was used with a gradient elution of heptane (A) and ethyl acetate (B) (0–3 min: A, 3.01–9.5 min: 70/30 (A/B), 9.51–15 min: 60/40 (A/B), 15.01–20 min: A) at 2 ml/min. All data were analyzed using Waters Empower data acquisition and analysis software. Radioactive compounds: Silica gel chromatography was conducted with SepPack silica gel cartridges (Oasis, France). Radio-TLC was carried out on precoated silica gel 60F254 TLC plastic sheets (Merck, Darmstadt, Germany), was eluted with heptane/acetone 3:2 and was examined using a Typhoon 9410 Variable Mode Imager (GE Healthcare Bioscience). HPLC analysis was performed using a Waters HPLC System equipped with a Waters 486 Tuneable Absorbance Detector and a Packard Bioscience Flow Scintillation Analyser 150 TR (Meriden, CT, US). A Waters Prep NovaPack HR Silica column was used with a gradient elution of heptane (A) and ethyl acetate (B) (0–3 min: A, 3.01–9.5 min: 70/30 (A/B), 9.51–15 min: 60/40 (A/B), 15.01–20 min.: A) at 2 ml/min. All data were analyzed using Waters Empower data acquisition and analysis software.

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

2.2. Organic synthesis 2.2.1. 1-benzyl-3-(4-nitrophenyl)imidazolium bromide 1 Benzyl bromide (712 μL, 6 mmol) and 4-nitrophenylimidazole (222 mg, 1.2 mmol.) were stirred in 10 ml of anhydrous THF under reflux for 45 h. The precipitate was then filtered, washed with THF and dried under reduced pressure to give 393 mg (91%) of 1 as an off white powder. 1H NMR (DMSO-d6, 400.13 MHz): δ 10.37 (t, 1H, J = 1.5Hz, N-CH-N), 8.58 (m, 1H, H(imidazolium backbone)), 8.55 (2H, m, H (nitrobenzyl)), 8.19 (1H, m, H(imidazolium backbone)), 8.16 (2H, m, H (nitrophenyl)), 7.60–7.55 (2H, m, Hm(benzyl)), 7.54–7.40 (3H, m, Hp + Ho(benzyl)), 5.60 (2H, s, CH2). 13C NMR (DMSO-d6,100.16MHz): δ 147.5 (N-CH-N), 139.3, 136.5, 134.2, 128.9, 128.8, 128.6, 125.5, 123.5, 123.0, 121.6, 52.5 (CH2). HRMS (FD/FI) calcd for [M-Br]+: 280.1086, [2M-Br]+: 639.1355; found [M-Br]+: 280.1070, [2M-Br] +: 639.1439. 2.2.2. (1-benzyl-3-(4-nitrophenyl)imidazolidene)(cyclooctadiene) rhodium(I) chloride 2 1 (44 mg, 0.122 mmol) and silver oxide (14 mg, 0.061 mmol.) were stirred in 4 ml of methylene chloride at room temperature for 2 h. The resultant mixture was filtered over a celite pad. [Rh(COD) Cl]2 (30 mg., 0.061 mmol.) was then added. The solution was stirred 18 h at room temperature. Volatiles were removed under reduced pressure, and the residue was purified by silica gel chromatography (99.5/0.5 DCM/MeOH). After concentration, heptane was added until precipitates appeared. The product was recovered by filtration and dried under reduced pressure. Forty-four milligrams (65%) of 2 was then obtained as a yellow powder. 