FLUORESCENT PROBE AND APPLICATIONS THEREOF

Abstract

##STR00001##

The invention relates a fluorescent compound of formula I: wherein A is selected from P, P═O and N; .Math.R.sup.1 is a residue comprising an oxygen atom.Math.R.sup.2 is a residue comprising an oxygen atom, or a halogen, .Math.R.sup.3, R.sup.4 and R.sup.5 are alkyls, possibly substituted or a salt or a solvate thereof. The compound is useful as a fluorescent probe sensitive to membrane fluidity and for diagnosing cancer.

Claims

1. Fluorescent compound of formula I: ##STR00024## wherein A is selected from P, P═O and N; R.sup.1 is selected from the group comprising hydrogen atom, NO.sub.2, NR.sup.6.sub.2, CH.sub.3, OR.sup.6, SO.sub.3H; Cl, F, CO.sub.2H, CONR.sup.6.sub.2, O.sub.2CR.sup.6, NHCOR.sup.6; NHCO.sub.2R.sup.6 and NHCONR.sup.6.sub.2 R.sup.2 is selected from the group comprising hydrogen atom, NO.sub.2, NH.sub.2, CH.sub.3, OH, SO.sub.3H and halogen; R.sup.3 is a C3-C25 alkyl, linear or branched, saturated or not, cyclic or not; R.sup.4 and R.sup.5 are selected from the group consisting of: R.sup.4 and R.sup.5 are either together and independently from each other a C2-C6 alkyl, linear or branched, saturated or not, cyclic or not, substituted by at least a hydroxyl group, R.sup.4 is a C3-C6 alkyl linear or branched, saturated or not, cyclic or not, substituted by at least two hydroxyl groups and R.sup.5 is a C1-C6 alkyl linear or branched, saturated or not, cyclic or not, not substituted, and R.sup.5 is a C3-C6 alkyl linear or branched, saturated or not, cyclic or not, substituted by at least two hydroxyl groups and R.sup.4 is a C1-C6 alkyl linear or branched, saturated or not, cyclic or not, not substituted; and R.sup.6 is selected in the group comprising hydrogen atom and a C1-C6 alkyl, linear or branched, saturated or not, cyclic or not, or a salt or a solvate thereof, preferably said fluorescent compound being of formula I-a: ##STR00025## wherein R.sup.3, R.sup.4 and R.sup.5 are as defined above.

2. The fluorescent compound according to claim 1 wherein said fluorescent compound is of formula I-a: ##STR00026## wherein: A is N; R.sup.3 is as recited in claim 1; R.sup.4 and R.sup.5 are selected from the group consisting of: R.sup.4 and R.sup.5 are either together and independently from each other a C2-C6 alkyl linear saturated, substituted by at least a hydroxyl group, R.sup.4 is a saturated C3-C6 alkyl, linear or branched, substituted by at least two hydroxyl groups and R.sup.5 is a C1-C6 alkyl, linear or branched, saturated, not substituted, and R.sup.5 is a C3-C6 alkyl, linear or branched, saturated substituted by at least two hydroxyl groups and R.sup.4 is a C1-C6 alkyl, linear or branched, saturated, not substituted.

3. The fluorescent compound according to claim 1, wherein R.sup.4 and R.sup.5 are together and independently from each other ##STR00027## or R.sup.4 is ##STR00028## and R.sup.5 is ##STR00029## or R.sup.4 is ##STR00030## and R.sup.5 is ##STR00031## wherein n and m are together and independently from each other an integer from 1 to 4, and wherein the sum of n and m is an integer from 2 to 5.

4. The fluorescent compound according to claim 1 of the following formula: ##STR00032##

5. Method for in vitro and ex vivo labelling a lipid membrane or part thereof comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of the said lipid membrane or part thereof labelled by the said fluorescent compound.

6. The method for in vitro and ex vivo labelling a lipid membrane according to claim 5, wherein the lipid membrane or part thereof is selected from a lipid monolayer membrane and a lipid bilayer membrane.

7. The method in vitro and ex vivo for labelling a lipid membrane according to claim 5, wherein the lipid membrane is a eukaryote cell lipid membrane.

8. (canceled)

9. Method for in vitro and ex vivo evaluating the degree of fluidity of a lipid membrane, said method comprising the steps of: a) labelling a lipid membrane or part thereof with a fluorescent compound according to claim 1 by a method comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of said lipid membrane or part thereof labelled by the said fluorescent compound, b) determining a generalized polarization parameter GP of the lipid membrane or part thereof by the following formula $\mathrm{GP}=\frac{I_{\mathrm{gel}}-I_{\mathrm{lc}}}{I_{\mathrm{gel}}+I_{\mathrm{lc}}}$ wherein I.sub.gel and I.sub.lc are intensities of fluorescence measured by fluorimetry from the said lipid membrane or part thereof labeled by said fluorescent compound, the measurement of said intensity I.sub.gel being performed at a wavelength characteristic of said fluorescent compound in gel lipid phase, and the measurement of said intensity lie being performed at a wavelength characteristic of said fluorescent compound in liquid crystal lipid phase, c) concluding that the lipid membrane is fluid if GP is lower or equal to 0, or concluding that the first lipid membrane is rigid if GP is higher than 0.

