Alkoxyamines for the treatment of cancers

Abstract

The present invention relates to alkoxyamines of general formula (I), and to compounds of general formula (IIa), (IIb), (IIc), (IId), IIe), (IIf) or (IIg), as such and for the treatment of cancers. ##STR00001##

Claims

1. Compound of general formula (IIa), (llb), (IIc), (IId), (IIe), (IIf) or (IIg): ##STR00017## wherein R.sup.1 and R.sup.2 which may be identical or different are chosen from: a secondary alkyl group ##STR00018## or a tertiary alkyl group ##STR00019## wherein X, Y and Z which may be identical or different are chosen amongst a linear or ramified alkyl radical including short and long carbon chains having from 1 to 40 carbon atoms, X, Y and/or Z may be substituted by a functional group chosen from hydroxyl, amine, mercaptan, azide, halogenure, carbonyl chosen from the group consisting of: aldehyde, amide, ketone, acid, ester, and their thio derivatives, aromatic, heteroaromatic, heterocycles, vinyl, alkyne, phosphoryl optionally substituted by a (C.sub.1-C.sub.4) alkoxv radical; a 5 to 12 membered ring, said ring being heterocyclic or not, which can carry the various functions mentioned above; wherein R.sup.1 and/or R.sup.2 are optionally substituted by a stabilizing group or an addressing group W selected from the group consisting of: peptides, sugars, steroids, fatty acids, polyketones, polyphenols, prostaglandins, lipids, bio-receptors, and antigens; R.sup.1 and R.sup.2 being different from R.sup.3; R.sup.3 is chosen from: (i) a secondary alkyl radical ##STR00020## or a tertiary alkyl radical ##STR00021## wherein X, Y and Z may be identical or different, X and Z are a linear or ramified alkyl radical including short and long carbon chains having from 1 to 40 carbon atoms and may be substituted by a functional group chosen from hydroxyl, amine, mercaptan, azide, halogenure, carbonyl chosen from the group consisting of: aldehyde, amide, ketone, acid, ester, and their thio derivatives, aromatic, heteroaromatic, heterocycles, vinyl, alkyne; Y is chosen from an aromatic group or heteroaromatic group, a carbonyl function optionally in a protected form such as enol, acetate, acetals, enamine, an easily oxidable function, a vinyl, an alkene function including short and long carbon chains having from 1 to 40 carbon atoms, an alkyne function including short and long carbon chains having from 1 to 40 carbon atoms, a function thiocarbonyl, a function imine, a function oxime or a function cyano; ##STR00022## wherein Z.sub.1 is selected from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function CO, a thiocarbonyl function CS, an amine or an imine function C(NH)or C(N[C.sub.1-C.sub.4 alkyl])- or an oxime function C(NOH); Z.sub.2 is selected from a single bond, O, S, NR, with R being a linear or ramified C.sub.1-C.sub.40 alkyl radical, a aryl or a heteroaryl group, a linear or ramified C.sub.1-C.sub.40 vinyl radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, those radicals being optionally substituted by a carbonyl function CO, a thiocarbonyl function CS, an amine or an imine function C(NH)or C(N[C.sub.1-C.sub.4 alkyl])- or an oxime function C(NOH); Z.sub.3 is selected from a hydrogen atom; a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 vinyl radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally interrupted by at least one O, S, NH and optionally substituted by a carbonyl function CO, a thiocarbonyl function CS, an amine or an imine function C(NH)or C(N[C.sub.1-C.sub.4 alkyl])- or an oxime function C(NOH); G is either a hydrogen atom or an addressing or stabilizing group W selected from the group consisting of: peptides, sugars, steroids, fatty acids, polyketones, polyphenols, prostaglandins, lipids, bio-receptors, and antigens; ##STR00023## wherein Z.sub.1 and G are as described above; Z.sub.10 and Z.sub.11 are independently selected from O, S, NRand CR.sub.2- where R is as defined above; Z.sub.12 is selected from O, Sand NRwhere R is as defined above; Z.sub.13 is selected from OSO.sub.2R, a halogen atom, an ammonium group, a phosphate group, RSO.sub.2, OH, SH, OR and SR where R is as defined above; Z.sub.14 Z.sub.15 and Z.sub.16 are independently selected from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function CO, a thiocarbonyl function CS, an amine or an imine function C(NH)or C(N[C.sub.1-C.sub.4 alkyl])- or an oxime function C(NOH); ##STR00024## wherein Z.sub.1 Z.sub.12 Z.sub.13 and G are as described above; Z.sub.19 is selected from CH.sub.2, CHRand CR.sub.2where R is as defined above; Z.sub.20 and Z.sub.21 are independently selected from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally interrupted by at least one O, S, NH and optionally substituted by a carbonyl function CO, a thiocarbonyl function CS, an amine or an imine function C(NH)or C(N[C.sub.1-C.sub.4 alkyl])- or an oxime function C(NOH) and wherein W is independently of one another selected from the group consisting of: peptides, sugars, steroids, fatty acids, polyketones, polyphenols, prostaglandins, lipids, bio-receptors, and antigens.