1H NMR (CDCl3, 400.13 MHz): δ 8.59 (d, J = 8.4 Hz, H(nitrobenzyl)), 8.45 (d, J = 8.4 Hz, H (nitrobenzyl)), 7.67–7.35 (5H, m, H(benzyl)), 7.22 (1H, s, H (imidazolidene backbone)), 6.91 (1H, s, H(imidazolidene backbone)), 5.99 (2H, AB, J = 14.8 Hz, Δν = 37.1 Hz, CH2), 5.27–4.99 (2H, m, CH(COD)), 3.18 (1H, bs, CH(COD)), 2.63–2.47 (1H, m, CH (COD)), 2.40–2.28 (1H, m, CH2(COD)), 2.25–2.08 (2H, m, CH2 (COD)), 1.93–1.70 (3H, m, CH2(COD)), 1.70–1.51 (2H, m, CH2 (COD)). 13C NMR (CDCl3, 100.16 MHz): δ 185.7 (d, JC-Rh = 52 Hz, carbene), 146.9, 145.1, 135.5, 129.1 (CH(benzyl)), 128.6 (CH (benzyl)), 128.5 (CH(benzyl)), 125.1 (CH(nitrophenyl)), 124.3 (CH (nitrophenyl)), 122.2 (CH(imidazolidene)), 121.0 (CH(imidazolidene)), 99.5 (d, JC-Rh = 7Hz, CH(COD)), 98.7 (d, JC-Rh = 7 Hz, CH (COD)), 69.3 (d, JC-Rh = 14.0 Hz, CH(COD)), 68.8 (d, JC-Rh = 14.0 Hz, CH(COD)), 55.6 (CH2), 32.7 (CH2(COD)), 32.3 (CH2(COD)), 28.7 (CH2 (COD)), 28.6 (CH2(COD)). HRMS (FD/FI) calcd for [M] +: 525.0690, found [M] +: 525.0652. HPLC: tR = 11.7min. 2.2.3. (1-benzyl-3-(4-nitrophenyl)imidazolidene)(cyclooctadiene) rhodium(I) iodide 3a Route A 1 (44 mg, 0.122 mmol.) and silver oxide (44 mg, 0.061 mmol.) was stirred in 4 ml of methylene chloride at room temperature for 2 h. The mixture was then filtered over a pad of celite. [Rh(COD)Cl]2 (30 mg, 0.061 mmol.) and KI (50 mg, 0.30 mmol) were then added. The solution was stirred 18 h at room temperature. Volatiles were removed under reduced pressure, and the residue was purified by silica gel chromatography using methylene chloride as elution phase. After concentration, heptane was added until precipitate appeared. The product was recovered by filtration and dried under vacuum. Forty-four milligrams (58%) of 3 was then obtained as an orange powder. Route B 2 (10 mg, 0.019 mmol) and KI (10 mg, 0.06 mmol) were stirred 20 h at room temperature in 1 ml of methylene chloride. The volatile materials were removed under reduced pressure, and the residue was purified by silica gel chromatography using methylene chloride as elution phase. After concentration, heptane was added until precipitate appeared. The product was recovered by filtration and dried under vacuum. Nine milligrams (77%) of 3 was then obtained as orange powder. 1H NMR (CDCl3,