10. Method for in vitro and ex vivo evaluating the degree of fluidity of a lipid membrane, said method comprising the steps of: a) labelling a lipid membrane or part thereof with a fluorescent compound according to claim 1 by a method comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of said lipid membrane or part thereof labelled by the said fluorescent compound, b) determining a generalized polarization parameter GP of the lipid membrane or part thereof by the following formula $\mathrm{GP}=\frac{I_{\mathrm{gel}}-I_{\mathrm{lc}}}{I_{\mathrm{gel}}+I_{\mathrm{lc}}}$ wherein I.sub.gel and I.sub.lc are intensities of fluorescence measured by fluorimetry from the lipid membrane or part thereof labeled by said fluorescent compound, the measurement of said intensity I.sub.gel being performed at a wavelength characteristic of said fluorescent compound in gel lipid phase, and the measurement of said intensity I.sub.lc being performed at a wavelength characteristic of said fluorescent compound in liquid crystal lipid phase, c) comparing said GP with a reference generalized polarization parameter GP.sub.R from a reference lipid membrane, and d) concluding that the lipid membrane is more fluid or has a higher fluidity compared to the reference lipid membrane if GP is lower than GP.sub.R, concluding that the first lipid membrane is more rigid or has a lower fluidity compared to the reference lipid membrane if GP is higher than GP.sub.R, or concluding that the first lipid membrane has the same fluidity or degree of fluidity compared to the reference membrane if GP is equal to GP.sub.R.

11-12. (canceled)

13. Method for the in vitro and ex vivo diagnosis of a cancer in an individual, said method comprising the steps of: a) labelling a cell membrane or part thereof in a sample of an individual with a fluorescent compound according to claim 1 by a method comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of said lipid membrane or part thereof labelled by the said fluorescent compound, b) determining a generalized polarization parameter GP of said cell membranes or part thereof by the following formula: $\mathrm{GP}=\frac{I_{\mathrm{gel}}-I_{\mathrm{lc}}}{I_{\mathrm{gel}}+I_{\mathrm{lc}}}$ wherein I.sub.gel and I.sub.lc are intensities of fluorescence measured by fluorimetry from cell membranes or part thereof labeled by said fluorescent compound, the measurement of said intensity I.sub.gel being performed at a wavelength characteristic of said fluorescent compound in gel lipid phase, and the measurement of said intensity lie being performed at a wavelength characteristic of said fluorescent compound in liquid crystal lipid phase, c) comparing said GP with a reference generalized polarization parameter GP.sub.R of said cell membranes or part thereof, and d) concluding that said individual is afflicted by a cancer if said GP is lower than said GP.sub.R, or concluding that said individual is not afflicted by a cancer if said GP is equal or higher than said GPR.

14. (canceled)

15. Method for in vitro and ex vivo evaluation of the response to a therapeutic treatment of a patient afflicted with a cancer, said therapeutic treatment being directed to said cancer, said method comprising the steps of: a) labelling a cell membrane or part thereof in a sample from a tumour of said patient obtained after said therapeutic treatment, with a fluorescent compound according to claim 1 by a method comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of said lipid membrane or part thereof labelled by the said fluorescent compound, b) determining a generalized polarization parameter GP of said cell membrane or part thereof by the following formula: $\mathrm{GP}=\frac{I_{\mathrm{gel}}-I_{\mathrm{lc}}}{I_{\mathrm{gel}}+I_{\mathrm{lc}}}$ wherein I.sub.gel and I.sub.lc are intensities of fluorescence measured by fluorimetry from said cell membrane or part thereof labeled by said fluorescent compound, the measurement of said intensity I.sub.get being performed at a wavelength characteristic of said fluorescent compound in gel lipid phase, and the measurement of said intensity lie being performed at a wavelength characteristic of said fluorescent compound in liquid crystal lipid phase, c) comparing said GP with a reference generalized polarization parameter GP.sub.R of cell membranes or part thereof, said GP.sub.R being a prior determination of GP according to steps a) and b) from cell membrane or part thereof in a sample from the tumour of said patient obtained before said therapeutic treatment, and d) concluding that said patient is considered to be responsive to the said therapeutic treatment if GP is higher than GP.sub.R, or concluding that patient is not considered to be responsive to the said therapeutic treatment if GP is lower or equal to GP.sub.R.

16. Method for quantification of lipids organized in at least one lipid membrane or part thereof in an aqueous solution, comprising the steps of: a) labelling a lipid membrane or part thereof with a fluorescent compound according to claim 1 by a method comprising a step of contacting a fluorescent compound according to claim 1 with a lipid membrane or part thereof, and a step of detecting the fluorescence of said lipid membrane or part thereof labelled by the said fluorescent compound, b) measuring the intensity of fluorescence I.sub.lc by fluorimetry, typically by using a spectrofluorometer or a photomultiplicator, at a wavelength characteristic of said fluorescent compound in gel lipid phase, c) determining the concentration of said at least one lipid membrane or part thereof, by reporting the value obtained in step b) on a calibration curve of the fluorescence intensity as function of the concentration, the calibration curve being a hyperbolic function of the formula y=ax/(b+x) with y being the intensity of fluorescence and x the concentration of the at least one lipid membrane.

Description

LEGEND TO THE FIGURES

[0197] FIG. 1 represents the fluorescence emission spectra (arbitrary unit) of DIOLL in comparison with Laurdan and C-Laurdan in organic solvents of increasing polarity. Full lines represent DIOLL; dotted lines represent Laurdan; dash lines represent C-Laurdan. A represents the Toluene condition; B represents the Chloroform condition; C represents the DMSO condition; D represents the Ethanol condition.