2. Compounds according to claim 1 wherein R.sup.3 is selected from: ##STR00025## wherein Z.sub.1, Z.sub.2 and G are as described previously; and Z.sub.4 is chosen from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function (CO), a thiocarbonyl function (CS), an amine or an imine function (C(NH)or C(N[C.sub.1-C.sub.4 alkyl])-) and an oxime function (C(NOH)); ##STR00026## wherein Z.sub.1, Z.sub.4, R and G are as described previously; Z.sub.5 is selected from O, S, NHand NRwhere R is as defined above; Z.sub.6 is chosen from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function (CO), a thiocarbonyl function (CS), an amine or an imine function (C(NH)or C(N[C.sub.1-C.sub.4 alkyl])-) and an oxime function (C(NOH)); and Z.sub.7 is selected from OR, SR, OH and SH where R is as defined above; ##STR00027## wherein Z.sub.1, Z.sub.4, Z.sub.6 and G are as described above; Z.sub.8 is selected from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function (CO), a thiocarbonyl function (CS), an amine or an imine function (C(NH)or C(N[C.sub.1-C.sub.4 alkyl])-) and an oxime function (C(NOH)); and Z.sub.9 is selected from H and R where R is as defined above; ##STR00028## wherein Z.sub.1, Z.sub.6, Z.sub.10, Z.sub.13 and G are as described above; Z.sub.17 is selected from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally interrupted by at least one O, S, NH and optionally substituted by a carbonyl function (CO), a thiocarbonyl function (CS), an amine or an imine function ((C(NH)or C(N[C.sub.1-C.sub.4 alkyl])-), an oxime function (C(NOH)) ; an aryl or a heteroaryl group optionally linked to at least one O, S, NH; Z.sub.18 is selected from O, S, NH, NR, CH.sub.2, CHRand CR.sub.2where R is as defined above; ##STR00029## wherein Z.sub.1, Z.sub.4, Z.sub.5, Z.sub.6, Z.sub.13 and G are as described above; Z.sub.22 and Z.sub.23 are selected independently from a hydrogen atom, a linear or ramified C.sub.1-C.sub.40 alkyl radical, a linear or ramified C.sub.1-C.sub.40 alkene radical, a linear or ramified C.sub.1-C.sub.40 alkyne radical, an aryl group, a heteroaryl group, those radicals being optionally substituted by a carbonyl function (CO), a thiocarbonyl function (CS), an amine or an imine function (C(NH)or C(N[C.sub.1-C.sub.4 alkyl])-) or an oxime function (C(NOH)).

3. Pharmaceutical composition comprising at least one compound according to claim 1 and a physiologically acceptable vehicle.