e25

400.13 MHz): δ 8.52–8.61 (2H, m, H(nitrobenzyl)), 8.49–8.38 (2H, m, H(nitrobenzyl)), 7.55–7.36 (5H, m, H(benzyl)), 7.27 (1H, d, J = 2.0 Hz, H(imidazolidene backbone)), 6.91 (1H, d, J = 2.0 Hz, H(imidazolidene backbone)), 6.91 (1H, d, J = 2.0 Hz, H(imidazolidene backbone)), 5.90 (2H, AB, J = 15.2 Hz, Δν = 82.2 Hz, CH2), 5.40–5.28 (2H, m, CH(COD)), 3.42–3.32 (1H, m, CH(COD)), 2.78–2.70 (1H, m, CH(COD)), 2.33–2.20 (1H, m, CH2(COD)), 2.19–2.05 (2H, m, CH2(COD)), 1.93–1.78 (2H, m, CH 2 (COD)), 1.71–1.42 (3H, m, CH 2 (COD)). 13 C NMR (CDCl 3 , 100.16MHz): δ 185.6 (d, JC-Rh = 50 Hz, C(Carbene)), 146.7, 145.1, 135.4, 129.1 (CH(benzyl)), 128.7 (CH(benzyl)), 128.5 (CH(benzyl)), 124.6 (CH(nitrophenyl)), 124.2 (CH(nitrophenyl)), 122.4 (CH (imidazolidene)), 121.3 (CH(imidazolidene)), 97.4 (d, JC-Rh = 7.0 Hz, CH(COD)), 96.7 (d, JC-Rh = 7.0 Hz, CH(COD)), 72.6 (d, JC-Rh = 14 Hz, CH(COD)), 72.3 (d, JC-Rh = 14 Hz, CH(COD)), 56.1 (CH2), 32.3 (CH2 (COD)), 31.5 (CH2(COD)), 29.7 (CH2(COD)), 29.5 (CH2(COD)), 29.2 (CH2 (COD)). HRMS (FD/FI) calcd for [M] +: 617.0047, found [M]+: 616.9965. HPLC: tR = 8.2 min. 2.3. Radiolabeling 2.3.1. [ 125I](1-benzyl-3-(4-nitrophenyl)imidazolidene)(cyclooctadiene) rhodium(I) iodide 3b To 16 μL of aqueous solution of [ 125I]NaI (0.12–3.7 MBq) were added 40 μL of 2 in ACN (21,5 μmol/ml). After 30 s of vigorous stirring, the mixture was incubated at 100°C during 5 min HPLC tR = 8 min TLC: Rf = 0.45. 2.3.2. [211At](1-benzyl-3-(4-nitrophenyl)imidazolidene)(cyclooctadiene) rhodium(I) astatide 4 To 10 μL of astatine in DIPE (0.75–1.05 MBq) were added 8 μL of cysteine (58 mM in distilled water). After 30 s of vigorous stirring, 40 μL of 2 in ACN (5.7 mM) were added. The mixture was stirred and incubated 15 min at 60°C (88%) or 5 min at 100°C (95%). TLC: Rf = 0.52. HPLC: tR = 6.8 min. 2.4. Serum stability Rh-At complex 4 (7.5–10.5 MBq) was prepared according to the previously described procedure. The resulting organic solution was evaporated until a final volume of 50 μL. The complex was purified by silica gel chromatography using DCM/MeOH (99/1) as a mobile phase. Activity was counted for each 500 μL aliquots, and fractions of interest were concentrated under nitrogen flux. One milliliter of human serum was then added. The solution was gently stirred and divided into two aliquots incubated for 15 h resp. at 4°C and 37°C. Then, serum proteins were precipitated by adding 750 μl of organic solvents mixture (ACN/DCM 100/1). Supernatant was separated by centrifugation at 4000 rpm and controlled by TLC. 3. Results and discussion Despite the huge potential of alpha immunotherapy particularly with astatinated antibodies fragments, only few alternatives are currently purposed to attach astatine to the vector. An effort has to be made for the development of new methodologies particularly given the fact that in vivo instability of the radiolabeling, which depends highly on the used prosthetic group, is not completely resolved for these vectors. It has been previously shown that astatide is a soft ligand able to form highly stable complexes with many cations of transition elements according to the Pearson's theory. Rhodium N-heterocyclic carbenes complexes, which have been described as strongly stabilized Rhodium (+ I) species, were selected to prepare potentially stable astatinated complexes. We then focused on the synthesis and the radiolabeling of the (1-benzyl-3-(4-nitrophenyl) imidazolidene) (cyclooctadiene) rhodium(I) chloride 2 for the radiolabelling with

e26

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

Scheme 1. Synthesis of rhodium NHC chloride and iodide complexes.

Scheme 2. Synthesis of rhodium NHC astatide complex.

At(− I) by ligand substitution. The N-benzyl-N’-(4-nitrophenyl) imidazolidene ligand was designed to sterically allow substitutions with bulky anions such as astatide but with sufficient steric hindrance to enhance the NHC complex stability. As electronic effects can dramatically change the stabilization properties of NHC ligands [34], the nitrophenyl tag has been chosen to model the introduction of electron-withdrawing groups such as activated esters, maleimides or isothiocyanates conventionally used for bioconjugation with proteins. Moreover, this moiety can be easily reduced and modified for further functionalization. This complex was easily synthesized from the commercial pnitrophenyl imidazole (Scheme 1). The quaternization with benzyl bromide led to the formation of the imidazolium bromide salt 1 in high yield (91%). Rh(+I) NHC complexes are usually synthesized from Rh (+ I) metallic precursors and free carbenes generated in situ by imidazolium deprotonation. However, relatively drastic conditions (particularly strong bases) are needed for this synthesis. The “silver carbene” transmetallation route was then chosen for the synthesis of