[0198] FIG. 2A represents the fluorescence emission spectra (arbitrary unit) at excitation wavelength 390 nm of Laurdan probe in DMPC liposomes at different temperatures conditions (y-axis is the normalized fluorescent intensity; x-axis is the emission wavelength in nm). A presents the fluorescence emission spectra at 5° C.; B represents the fluorescence emission spectra at 9° C.; C represents the fluorescence emission spectra at 13° C.; D represents the fluorescence emission spectra at 18° C.; E represents the fluorescence emission spectra at 23° C.; F represents the fluorescence emission spectra at 28° C.; G represents the fluorescence emission spectra at 33° C.; H represents the fluorescence emission spectra at 40° C.; I represents the fluorescence emission spectra at 55° C.; J represents the fluorescence emission spectra at 65° C.

[0199] FIG. 2B represents the fluorescence emission spectra (arbitrary unit) at excitation wavelength 390 nm of C-Laurdan probe in DMPC liposomes at different temperatures conditions (y-axis is the normalized fluorescent intensity; x-axis is the emission wavelength in nm). A presents the fluorescence emission spectra at 5° C.; B represents the fluorescence emission spectra at 9° C.; C represents the fluorescence emission spectra at 13° C.; D represents the fluorescence emission spectra at 18° C.; E represents the fluorescence emission spectra at 23° C.; F represents the fluorescence emission spectra at 28° C.; G represents the fluorescence emission spectra at 33° C.; H represents the fluorescence emission spectra at 40° C.; I represents the fluorescence emission spectra at 55° C.; J represents the fluorescence emission spectra at 65° C.

[0200] FIG. 2C represents the fluorescence emission spectra (arbitrary unit) at excitation wavelength 390 nm of DIOLL probe in DMPC liposomes at different temperatures conditions (y-axis is the normalized fluorescent intensity; x-axis is the emission wavelength in nm). A presents the fluorescence emission spectra at 5° C.; B represents the fluorescence emission spectra at 9° C.; C represents the fluorescence emission spectra at 13° C.; D represents the fluorescence emission spectra at 18° C.; E represents the fluorescence emission spectra at 23° C.; F represents the fluorescence emission spectra at 28° C.; G represents the fluorescence emission spectra at 33° C.; H represents the fluorescence emission spectra at 40° C.; I represents the fluorescence emission spectra at 55° C.; J represents the fluorescence emission spectra at 65° C.

[0201] FIG. 2D represents the GP values (arbitrary unit) of Laurdan (A), C-Laurdan (B) and DIOLL (C) probes as a function of temperature (in ° C.).

[0202] FIG. 2E represents the fluorescence emission spectra (arbitrary unit) at excitation wavelength 390 nm of Laurdan (a), C-Laurdan (b) and DIOLL probe (c) in POPC liposomes at 21° C. (y-axis is the normalized fluorescent intensity; x-axis is the emission wavelength in nm). Liposome concentration is 40 μg/ml and probe concentration is 0.2 μM.

[0203] FIGS. 3A and 3B represent the membrane fluidity estimation with DIOLL probe using fluorescence spectroscopy and flow cytometry. In FIG. 3A, GP values (y-axis, arbitrary unit) are measured from the spectra of DIOLL inserted in liposomes of increasing fluidity. In FIG. 3B, the number of events (y-axis) as a function of the ratio 525/450 fluorescence intensity values (x-axis) is measured from liposomes of increasing fluidity. For FIGS. 3A and 3B, A represents liposome DSPC; B represents liposome DPPC; C represents liposome DMPC; D represents liposome POPC, E represents liposome DOPC; F represents liposome DLPC.

[0204] FIG. 4 represents a fluorescence image of HMEC cells incubated with DIOLL probe. Excitation was performed at 405 nm. This image is an overlay of images recovered at emission 424 nm, 432 nm, 441, 468 nm, 477 nm, 486 nm, 495 nm, 504 nm and 513 nm. In this image, the shades of grey correspond to different fluorescence intensity obtained for each emission which leads to the differentiation of the fluidity state of the cell membranes: light grey corresponds to fluidic membrane while dark grey corresponds to rigid membrane.

[0205] FIG. 5 represents the mean GP values (arbitrary unit) obtained for different lipid membranes. The standard deviation is also represented for each mean GP values. Data sets are representative of at least 3 independent measurements. A represents the plasma membrane, B represents the endoplasmic reticulum, C represents mitochondrial membrane and D represents the nuclear membrane.

[0206] FIG. 6A represents the fluorescence emission intensity (arbitrary unit) of DIOLL probe inserted in the lipid bilayer of POPC liposomes as a function of the emission wavelength (in nm). Excitation was performed at 390 nm. 2 μM of DIOLL probe was added to increasing liposome concentrations. 1 represents 0 μg.ml.sup.−1, 2 represents 5 μg.ml.sup.−1, 3 represents 10 μg.ml.sup.−1, 4 represents 15 μg.ml.sup.−1, 5 represents 20 μg.ml.sup.−1, 6 represents 30 μg.ml.sup.−1, 7 represents 40 μg.ml.sup.−1, 8 represents 60 μg.ml.sup.−1 and 9 represents 80 μg.ml.sup.−1 POPC liposomes.