4. Method for in vivo monitoring the curing of a solid tumor comprising the steps of: (a) administering at least a compound according to claim 1; and (b) visualizing the site at which prodrug activation occurs and the amount of drug deposit inside the tumor by Magnetic Resonance Imaging (MRI) enhanced by dynamic nuclear polarization or Electron Paramagnetic Resonance Imaging (EPRI).

5. Method for treating a solid tumor comprising the step of: administering a therapeutically effective amount of a compound of claim 1 to a subject in need thereof.

Description

(1) FIGS. 1 to 7 illustrate the results of assays described in Example II:

(2) FIG. 1 shows the viability of cells U87 treated during 1 hour with several concentrations of the tested alkoxyamine of Example I previously hydrolyzed;

(3) FIG. 2 shows the rate of oxidative stress in U87 cells treated 1 hour with 3 mM of the alkoxyamine of Example I (ALK-1);

(4) FIG. 3 shows the viability of cells U87 treated during 1 hour with 3 mM of the alkoxyamine of Example I with and without scavenger;

(5) FIG. 4 shows the viability of cells U87 at several times after treatment with several concentrations of the tested alkoxyamine of Example I previously hydrolyzed.

(6) FIG. 5 illustrates mitochondria perturbations induced by ALK-1. ALK-1 effects on the mitochondria were measured using TMRE (grey bars) illustrating the variations in mitochondrial membrane potential and NAO (dashed bars) to assess the changes in mitochondrial morphology. U87 cells were incubated with concentrations of ALK-1 between 0.1 to 1.5 mM and mitochondria perturbations were observed by flow cytometry at 3 h (A), 24 h (B) or 72 h (C) observation times (n=3);

(7) FIG. 6 illustrates Quantitative analysis of cell apoptosis induced by ALK-1 given as the percentage of U87 cells in different stages against concentrations of ALK-1 at 3 h (A), 24 h (B) and 72 h (C). Percentages of U87 cells stages were determined by PI and annexin-V staining (viable cells, IP/AV, grey bars; early apoptotic cells, IP/AV+, blue bars; and late apoptotic/necrotic cells, PI+/AV+, orange bars). The percentages of positive cells for caspase activity (caspase+) were represented by the superimposed red line ().

(8) FIG. 7 illustrates the monitoring the ALK-1 homolysis by OMRI. The release of SG-1 nitroxide free radical upon 0.8 mM ALK-1 homolysis at 37 C. was monitored by Overhauser-enhanced Magnetic Resonance Imaging. (A) MRI of ALK-1 homolysis in the absence (top) or presence (bottom) of EPR irradiation. (B) Overhauser enhancement (signal with EPR on/signal with EPR off) over time. The inset shows the EPR spectrum of SG-1. The red star indicates the frequency used for EPR irradiation.

EXAMPLES

I. Synthesis of an Alkoxyamine of General Formula (I): N-methyl 4-(1-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethylpropyl)amino)oxy)ethyl)pyridin-1-ium 4-methylbenzenesulfonate (ALK-1)

(9) To a stirred suspension of CuBr (270 mg, 1.88 mmol, 0.55 equiv.) and Cu (239 mg, 3.76 mmol, 1.1 equiv) in degassed benzene (4.0 mL) was added N,N,N,N,N-pentamethyldiethylenetriamine (393 L, 1.88 mmol, 0.55 equiv.). The resulting mixture was stirred under argon at room temperature for 30 min then a solution of 4-(1-bromoethyl)pyridine1 (700 mg, 3.76 mmol, 1.1 equiv.) and SG1 (1.0 g, 3.42 mmol, 1.0 mmol) in degassed benzene (4.0 mL) was slowly added. The mixture was stirred overnight under argon. It was then diluted with ethyl acetate, filtered and washed several times with saturated aqueous ammonia solution, water and brine. After drying with Na2SO4, filtration and concentration, column chromatography on silica gel (eluent:gradient of ethyl acetate/pentane) gave diethyl(1-(tert-butyl(1-(pyridin-4-yl)ethoxy)amino)-2,2-dimethylpropyl)phosphonate.