chlorinated rhodium complex 2 under reaction conditions compatible with further functionalization [35]. In a first step, the silver carbene complex was obtained from 1 using photosensitive silver oxide in degased dichloromethane. Then, 2 was obtained in 65% yield by transmetallation with the rhodium precursor [Rh(COD)Cl]2. In order to establish chromatographic references (HPLC and TLC) for astatinated complex analyses, the iodinated complex 3a was also synthesized using either the silver carbene route or the halide exchange from 2. No reaction occurred when adding 2 to wet harvested astatine-211 in di-isopropyl ether even after 2 h at 60°C. It was therefore hypothesized that astatine, obtained in DIPE/3M nitric acid solution, was in an oxidized form and that a reduction step was necessary (Scheme 2). Cysteine was selected as it can reduce iodinated species independently of the pH value [36, 37]. Considering that astatine oxidation state may change rapidly in organic solvents [38, 39], an aqueous solution of this reducing agent was used. Then, reduction occurred in a biphasic system with a 2.4 M nitric acid solution as aqueous phase. The best reduction conditions were determined by testing cysteine amounts from 28.9 nmol to 57.8 nmol/μl of astatine. The reaction occurred despite traces of nitric acid, which is considered as a good oxidizing agent for thiols [40]. Moreover, astatide formation is observed even if cysteine molar amounts are lower than that of nitric acid (Fig. 1). We hypothesized that a fast kinetics of astatide formation explains this surprising result while sulfhydryl group oxidation would require

Table 1 Influence of the reducing agent quantity.a

Fig. 1. Chromatographic profiles of reduced astatine using various ratios of cysteine. Astatine (10 μl) was added to cysteine (0 (a), 289 (b), 404 (c), 462 (d), 578 (e) μmol. in 8 μl distill. water) and after 30 s. stirring, the reaction was controlled by TLC.

Cysteine (nmol)

0

289

347

404

462

578

RCY (%)

0

22.7±0.5

38.7 ± 1.8

54.8 ± 2.3

79.3 ± 0.2

77.3 ± 2.6

a

1) At-211 (10 μL, 0.7–1.1 MBq), cysteine (58 mM in distill. water), vortex 30 s. 2) 2 (40 μl, 5.7 mM in CH3CN), 60 °C, 60 min.

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

e27

Fig. 2. TLC (silica gel, Heptane/Acetone (3:2)) for iodinated Rh-NHC complex 3 (a), reducted astatine (b) and astatinated Rh-NHC complex 4 (c) and HPLC chromatographic profiles of Rh-NHC complexes 2, 3a, 3b and 4.

0.48). The best radiochemical yield was obtained using 46.2 nmol of cysteine/μl of astatine solution. Close HPLC retention times (resp 6.8 min and 8.2 min) were obtained for this new species and the iodinated analogue (Fig. 2). The shift between these two retention times may be attributed to their respective polarity but also to remaining nitric acid traces that dramatically influence the retention time. Kinetic studies (Table 2) showed that high radiolabeling yields were obtained within 15 min at 60°C (88%) or within 5 min at 100°C

a longer reaction time (30 min vs. 30 s needed for astatide formation). We have also considered that the large excess of cysteine compared to astatine can play a role. Astatide generated under these reductive conditions was then used for the radiolabeling of the chlorinated precursor 2. The substitution reaction was first tested using a large excess of precursor (21.5 nmol/μl of astatine solution) with heating 60 min at 60°C (Table 1). A new compound was detected by radio-TLC at Rf 0.52 in good agreement with the iodinated rhodium complex 3a (Rf Table 2 Radiolabeling kinetic studies.a Reaction time (min.)