[0207] FIG. 6B represents the maximum fluorescence intensity (arbitrary unit) at emission at 490 nm as function of lipid concentration (μg.ml.sup.−1) of POPC liposomes labelling by DIOLL probe. Experimental data were fitted to a hyperbolic function: y=2.8*10.sup.6*x/(27+x).

[0208] FIG. 6C represents the surface under the peak of fluorescence (arbitrary unit), calculated from FIG. 6A curves, as function of the lipid concentration (μg.ml.sup.−1) of POPC liposomes labelling by DIOLL probe. Experimental data were fitted to a hyperbolic function: y=29625*x/(25+x).

[0209] FIG. 7A represents the fluorescence emission intensity (arbitrary unit) of DIOLL probe inserted in the lipid bilayer of DMPC liposomes as a function of the emission wavelength (in nm). Excitation was performed at 390 nm. 2 μM of DIOLL probe was added to increasing liposome concentrations. 1 represents 0 μg.ml.sup.−1, 2 represents 5 μg.ml.sup.−1, 3 represents 10 μg.ml.sup.−1, 4 represents 15 μg.ml.sup.−1, 5 represents 20 μg.ml.sup.−1, 6 represents 30 μg.ml.sup.−1, 7 represents 40 μg.ml.sup.−1, 8 represents 60 μg.ml.sup.−1 and 9 represents 80 μg.ml.sup.−1 DMPC liposomes.

[0210] FIG. 7B represents the fluorescence emission intensity (arbitrary unit) of DIOLL probe inserted in the lipid bilayer of DPPC liposomes as a function of the emission wavelength (in nm). Excitation was performed at 390 nm. 2 μM of DIOLL probe was added to increasing liposome concentrations. 1 represents 0 μg.ml.sup.−1, 2 represents 5 μg.ml.sup.−1, 3 represents 10 μg.ml.sup.−1, 4 represents 15 μg.ml.sup.−1, 5 represents 20 μg.ml.sup.−1, 6 represents 30 μg.ml.sup.−1, 7 represents 40 μg.ml.sup.−1, 8 represents 60 μg.ml.sup.−1 and 9 represents 80 μg.ml.sup.−1 DPPC liposomes.

[0211] FIG. 7C represents the surface under the peak of fluorescence (arbitrary unit), calculated from FIG. 7A curves, as a function of the lipid concentration (μg.ml.sup.−1) of DMPC liposomes labelling by DIOLL probe.

[0212] FIG. 7D represents the surface under the peak of fluorescence (arbitrary unit), calculated from FIG. 7B curves, as a function of the lipid concentration (μg.ml.sup.−1) of DPPC liposomes labelling by DIOLL probe.

[0213] FIG. 7E represents the cross calculations of concentrations (μg.ml.sup.−1) of lipids in POPC liposomes (1), in DMPC liposomes (2) and DPPC liposomes (3) using POPC (white bars), DMPC (light grey), DMPC (grey) standard solutions or by Stewart method (black bars). Error bars represent standard deviation (SD) obtained with 3 independent measurements.

[0214] FIG. 8 represents the cholesterol effect on liposome quantification with DIOLL. Quantification of POPC-cholesterol liposomes (μg.ml.sup.−1) is performed using POPC liposomes as standard (white bars): 10 (x-axis) represents POPC-cholesterol liposomes with 10 mass % of cholesterol; 20 (x-axis) represents POPC-cholesterol liposomes with 20 mass % of cholesterol; 30 (x-axis) represents POPC-cholesterol liposomes with 30 mass % of cholesterol. Phospholipid content for each POPC-cholesterol liposome was determined by Stewart method (black bars). Dotted bars represent total lipid content (phospholipid as determined by Stewart method and 10, 20 or 30% cholesterol) for each POPC-cholesterol liposome. Error bars are SD obtained with 3 independent measurements.

[0215] FIG. 9 represents the charge effect on liposome quantification with DIOLL probe. Quantification of POPC-POPS (1:1 molar ratio) liposomes (μg.ml.sup.−1) is performed using POPC liposomes as standard (white bars) and by Steward method (black bars). Error bars are SD obtained with 3 independent measurements.

[0216] FIG. 10 represents the quantification of exosomes. The exosome preparation was quantified (μg.ml.sup.−1) using DIOLL probe and POPC liposomes as standard (white bars). Lipids were extracted with chloroform and quantified with Steward method (black bars). Error bars are SD obtained with 3 independent measurements.

[0217] FIG. 11 represents the quantification of GPMV. The GPMV preparation was quantified (μg.ml.sup.−1) using DIOLL probe and POPC liposomes as standard (white bars). Lipids were extracted with chloroform and quantified with Steward Method (black bars). Error bars are SD obtained with 3 independent measurements.

[0218] FIG. 12A represents the effect of temperature on fluorescence emission of DIOLL-NE. Fluorescence emission spectra at excitation wavelength 390 nm of DIOLL-NE is measured at different temperatures conditions (y-axis is the fluorescent intensity in arbitrary unit; x-axis is the emission wavelength in nm). A presents the fluorescence emission spectra at 5° C.; B represents the fluorescence emission spectra at 7° C.; C represents the fluorescence emission spectra at 9° C.; D represents the fluorescence emission spectra at 11° C.; E represents the fluorescence emission spectra at 13° C.; F represents the fluorescence emission spectra at 15° C.; G represents the fluorescence emission spectra at 17° C.; H represents the fluorescence emission spectra at 19° C.; I represents the fluorescence emission spectra at 21° C.; J represents the fluorescence emission spectra at 23° C. K represents the fluorescence emission spectra at 25° C. L represents the fluorescence emission spectra at 27° C. M represents the fluorescence emission spectra at 37° C.