(10) To a stirred solution of diethyl(1-(tert-butyl(1-(pyridin-4-yl)ethoxy)amino)-2,2-dimethylpropyl)phosphonate (500 mg, 1.25 mmol, 1.0 equiv.) in THF (12.5 mL) was added methyl sulfonate (257 mg, 1.38 mmol, 1.1 equiv.). The resulting mixture was stirred at room temperature under argon for 12 h. It was then concentrated in vacuo and triturated with an ether/pentane mixture (1/1 v/v) to give 610 mg (1.04 mmol, 83% yield) N-methyl 4-(1-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethylpropyl)amino)oxy)ethyl)pyridin-1-ium 4-methylbenzenesulfonate:

(11) ##STR00016##

(12) .sup.1H NMR (400 MHz, CDCl.sub.3): =9.21 (br d, J=5.3 Hz, 2H, min), 9.09 (d, J=6.0 Hz, 2H, Maj), 7.90 (d, J=6.0 Hz, 2H, Maj), 7.78 (br d, J=5.3 Hz, 2H, min), 7.73 (br d, J=8.0 Hz, 2H, min and Maj), 7.12 (br d, J=8.0 Hz, 2H, min and Maj), 6.33 (br s, 2H, min), 6.27 (d, J=13.9 Hz, 1H, Maj), 6.23 (d, J=13.9 Hz, 1H, Maj), 5.28-5.19 (m, 1H, min and Maj), 4.57 (s, 3H, min), 4.54 (s, 3H, Maj), 4.39-4.24 (m, 2H, min), 4.11-3.98 (m, 2H, min), 3.96-3.86 (m, 2H, Maj), 3.75-3.65 (m, 2H, Maj), 3.43 (d, J=27.2 Hz, 1H, Maj), 3.23 (d, J=26.4 Hz, 2H, min), 2.32 (s, 3H, min and Maj), 1.56 (d, J=6.8 Hz, 3H, min), 1.51 (d, J=6.8 Hz, 3H, Maj), 1.35-1.29 (m, 6H, min), 1.22-1.18 (m, 3H, Maj), 1.19 (s, 9H, min and Maj), 1.12 (s, 9H, Maj), 1.12 (t, J=7.1 Hz, 3H, Maj), 0.87 (s, 9H, min). .sup.13C NMR (75 MHz, CDCl.sub.3): =160.8, 152.7, 151.1, 148.0, 147.8, 144.5, 143.5, 142.9, 137.4, 130.6, 128.5, 127.1, 126.5, 124.4, 124.0, 121.1, 120.6, 75.7, 75.3, 68.1 (d, J=139.0 Hz), 67.5 (d, J=139.2 Hz), 60.3-59.8 (m), 58.4 (d, J=7.5 Hz), 67.6 (d, J=7.5 Hz), 34.1-33.7 (m), 29.3, 29.2 (d, J=5.8 Hz), 28.6-28.5 (m), 26.7, 22.5, 20.1, 19.8, 19.7, 15.4-14.6 (m). .sup.31P NMR (162 MHz, CDCl.sub.3): =24.0 (min), 23.4 (Maj). HRMS (ESI) m/z calcd for C.sub.21H.sub.40N.sub.2O.sub.4P.sub.1 (M).sup.+ 415.2720, found 415.2711.

II. Use of an Alkxyamine According to Example I (ALK-1) to Kill Glioma Cells Through the Generation of a Reactive Alkyl Radical

(13) II.A. Material & Method

(14) Cell culture: Human glioblastoma-astrocytoma cell line U87 MG from the American Type Culture Collection (reference ATCC-HTB-14) were cultured in Dulbecco's modified eagles medium (DMEM, Gibco Corp) supplemented with 10% fetal calf serum (FCS, Gibco Corp), in a humidified atmosphere with 5% CO2 at 37 C.