5

10

15

30

45

60

R.T. (% RCY) 37 °C (% RCY) 60 °C (% RCY) 100 °C (% RCY)

27.2 ± 0.3 95.1 ± 2.1

73.3 ± 0.1 17.9 ± 2.2

1.1 ± 0.1 23.4 ± 5.2 88.1 ± 5.4 -

12.3 ± 2.2 26.6 ± 4.0 86.8 ± 2.9 -

12.1 ± 0.85 74.9 ± 0.3 88.8 ± 0.3 -

18.6 ± 3 76.1 ± 1.8 83.6 ± 5.0 -

a

1) At-211 (10 μL, 0.7–1.1 MBq), cysteine (8 μl, 58 mM in distill. water), vortex 30 s. 2) 2 (40 μl, 5.7 mM in CH3CN).

Table 3 Influence of the precursor amounts.a 2 (nmol.)

215

108

27

11

5

3

0,3

0,05

RCY (%)

88.1 ± 5.4

87.4 ± 0.1

82.0 ± 0.6

84.3 ± 4.0

77.4 ± 3.3

63.9 ± 1.6

15.7 ± 0.1

11.5 ± 3.0

a

1) At-211 (10 μL, 0.7–1.1 MBq), cysteine (8 μl, 58 mM in distill. water), vortex 30 s. 2) 2 (40 μl), 60 °C, 15 min.

e28

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

Fig. 3. Astatide complex 4 (10 MBq) was synthesized under previously described conditions (a, RCP: 82.7%). The complex was purified using a silica gel cartridge (b, RCP: 90.8%). After purification and volatiles elimination, human serum was added to 4, and incubation at 4 °C or 37 °C occurred during 15 h. Serum proteins were precipitated, supernatant was separated and 4 chromatographic profiles were controlled by radio-TLC (c, RCP: 90.2%).

(95%). A longer reaction time was needed at 37°C to obtain acceptable yields (60 min, 76%). No significant formation of the astatinated product 4 was observed when the radiolabeling test was performed at RT, even after 2 h. The quantity of precursor was then decreased from 215 to 0.05 nmol per 10 μl of astatine (Table 3), and high yields were obtained for concentrations above 0.5 nmol/μl of astatine solution. These results were comparable with those obtained by Bourgeois and co-workers for SAB preparation using wet harvested astatine [41]. Another potential advantage of this radiolabeling method should be highlighted. “At(+ I)” species reactivity in halodemetalation reactions decreases over time leading to serious variations in radiolabeling yields as demonstrated by Zalutsky and co-workers [39]. The authors assigned this effect to solvent radiolysis and formation of reductive species that changes the predominant chemical form (or oxidation state) of astatine. Comparatively, the chlorinated rhodium complex 2 was radiolabeled with astatine 12–15 h after extraction. No yield or purity variation was observed in comparison with “fresh” astatine. In addition astatine can also be reduced and then used 12–15 h after production. We have then shown that the reduced astatine was stable and reactive even 24 h after extraction, contrarily to At(+ I). Finally, stability of the astatine complex in organic solution was tested after 15 h of incubation at RT. No chromatographic profile modification was observed. The astatinated Rhodium complex was also incubated for 15 h at 4°C and 37°C in human serum. No degradation was observed either at 4°C than at 37°C, which proved in vitro chemical and enzymatic stability (Fig. 3).