[0219] FIG. 12B represents the Generalized Polarization (GP) of DIOLL probe inserted in DIOLL-NE as a function of temperature.

[0220] FIG. 13 represents microscopic spectral imaging of HMEC cells after a dynamic change of their membrane fluidity through the supply in their medium of different type of lipids. Horse serum contains high amount of polyunsaturated lipids whereas palmitate is a well-known saturated lipid increasing the rigidity of the membrane. Scale fluidity and the overall GP of cells after 16 h of treatment in indicated. GP was determined by measuring the fluorescence spectrum of each relevant pixel for membrane staining and by calculating the median of obtained GPs for all these analysed pixels.

[0221] FIGS. 14A and 14B report fluidity measurement of control HMEC (black lane in 14A and (1) in 14B) and their dedifferentiating counterparts (grey lane in 14A and (2) in 14B) where ZEB1 is activated with DIOLL probe using fluorescence spectroscopy at different temperature and flow cytometry, respectively. In FIG. 14A, the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of HMEC cells is represented as a function of the temperature at which the I.sub.440 and I.sub.400 were recorded. Error bars are SD obtained with 3 independent measurements. In FIG. 14B, the number of events (y-axis) as a function of the ratio 525/450 fluorescence intensity values (x-axis) is measured.

[0222] FIG. 15 represents the survival test by resazurin of former HMEC, Metastable-generated cell-line and one differentiated cell population (Myoepithelial-like) issue from this Metastable cell line before or after the induction of muted pro-oncogenic protein RAS.

[0223] FIG. 16 represents the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of the cell lines used in FIG. 13 at 25° C.

[0224] FIG. 17 represents a multivariate plot generated by Principal Component Analysis (PCA) of gene expression measured in 50 mammary cell lines and corresponding to two genetic signatures associated to genetic plasticity, and lost of epithelial identity.

[0225] Signatures were established from the RNAseq analysis of the metastable cell line. Intrinsic molecular subtype of each cell line is indicated wherein A represents luminal A cell lines, B represents luminal B cell lines, C represents claudin-low cell lines, D represents her2 cell lines and E represents basal cell lines.

[0226] FIG. 18 represents the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of indicated cancerous cell lines at 31° C. at which I.sub.440 .sup.and I.sub.490 where recorded. Error bars are SD obtained with 4 independent measurements. Significant p-value (T-test) is indicated (*<0,05, **<0,01, ***<0,001).

[0227] FIG. 19 represents the the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane at 37° C. of highly plastic claudin-low cell line (MDAMB231) and lower plastic claudin-low cell line (HS578T) according to previous PCA of genetic plasticity. Error bars are SD obtained with 3 independent measurements. Significant p-value (T-test) is indicated (*<0,05, **<0,01, ***<0,001)

[0228] FIG. 20 represents the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of basal HCC1954 cell line and its more differentiated mesenchymal counterpart produced through nutritive deprivation.

[0229] FIG. 21 represents the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of several prostate cell lines as a function of the temperature at which the I.sub.440 and I.sub.490 were recorded. Black squares represent C4-2B mesenchymal cell line. Black spheres represent LNCaP cell line. Grey triangles represent WPMY-1 cell line. Grey diamonds represent PC-3 cell line.

[0230] FIGS. 22A and 22B represent the distinguishing of exosomes from normal differentiated HMEC and their differentiating counterpart 7 days after ZEB1 activation, with DIOLL probe using fluorescence spectroscopy and flow cytometry, respectively. In FIG. 22A, the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of exosomes from HMEC cells is represented as a function of the temperature at which the I.sub.440 and I.sub.400 were recorded. Black spheres represent exosomes from differentiated HMEC cells. Grey spheres represent exosomes from dedifferentiating HMEC cells. In FIG. 22B, the number of events (y-axis) as a function of the ratio 525/450 fluorescence intensity values (x-axis) is measured from exosomes from differentiated HMEC (1) and exosomes from dedifferentiating HMEC (2).

[0231] FIGS. 23A and 23B represent the distinguishing of exosomes from HMEC cells mediated from 0 to 12 days by ZEB1 transcription factor, with DIOLL probe using fluorescence spectroscopy and flow cytometry, respectively. In FIG. 23A, the Generalized Polarization (GP) (arbitrary unit) of DIOLL probe inserted in the lipid membrane of exosomes from the mediated HMEC cells is represented as a function of the temperature at which the I.sub.440 and I.sub.490 were recorded. Black spheres represent exosomes from original HMEC cells. Black squares represent exosomes from HMEC cells mediated for 3 days. Black triangles exosomes from HMEC cells mediated for 7 days. Grey spheres represent exosomes from HMEC cells mediated for 10 days. Grey squares with a black star represent exosomes from HMEC cells mediated for 12 days. In FIG. 23B, the number of events (y-axis) as a function of the ratio 525/450 fluorescence intensity values (x-axis) is measured for exosomes from original HMEC cells (1), exosomes from HMEC cells mediated for 3 days (2), exosomes from HMEC cells mediated for 7 days (3), exosomes from HMEC cells mediated for 10 days (4) and exosomes from HMEC cells mediated for 12 days (5). For more readability, the origin point of y-axis is moved for each condition 1 to 5.