(15) Viability Test: The viability study of U87 cells treated or not with the alkoxyamine was measured by the LIVE/DEAD Viability/Cytotoxicity Kit (Molecular Probes, Invitrogen, Life Technologies). Briefly, the cells were allowed to grow during 24 h on 24 multiwell plates (Becton-Dickinson, 2 cm.sup.2/well). They were then treated with the alkoxyamine at concentrations ranging from 0.1 to 3 mM and tested for viability at distinct observation times (1, 3, 6, 24 or 72 h). To do this the adherent cells were dissociated from the plate by trypsinization and pelleted by centrifugation. Then, the cells were resuspended at 10.sup.6 cells/mL on calcein-am and ethidium homodimer-1 solution during 20 minutes at room temperature and analyzed on a Guava easyCyte flow cytometer/counter (Millipore). U87 cells viability was also measured after treatment with previously homolysed alkoxyamine. For these experiments, the alkoxyamine was incubated in the culture medium during 72 h at 6 mM without the cells. U87 cells were then treated with different concentrations of this homolysed alkoxyamine for 1 h and a viability test was performed as previously described. The viability test in the presence of the free radical scavenger was carried out with an alkoxyamine concentration of 3 mM and an incubation time of 1 h. U87 cells viability was measured versus increasing concentrations the radical scavenger, trihydroxyethylrutin (Santa Cruz Biotechnology, inc.) namely, 20, 50 or 100 mM.

(16) Oxidative stress: The cells were grown in a 24 multiwell plate as previously described. Then, cells were dissociated using 0.05% trypsin (Invitrogen, Carlsbad, Calif., USA) and suspended at a density of 110.sup.6 cells/mL in carboxy-H.sub.2DCFDA (Molecular Probes, Invitrogen, Life Technologies) at 50 M. After 30 minutes incubation at 37 C., cells were washed with phosphate buffered saline (PBS). Cells were centrifuged and resuspended with cell culture media containing or not 3 mM alkoxyamine with or without trihydroxyethylrutin at 20 or 50 mM. After incubation time the cells were washed and analyzed on a Guava easyCyte flow cytometer/counter.

(17) Mitochondrial alterations measurements. The red fluorescent dye tetramethylrhodamine ethyl ester (TMRE, Molecular Probes, Invitrogen, Life Technologies) was used to follow variations in mitochondrial membrane potential. This cationic probe accumulates in polarized mitochondria through the electrochemical gradient. Nonyl Acridine Orange (NAO, Molecular Probes, Invitrogen, Life Technologies) was then used as a complementary assay to assess the changes in mitochondrial morphology. Cells were plated on 24 multi-well plates (Becton-Dickinson, 2 cm.sup.2/well) during 24 h and treated with a range of ALK-1 concentrations (0.1; 0.25; 0.375; 0.5; 0.75; 1.0 and 1.5 mM). After 3, 24 or 72 h, cells were resuspended in TMRE (250 nM) or NAO (500 nM) and incubated 30 minutes at 37 C., 5% CO.sub.2. Incubation with 0.5 mM of Carbonyl cyanide 3-chlorophenylhydrazone (CCCP, Sigma-Aldrich) was performed as positive control of the mitochondrial depolarization staining. Then, cells were washed and analyzed by flow cytometry. All the experiments were performed three times.

(18) Cell death analysis by annexin V/propidium iodide staining. Cell death was detected by annexin V-FITC (Molecular Probes, Invitrogen, Life Technologies) binding to exposed phosphoserine (PS) residues at the surface of cells. Cells were treated with ALK-1 at concentrations ranging from 0.1 to 1.5 mM at 37 C. and tested at distinct observation times (3, 24 or 72 h). After treatment, cells were re-suspended in staining buffer containing propidium iodide (PI, 2 g/mL) and annexin V-FITC. Double-labeling was performed at room temperature for 15 min in darkness. Then, the percentage of viable (IP/AV), early apoptotic (IP/AV+) and late apoptosis/necrotic cells (IP+/AV+) was quantified by flow cytometry. These experiments were performed twice.