Comparative reactivity studies were also conducted with this new precursor radiolabelled with [ 125I]-iodide. Previously determined optimized reaction conditions (5 min at 100°C or 15 min at 60°C) were used. In a first step, the same conditions as for astatine were tested with iodide diluted in a DIPE/3M nitric acid solution, followed by the two step radiolabeling reaction (reduction with cysteine then substitution). The formation of the complex was not observed, and a mixture of unreacted iodinated species probably in an oxidized chemical form was obtained, which again supports the dissimilar redox behavior for astatine and iodine. Thus commercial solutions of iodide were used. As a slightly acidic media may be necessary for the radiolabeling reaction when prosthetic groups bearing a reactive function such as activated esters are used, the reaction was tested for pH values from 1 to 11. The complex 3b was formed in acceptable yields (up to 85%) but only in a basic or moderately acidic media highlighting another difference of reactivity between astatide and iodide (Fig. 4). This reactivity could be explained by the lower availability of iodide compared to astatide at acidic pH considering their respective basicities. In addition, astatine is theoretically more nucleophilic than iodine. As shown in Table 4, the radiolabeling yield also depends on the water/organic phase ratio which tend to design the iodide solubilization as a limiting factor. This hypothesis has been confirmed by the increase of the radiolabeling yield when a phase transfer catalyst such as the tetrabutyl ammonium bromide is added. These experimental observations underline the iodide and astatide chemical behavior differences, which are rarely highlighted in the literature as iodine is commonly admitted as a model for astatine chemistry [42]. 4. Conclusion This study demonstrates the possibility to radiolabel straightforwardly an N-heterocyclic carbene rhodium complex with astatine and iodine. The astatinated complex was obtained in good yields under relatively mild conditions and exhibited good chemical and enzymatic stability. This study also underlines the greater facility of working Table 4 Radiolabeling studies with iodine-125.

Fig. 4. To 16 μl of aqueous solution of [125I]NaI (0.12–3.7 MBq) were added 40 μl of 2 (5.7 mM in CH3CN). After 30 s. of vigorous stirring, the mixture was incubated at 5 min 100 °C. The RCY for 3b was defined as a function of the radioactive aqueous phase pH values.

[125I] NaI (μl)a

1

8

8

RCY (%)

56.5 ± 1.8

68.3 ± 10.2

73.3 ± 2.3

a

b

16

16b

72.8 ± 13.2

82.4 ± 1.8

[125I] NaI (120 kBq) in NaOH 0.048 M, 2 (40 μl, 5.7 mM in CH3CN), 100 °C, 5 min; b with NBu4Cl.