[0232] FIG. 24 represents the analysis by flow cytometry of primary cancer cells issue from mouse mammary tumor . After tumour dilaceration and digestion, cancer cells were incubated overnight with DIOLL probe to measure membrane fluidity and antibodies to quantify surface makers (CD24 and CD29). In FIG. 24, the different cell populations (normal and malignant) were identified via the surface markers CD24 and CD29 whereas membrane fluidity was determinated for each cell populations wherein A represents the stromal cells, B represents the erythrocytes, C represents the low plastic tumour cells and D represents the highly plastic malignant cells.

[0233] FIG. 25A represents DIOLL fluorescence emission spectra of DIOLL inserted in influenza virus membranes at 21° C. and 37° C.

[0234] FIG. 25B represents GP calculated from spectra in FIG. 25A.

EXAMPLES

Example 1

Synthesis of DIOLL Probe

1. Chemicals and Reagents

[0235] Lauroylchloride and nitrobenzene were purchased from Acros. 2-bromonaphthalene, aluminium chloride, N-methylallylamine, 2-dicyclohexylphosphino2′4′6′triisopropylbiphenyl (X-Phos), Palladium acetate (Pd (Ac)2), N-methylmorpholine oxide monohydrate, catalytic OsO4 2.5% solution in tert-butanol and anhydrous dioxane were purchased from Sigma-Aldrich. Potassium tert-butanolate was purchased from Lancaster. Nitrobenzene and anhydrous dioxane were used without further purification. Flash chromatographies were performed on silica gel purchased from Acros Organics.

2. Synthesis

[0236] Reactions were carried out in dried glassware and under argon atmosphere. All reactions were magnetically stirred and monitored by thin-layer chromatography (TLC) on silica gel 60 F254 (Merck) plates using UV light to visualize the compounds. Yields refer to chromatographically and spectroscopically pure compounds unless otherwise noted. .sup.1H and .sup.13C NMR spectra were recorded at 23° C. using Bruker Avance DRX300, DRX400 or DRX500. Chemical shifts are reported relative to the residue peaks of the solvent (CDCl3: 7.26 ppm for .sup.1H and 77.16 ppm for .sup.13C). The following abbreviations are used to denote the multiplicities: s=singlet, d=doublet, dd=doublet of doublets, t=triplet, m=multiplet and br s=broad singlet. NMR solvents were purchased from Euriso-Top (SaintAubin, France). Low resolution mass spectra (ESI) and HRMS were recorded in positive ion mode using a Thermo Finnigan LCQ spectrometer.

a. Synthesis of 1-(6-bromonaphthalen-2-yl) dodecan-1-one

[0237] 2-bromonaphthalene (2 g, 9.7 mmol) was dissolved in 9.6 mL of nitrobenzene under an Argon atmosphere. The solution was stirred in an ice bath. Lauroylchloride (2.8 mL, 11.64 mmol, 1.2 equiv.) was added drop wise. The reaction mixture was treated with aluminium chloride (1.4 g, 10.67 mmol, 1.1 equiv.) and stirred at room temperature for at least 18 hours. The reaction mixture was quenched with distilled water and extracted three times with ethylacetate. The organic phases were combined, washed with brine, dried over sodium sulphate and concentrated under reduced pressure. The product was purified by crystallization from cold ethanol to obtain a yellowish powder; yield 1.6305 g (43.4%).

[0238] .sup.1H NMR (300 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=6.6 Hz, CH.sub.3); 1.26-1.37 (m, 16H, CH.sub.2); 1.79 (m, 2H, CH.sub.2); 3.07 (t, 2H, J=7.5 Hz, CH.sub.2CO); 7.62 (d, 1H, J=8.7 Hz, H ar); 7.81 (m, 2H, H ar); 8.05 (m, 2H, H ar); 8.42 (s, 1H, H ar).

[0239] .sup.13C NMR: δ=14.1; 22.7; 24.5; 29.4; 29.5; 29.6; 29.7; 31.9; 38.8; 122.7; 125.2; 127.5; 129.4; 129.9; 130.2; 131.0; 134.8; 136.4; 200.2.

[0240] HRMS: m/z (MH+) C.sub.22H.sub.30BrO calculated: 389.1475, measured: 389.1471.

b. Synthesis of 1-(6-(allyl(methyl)amino)naphthalen-2-yl)dodecan-1-one (NMAAL)

[0241] An oven-dried Schlenk flask was evacuated and backfilled with argon. The flask was charged with 1-(6-bromonaphthalen-2-yl) dodecan-1-one (200 mg, 0.514 mmol); (Pd (Ac).sub.2) (9 mg, 41.1 μmol, 0.08 equiv.); X-Phos (39 mg, 82.2 μmol, 0.16 equiv.) and cesium carbonate (503 mg, 1.54 mmol, 3 equiv.). The Schlenk flask was capped with a rubber septum, and then evacuated and backfilled with argon three times. N-methylallylamine (64 μL, 0.668 mmol, 1.3 equiv.) and anhydrous dioxane (1.5 mL) were added through the septum. The mixture was heated to 100° C. in an oil bath and stirred for 15 hours. The mixture was cooled down to room temperature, diluted with ethylacetate, filtered through Celite and concentrated under reduce pressure to obtain a brown solid product. The crude product was purified by flash chromatography on silica gel eluting with ethylacetate/petroleum ether=3/97. A yellow powder (51.5 mg, yield 21%) was obtained. The final product is pure at 80%.