(19) Apoptosis detection through caspase-3/-7 assay. Quantitative assessment of apoptotic cells was also conducted by the detection of caspase activity using the Vybrant FAM Caspase-3 and -7 Assay Kit (Molecular Probes, Invitrogen, Life Technologies). Briefly, cells were treated with the previously described ALK-1 concentrations during 3, 24 or 72 h. After the alkoxyamine treatment, the cells were dissociated from the support, washed and incubated 1 h at 37 C. in FLICA working solution. Then, the cells were washed and green fluorescence was measured by flow cytometry. These experiments were performed twice.

(20) Overhauser-Enhanced Magnetic Resonance Imaging.

(21) EPR Cavity and MRI devices. The OMRI experiment were done in a C-shaped 0.2 T MRI system (Magnetom Open Viva, Siemens, Erlangen, Germany) and a resonant TE011 transverse electric mode EPR cavity setup (Bruker, Wissembourg, France) as described previously. The EPR cavity, placed at the center of the magnet, was used to saturate the electron spin transition of the nitroxide SG1 produced upon ALK-1 homolysis. A homemade saddle-shaped MRI coil (28 mm in diameter and 29 mm in length) in the EPR cavity was used for imaging.

(22) Electron spin saturation was carried out at 5.4573 GHz, corresponding to the first line at high field from the center of the EPR spectrum. The proton frequency was 8.24 MHz. Sample temperature was kept at 37 C.

(23) OMRI experiments were performed in two NMR tubes (4 mm inner diameter): one containing 0.8 mM of SG1 nitroxide in phosphate buffer saline (not shown) and the other with 0.8 mM of ALK-1 in DMEM, 10% FCS. ALK-1 homolysis was followed for 48 hours.

(24) Pulse sequences. 2D MRI images were acquired with a standard gradient echo sequence, which was synchronized to an external pulse generator for electron spin saturation. The EPR pulse time was 260 ms long, followed immediately by the MRI sequence. This sequence had the following parameters: TE (echo time)=10 ms; TR (repetition time) minimal=27 ms; Effective TR=300 ms; Field of view=2222 mm; Matrix size=6464; Slice thickness=5 mm; Spatial resolution=0.340.34 mm, Number of averages=2 and an acquisition time=22 s. All MR adjustments were done manually, using the same fixed receiver amplification gain for both measurements, without (S.sub.off) and with (S.sub.on) HF irradiation, so that signals can be directly compared and Overhauser enhancements (S.sub.on/S.sub.off) calculated.

(25) Post-processing. All signal intensity measurements were made with ImageJ imaging software (ImageJ, National Institutes of Health, USA). Signal intensity was measured in a rectangular region of interest of 2 mm.sup.2 positioned in the NMR tube area. Curve fitting and t.sub.1/2 measurement were carried out with IGOR Pro (Wavemetrics, Lake-Oswego, Oreg., USA).

(26) II.B. Results

(27) N-methyl 4-(1-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethylpropyl)amino)oxy)ethyl)pyridin-1-ium 4-methylbenzenesulfonate (ALK-1) has an homolysis half-life time of 50 mn at 37 C.; 3 mM of this compound is applied one hour to U87 glioblastoma cell cultures in the presence of increasing concentrations of a non toxic polyphenolic free radical scavenger. The viability of the cells was then measured by cytometry with the combined ethidium bromide/calcein tests. The viability diagram shows that at one hour (one half-life of the alkoxyamine) 3 mM of this alkoxyamine kills approximately 75% of the cells.