H. Rajerison et al. / Nuclear Medicine and Biology 41 (2014) e23–e29

with At (− I) considering its high stability over time. These preliminary results have demonstrated the promise of this new radiolabeling method. Acknowledgments This work was supported in part by the TARCC project (contract no. Health-F2-2007-201962 of the Seventh Research Framework Program of the European Commission). We also thank the region “Pays de la Loire” (NUCSAN project), the French National Cancer Institute (project n°R06101NG) and the French National Agency for Research (“Investissements d’Avenir” n°ANR-11-LABX-0018-01 Labex IRON and ANR “alpha-RIT/beta-PET”) for financial supports. References [1] (a) Zalutsky MR, Reardon DA, Pozzi OR, Vaidyanathan G, Bigner DD. Targeted alpha-particle radiotherapy with 211At-labeled monoclonal antibodies. Nucl Med Biol 2007;34:779–85. (b) Guérard F, Gestin JF, Brechbiel MW. Production of [(211)At]-astatinated radiopharmaceuticals and applications in targeted a-particle therapy. Cancer Biother Radiopharm 2007;28:1–20. [2] Chérel M, Davodeau F, Kraeber-Bodéré F, Chatal JF. Current status and perspectives in alpha radioimmunotherapy. Q J Nucl Med Mol Imaging 2006;50:322–9. [3] Sture Lindegren S, Bäck T, Jensen HJ. Dry-distillation of astatine-211 from irradiated bismuth targets: a time-saving procedure with high recovery yields. Appl Radiat Isot 2001;55:157–60. [4] Yordanov AT, Pozzi O, Carlin S, Akabani G, Wieland B, Zalutsky MR. Wet harvesting of no-carrier-added 211At from an irradiated 209Bi target for radiopharmaceutical applications. J Radioanal Nucl Chem 2004;262:593–9. [5] Visser GWM, Diemer EL. Inorganic astatine chemistry: formation of complexes of astatine. Radiochim Acta 1983;33:145–51. [6] Aaij C, Tschroots RJM, Lindner L, Feltkamp TEW. The preparation of astatine labelled proteins. Int J Appl Radiat Isot 1975;26:25–30. [7] Vaughan ATM, Fremlin JH. The preparation of astatine labelled proteins using an electrophilic reaction. Int J Nucl Med Biol 1978;5:229–30. [8] Pillai KKR, McLaughlin WW, Lambrecht RR, Bloomer WW. Carrier-free astatination of steroid hormones. J Label Compd Radiopharm 1987;24:1117–22. [9] Visser GWM, Diemer EL, Kaspersen FM. The preparation and stability of 211Atastato-imidazoles. Int J Appl Radiat Isot 1980;31:275–8. [10] Vaidyanathan G, Affleck DJ, Bigner DD, Zalutsky MR. N-succinimidyl 3-[211At] astato-4-guanidinomethylbenzoate: an acylation agent for labeling internalizing antibodies with α-particle emitting 211At. Nucl Med Biol 2003;30:351–9. [11] Talanov VS, Garmestani K, Regino CAS, Milenic DE, Plascjak PS, Waldmann TA, et al. Preparation and in vivo evaluation of novel stabilized linkers for 211At labeling of protein. Nucl Med Biol 2006;33:469–80. [12] (a) Harrison A, Royle L. Preparation of a 211At-IgG conjugate which is stable in vivo. Int J Appl Radiat Isot 1984;35:1005–8. (b) Zalutsky MR, Narula AS. Appl Radiat Isot 1984;39:227–32. [13] Hadley SW, Wilbur DS, Gray MA, Atcher RW. Astatine-211 labeling of an antimelanoma antibody and its fragment using N-succimimidyl p-(211At) astatobenzoate: comparisons in vivo with the p-(125I)iodobenzoyl conjugate. Bioconjug Chem 1991;2:171–9. [14] Vasaros L, Norseev Y, Nhan DD, Khalkin VA, Huan NQ. Determination of the dissociation energy of the C-At bond in substituted astatobenzenes. J Radioanal Nucl Chem Lett 1984;87:31–9. [15] Wilbur DS, Chyan MK, Hamlin DK, Vessella RL, Wedge TJ, Hawthorne MF. Reagents for astatination of biomolecules. 2. Conjugation of anionic boron cage pendant groups to a protein provides a method for direct labeling that is stable to in vivo deastatination. Bioconjug Chem 2007;18:1226–40. [16] Guerard F, Rajerison H, Faivre-Chauvet A, Barbet J, Meyer G, Da Silva I, et al. Radiolabelling of proteins with stabilised hypervalent astatine-211: feasibility and stability. J Nucl Med 2011;52:1486.