[0242] .sup.1H NMR (300 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=5.1 Hz, CH.sub.3); 1.25-1.37 (m, 16H, CH.sub.2); 1.77 (m, 2H, CH.sub.2); 3.03 (t, 2H, J=7.7 Hz, CH.sub.2CO); 3.10 (s, 3H, NCH.sub.3); 4.07 (d, 2H, J=3.6 Hz, CH.sub.2N) ; 5.17-5.22 (m, 2H, CH.sub.2═CH); 5.89 (m, 1H, CH═CH.sub.2); 6.96 (br s, 1H, H ar); 7.16 (dd, 1H, J=6.6 Hz, J=1.2 Hz, H ar); 7.64 (d, 1H, J=6.6 Hz, H ar); 7.80 (d, 1H, J=6.9 Hz, H ar); 7.93 (dd, 1H, J=6.3 Hz, J=1.2 Hz, H ar); 8.32 (s, 1H, H ar).

[0243] .sup.13C NMR: δ=14.1; 22.7; 24.9; 29.3; 29.5; 29.6; 29.7; 31.9; 38.3; 38.4; 55.1, 116.2; 124.7; 126.1; 129.7; 130.7; 133.0; 137.6; 200.2.

[0244] HRMS: m/z (MH.sup.+) C.sub.26H.sub.38NO calculated: 380.2948, measured: 380.2938.

c. Synthesis of 1-(6-((2,3-dihydroxypropyl)(methyl)amino)naphthalen-2-yl)dodecan-1-one (DIOLL probe, compound of formula 12, referred herein as “DIOLL”)

[0245] The NMAAL (80% purity) product is disposed in 25 mL glassware. 5 mL of acetone, 2 mL of water, 141 mg of N-methylmorpholine oxide monohydrate and 50 μL of catalytic 0504 (100 μL of a 2.5% solution in tert-butanol) are added and the reaction is stirred overnight at room temperature. The mixture is then diluted in ethylacetate, extracted with water (2×30 mL) and washed with brine (30 mL). The organic phase is dried over magnesium sulphate and concentrated under reduce pressure to obtain a crude product purified by flash chromatography on silica gel. Elution is carried first with 100 mL of ethylacetate/petroleum ether (10/90) to eliminate contaminants, then with 100 mL ethylacetate/petroleum ether (50/50) to eliminate intermediate products of reaction and finally with 100 mL of ethylacetate 100% for the recuperation of a dark green pure final product (18.7 mg, yield 41.7%).

[0246] .sup.1H NMR (500 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=7.0 Hz, CH.sub.3); 1.25-1.41 (m, 16H, CH.sub.2); 1.77 (m, 2H, CH.sub.2); 3.03 (t, 2H, J=7.5 Hz, CH.sub.2CO); 3.14 (s, 3H, CH.sub.3N) ; 3.51 (dd, 1H, J=15 Hz, J=5.5 Hz); 3.61 (m, 2H); 3.82 (dd, 1H, J=11.5 Hz, J=3.5 Hz); 4.11 (m, 1H, CHO); 6.95 (br s, 1H, H ar); 7.22 (dd, 1H, J=9.5 Hz, J=2.5 Hz, H ar); 7.63 (d, 1H, J=8.5 Hz, H ar); 7.80 (d, 1H, J=9 Hz, H ar); 7.93 (dd, 1H, J=8.5 Hz, J=1.5 Hz, H ar); 8.32 (s, 1H, H ar).

[0247] .sup.13C NMR: δ=14.1; 22.7; 24.8; 29.4; 29.5; 29.6; 29.7; 31.9; 38.4; 39.5; 55.3; 64.2; 69.8; 105.9; 116.5; 124.8; 125.5; 126.3; 129.7; 130.9; 131.1 ; 137.5; 149.5; 200.3.

[0248] HRMS: m/z (MH.sup.+) C.sub.26H.sub.40NO.sub.3 calculated: 414.3003, measured: 414.2995.

Example 2

Labelling Lipid Membrane with DIOLL Probe

1. Material and Method

[0249] a. Chemicals

[0250] DIOLL probe was synthesis as described in Example 1.

[0251] Laurdan was purchased from GE healthsicences and C-Laurdan was synthesized as described below.

[0252] Lipids were purchased from Avanti Polar Lipids (Alabaster, Al.). PBS was purchased from Sigma-Aldrich. Ultrapure water was obtained with a PureLabOption Q7 system (Veolia Water STI, Decines, France).

[0253] Fetal bovine serum (FBS) was purchased from Eurobio (ref: CVFSVF 00-01), Dulbecco's Modified Eagle's Medium/Nutrient Mixture F-12 (DMEM/F-12) nutrient mixture (Ham) (ref: 31331-028) and penicillin-streptomycin were purchased from Gibco, human epidermal growth factor (EGF) was purchased from Promocell and insulin was purchased from NovoRapid.