(28) However most of the cells (80%) can be saved by the free radical scavenger. This strongly suggests that cell death occurs through a free radical mechanism. This conclusion is further strengthened by observing the oxidative stress created by 3 mM alkoxyamine at one hour with the intracellular probe H2DCFDA.

(29) The results (see FIGS. 1 to 4) clearly demonstrate that cell death occurs through an oxidative stress which is suppressed by the free radical scavenger. It is also correlated to the fact that alkoxyamines at the same concentration but previously homolysed do not carry any toxicity for U87 cells confirming that the freshly produced reactive alkyl radical is required to kill the cells. These preliminary results show that the concept works even with a slow homolysing alkoxyamine at reasonable concentrations generating low transient free radical concentrations.

(30) Mitochondria modifications. To address the ALK-1 effect on mitochondria, the mitochondrial potential was monitored using tetramethylrhodamine ester (TMRE). In addition, the mitochondrial morphology changes were observed through cardiolipin accessibility by nonyl acridine orange (NAO) staining. For all the tested times of incubation, the CCCP (carbonyl cyanide 3-chlorophenylhydrazone) uncoupling agent generated the expected drop of the mitochondrial potential (FIG. 5). Independently of the incubation time the effect of CCCP on the accessibility of the cardiolipins to the NAO appeared either neutral or positive. Starting from 0.5 mM ALK-1, the mitochondrial potential drops for all times of observation (FIG. 5A-C). Nevertheless, NAO staining displays diverging evolutions as a function of ALK-1 concentration at 24 h (FIG. 5B) and 72 h (FIG. 5C) of incubation. Indeed at 24 h, NAO staining increased for ALK-1 concentrations greater than 0.5 mM while the potential dropped. At 72 h, NAO staining evolution correlates with the mitochondrial potential variations. Commonly, the cardiolipin labeling is described as a mitochondrial mass marker. However, it also depends on the cardiolipin accessibility therefore it reflects the mitochondria morphology. A drop of both cardiolipin labeling and the mitochondrial potential suggests a loss of the mitochondrial content as observed after 72 h incubation in the presence of 0.375 to 0.5 mM (FIG. 5C). An increase of NAO staining associated to a decrease of the mitochondrial potential is better discussed as a change in the mitochondrial morphology as seen at 24 h between 0.5 and 1 mM (FIG. 5B). This is consistent with the fact that the mitochondrial content could not increase within 24 h especially on a stressful situation. A simultaneous increase of both parameters at 72 h and 0.25 mM reveals a transient growth of the mitochondrial mass compatible with the observation time (FIG. 5C).

(31) Apoptosis induction by ALK-1. U87 cells were treated with ALK-1 at various concentrations and observed at 3, 24 and 72 h. Then, cell apoptosis was studied looking at propidium iodide membrane permeability (PI), phosphatidyl serine translocation by annexin-V (AV) staining and caspase-3 and -7 activation (FIG. 6). PI and annexin-V staining discriminate the early apoptotic cells (PI/AV+) from the late apoptotic/necrotic cells (PI+/AV+). Caspase activation would discriminate the necrotic cells (caspase) from the apoptotic cells (caspase+). At 3 h (FIG. 5A), the percentage of PI/AV+ cells increased until 1 mM of alkoxyamine treatment. At 1.5 mM most cells were necrotic or apoptotic cells (PI+/AV+). Percentage of cells carrying activated caspases reached 92% at 1.5 mM ALK-1. At 24 h (FIG. 5B), the percentage of PI/AV+ cells increase from 0.5 mM to 0.75 mM ALK-1. At 1 mM, 94% of the cells were necrotic or apoptotic cells (PI+/AV+). Simultaneously, 88% of the cells were caspase positive. At 72 h (FIG. 5C), a slight rise in the level of PI/AV+ cells was observed at 0.25 mM and 0.375 mM of ALK-1. At 0.5 mM, most cells were PI+/AV+. At this observation time, the number of caspase positive cells varies almost linearly from the lowest ALK-1 concentration of 0.1 mM to reach 80% at 0.5 mM. From the time dependent study, two ALK-1 concentrations appeared noticeable. At 0.25 mM for the 24 h observation time cells did not seem to enter the apoptotic pathway (most cells are PI/AV and caspase). However at 72 h, cells displayed early apoptotic properties and were more than 40% caspase+. At 0.5 mM, cells entered into apoptosis no later than 24 h and after 72 h the remaining cells were massively in late apoptotic state with 80% of the cells being caspase+. These results demonstrated that ALK-1 induced the U87 cells death through apoptosis. It is worth to note that ALK-1 action ended after 5 times the half-life of homolysis, namely 5 hours. However, apoptosis appeared much later for the lowest concentrations. This strongly suggests that some irreversible alterations occurred within five hours which were able to trigger apoptosis between 24 to 72 h after the beginning of alkoxyamine treatment.