e29

[17] Champion J, Alliot C, Renault E, Mokili BM, Chérel M, Galland N, et al. Astatine standard redox potentials and speciation in acidic medium. J Phys Chem A 2010;114:576–82. [18] Visser GWM, Diemer EL, Kaspersen FM. The preparation and stability of astatotyrosine and astato-iodotyrosine. Int J Appl Radiat Isot 1979;30:749–52. [19] Meyer GJ, Roessler K, Stoecklin G. Reaction of aromatic diazonium salts with carrier-free radioiodine and astatine. Evidence of complex formation. J Am Chem Soc 1979;101:3121–3. [20] Pruszinski M, Bilewicz A, Was B, Petelenz B. Formation and stability of asta tidemercury complexes. J Radioanal Nucl Chem 2006;268:91–4. [21] Pruszyski M, Bilewicz A, Zalutsky MR. Preparation of Rh[16aneS4-diol]211At and Ir[16aneS4-diol]211At Complexes as Potential Precursors for Astatine Radiopharmaceuticals. Part I: Synthesis. Bioconjugate Chem 2008;19:958–65. [22] Wanzlick HW. Aspects of nucleophilic carbene chemistry. Angew Chem 1962;1: 7–80. [23] Hahn FE, Jahnke MC. Heterocyclic carbenes; synthesis and coordination chemistry. Angew Chem Int Ed 2008;47:3122–72. [24] Arduengo AJ, Davidson F, Dias HVR, Goerlich JR, Khasnis D, Marshall WJ, et al. An air stable carbene and mixed carbene “dimers”. J Am Chem Soc 1997;119: 12742–9. [25] Cavallo L, Correa A, Costabile C, Jacobsen H. Steric and electronic effects in the bonding of N-heterocyclic ligands to transition metals. J Organomet Chem 2005;690:5407–13. [26] Diez-Gonzälez S, Nolan SP. Stereoelectronic parameters associated with N-heterocyclic carbene (NHC) ligands: a quest for understanding. Coord Chem Rev 2007;251:874–83. [27] Leuthäuβer S, Schwarz D, Plenio H. Tuning the electronic properties of N-heterocyclic carbenes. Chem Eur J 2007;13:7195–203. [28] Peris E. Routes to N-heterocyclic carbenes complexes. Top Organomet Chem 2007;21:83–116. [29] Worm-Leonhard K, Meldal M. Green catalysts: solid-phase peptide carbene ligands in aqueous transition-metal catalysis. Eur J Org Chem 2008;31:5244–53. [30] Lemke J, Metzler-Nolte N. The synthesis of ruthenium and rhodium complexes with functionalized N-heterocyclic carbenes and their use in solid phase peptide synthesis. Eur J Inorg Chem 2008;21:3359–66. [31] Hindi KM, Siciliano TJ, Durmus S, Panzner MJ, Medvetz DA, Reddy D, et al. Synthesis, stability, and antimicrobial studies of electronically tuned silver acetate N-heterocyclic carbenes. J Med Chem 2008;51:1577–83. [32] Teyssot ML, Jarrousse AS, Manin M, Chevry A, Roche S, Norre F, et al. Metal-NHC complexes: a survey of anti-cancer properties. Dalton Trans 2009;35:6894–902. [33] Alliot C, Chérel M, Barbet J, Sauvage T, Montavon G. Extraction of astatine-211 in diisopropylether (DIPE). Radiochim Acta 2009;97:161–5. [34] (a) Sashuk V, Peeck LH, Plenio H. [(NHC)(NHCewg)RuCl2(CHPh)] complexes with modified NHCewg ligands for efficient ring-closing metathesis leading to tetrasubstituted olefins. Chem Eur J 2010;16:3983–93. (b) Leuthäusser S, Schmidts V, Thiele CM, Plenio H. π-Face Donor Properties of N-Heterocyclic Carbenes in Grubbs II Complexes. Chem Eur J 2010;14: 5465–81. (c) Ogle JW, Miller SA. Electronically tunable N-heterocyclic carbene ligands: 1,3-diaryl vs. 4,5-diaryl substitution. Chem Comm 2010;38:5728–30. [35] Zinner SC, Hermann WA, Kühn FE. Synthesis and characterization of asymmetric NHC complexes. J Organomet Chem 2008;693:1543–6. [36] Lucas CC, King EJ. The iodometric titration of cysteine and allied substances. Biochem J 1932;26:2076–89. [37] Simonsen DG. The oxidation of cysteine with iodine: formation of a sulfinic acid. J Biol Chem 1933;101:35–42. [38] Visser GM. Inorganic astatine chemistry: Part II. The chameleon behavior and electrophilicity of At-species. Radiochim Acta 1989;47:97–103. [39] Pozzi OR, Zalutsky MR. Radiopharmaceutical chemistry of targeted radiotherapeutics, part 3: α-particle–induced radiolytic effects on the chemical behavior of 211At. J Nucl Med 2007;48:1190–6. [40] Misra AK, Agnihotri G. Nitric acid mediated oxidative transformation of thiols to disulfides. Synth Commun 2004;34:1079–85. [41] Bourgeois M, Guerard F, Alliot C, Mougin-Degraef M, Rajerison H, Remaud-Le Saëc P, et al. Feasibility of the radioastatination of a monoclonal antibody with astatine-211 purified by wet extraction. J Labelled Compd Radiopharm 2008;15: 379–83. [42] Adam MJ, Wilbur DS. Radiohalogens for imaging and therapy. Chem Soc Rev 2005;34:153–63.