(32) Imaging of ALK-1 homolysis by OMRI. The homolysis of ALK-1 was followed by OMRI at 0.8 mM in culture cell medium at 37 C. The signal enhancement due the generated nitroxide radical was monitored for 47 h (FIG. 7). The signal amplification increased from 1-fold at 5 mM to 6-fold at 2825 min after the beginning of ALK-1 homolysis. The signal displays an asymptotic exponential growth as expected from the first order kinetics of homolysis and was thus fitted to the equation (1):
y=y.sub.0Ae.sup.kt (1)
with y.sub.0=5.90.2; A=5.30.3 and k=5.6 10.sup.3 min.sup.1. The experimental t.sub.1/2 was approximately 125 min. This t.sub.1/2 was fairly close to the value calculated from the homolysis activation energy. It afforded E.sub.a=109 kJ/mol, very close to the 106 kJ/mol value reported, taking into account the inaccuracy of the technique. The 2.5 times discrepancy is easily accounted for by the non-linear response of whole OMRI process due to the non-linear effect of the concentration of the nitroxide on the Overhauser enhancement and due to the lack of stability of the actual setup over 48 hours. These results demonstrated that the released nitroxide can be used as a reporter of the radical alkyl owing to the one-to-one stoechiometry.

(33) Although alkoxyamine ALK-1 was only an unrefined lead-compound, it displays a dose-dependent cytotoxic effect. This effect occurs through its homolysis and we showed that an in situ-released alkyl radical was required to induce cell death. Indeed, when the alkyl radical was scavenged, cell viability turned back to normal level. Several effects of alkoxyamines on cells that could lead to cell death were investigated. ALK-1 induced a strong oxidative stress which was also suppressed upon alkyl radical scavenging. Even at low concentrations persistent changes in the mitochondrial potential, mass and/or morphology were observed. Ultimately, as shown by the caspase-3 and -7 activation, membrane integrity alteration and phosphatidyl serine translocation cell death occurred by the apoptotic pathway.

(34) The observation of the viability versus the time after exposition to ALK-1 revealed a delayed toxicity, as compared with the completion of homolysis (about 5 h) and the life-time of the alkyl radical (a few milliseconds). This suggests that the released alkyl radical promptly induced cell alterations that committed the cells into an irreversible cell death process. The various time scales observed for oxidative stress, mitochondrial alterations and the development of cells apoptosis suggest the absence of a single sequential link between these phenomena. It rather indicates a direct action from the alkyl radical on each effect without excluding interactions.

(35) It has been shown here that the nitroxide radical released from the alkoxyamine homolysis could be efficiently detected by OMRI. This method is currently developed in order to enhance MRI specificity. The nitroxide stability would allow the monitoring of alkoxyamine homolysis in vivo, with an accurate real-time localization in 3D. Moreover, a longitudinal follow-up of the treatment could be achieved through standard MRI modality.