[0254] Dil is purchased from life technologies (ref: M7513), Er-tracker is purchased from life technologies, Mito tracker is purchased from life technologies.

b. C-Laurdan Synthesis
i. Synthesis of 1-(6-Bromonaphthalen-2-yl)dodecan-1-one and 1-(6-chloronaphthalen-2-yl)dodecan-1-one

[0255] 2-Bromonaphthalene (4 g, 19.3 mmol) was dissolved in 20 mL of nitrobenzene under an argon atmosphere. The solution was stirred in an ice bath. Aluminium chloride (2.836 g, 21.23 mmol, 1.1 equiv.) was added by portions and lauroyl chloride (5.036 mL, 23.16 mmol, 1.2 equiv.) was added drop- wise. The reaction mixture was stirred at room temperature for 18 hours. The reaction mixture was quenched with distilled water and extracted three times with ethyl acetate. The organic phases were combined, washed with brine, dried over sodium sulfate and concentrated under reduced pressure. The product was purified by crystallization from cold ethanol to obtain a white powder; yield 4.215 g (56%).

[0256] .sup.1H NMR (300 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=6.6 Hz, CH.sub.3); 1.26-1.37 (m, 16H, CH.sub.2); 1.79 (m, 2H, CH.sub.2); 3.07 (t, 2H, J=7.5 Hz, CH.sub.2CO); 7.62 (d, 1H, J=8.7 Hz, H ar); 7.81 (m, 2H, H ar); 8.05 (m, 2H, H ar); 8.42 (s, 1H, H ar). .sup.13C NMR: δ=14.1; 22.7; 24.5; 29.4; 29.5; 29.6; 29.7; 31.9; 38.8; 122.7; 125.2; 127.5; 129.4; 129.9; 130.2; 131.0; 134.8; 136.4; 200.2. HRMS: m/z (MH.sub.+) C.sub.22H.sub.30BrO calculated: 389.1475, found: 389.1471.

[0257] 2-Chloronaphthalene (1 g, 6.1 mmol) was dissolved in 6 mL of nitrobenzene under an argon atmosphere. The solution was stirred in an ice bath. Lauroylchloride (1.7 mL, 7.32 mmol, 1.2 equiv.) was added drop wise. The reaction mixture was treated with aluminium chloride (854 mg, 6.71 mmol, 1.1 equiv.) and stirred at room temperature for at least 18 hours. The reaction mixture was quenched with distilled water and extracted three times with ethylacetate. The organic phases were combined, washed with brine, dried over sodium sulphate and concentrated under reduced pressure. The product was purified by crystallization from cold ethanol to obtain a yellow powder; yield 1.06 g (50%). .sup.1H NMR (300 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=6.9 Hz, CH.sub.3); 1.26-1.41 (m, 16H, CH.sub.2); 1.79 (m, 2H, CH.sub.2); 3.07 (t, 2H, J=7.5 Hz, CH.sub.2CO); 7.49 (dd, 1H, J=8.7 Hz, J=2.1 Hz, H ar); 7.79-7.91 (m, 3H, H ar); 8.05 (dd, 1H, J=8.7 Hz, J=1.8 Hz, H ar); 8.43 (br s, 1H, H ar). .sup.13C NMR: δ=14.1; 22.7; 24.5; 29.3; 29.4; 29.5; 29.6; 31.9; 38.7; 125.1; 126.6; 127.6;

[0258] 127.7; 129.3; 130.8; 131.0; 134.3; 134.6; 136.0; 200.2. MS: m/z (MNa+)=367.2; HRMS: m/z (MH+) C.sub.22H.sub.30C.sub.10 calculated: 345.1980, measured: 345.1973.

ii. Synthesis of Ethyl 2[(6-dodecanoylnaphthalen-2-yl)methylamino]acetate

[0259] An oven-dried Schlenk flask was evacuated and backfilled with argon. The flask was charged with 1-(6-bromonaphthalen-2-yl) dodecan-1-one (1 g, 2.57 mmol), Pd(OAc)2 (46 mg, 206 mmol, 0.08 equiv.), X-Phos (196 mg, 411 mmol, 0.16 equiv.), cesium carbonate (3.35 g, 10.28 mmol, 4 equiv.), and sarcosine ethyl ester (790 mg, 5.14 mmol, 2 equiv.). The Schlenk flask was capped with a rubber septum, then evacuated and backfilled with argon three times. Anhydrous dioxane (7.5 mL) was added through the septum. The mixture was heated to 100° C. in an oil bath and was stirred for 18 hours. The mixture was cooled to room temperature, diluted with ethyl acetate, filtered through Celite, and concentrated under reduced pressure to obtain a yellow solid. The crude product was purified by flash chromatography on silica gel eluting with ethyl acetate/petroleum ether (10/90) to afford a yellow powder (569 mg, yield 52%).

[0260] .sub.1H NMR (300 MHz, CDCl.sub.3): δ=0.88 (t, 3H, J=7.2 Hz, CH.sub.3); 1.21-1.36 (m, 16H, CH.sub.2); 1.76 (m, 2H, CH.sub.2); 3.00 (t, 2H, J=7.5 Hz, CH.sub.2CO); 3.16 (s, 3H, CH.sub.3N); 4.16 (s, 2H, COCH.sub.2N); 4.18 (q, 2H, J=7.2 Hz, CH.sub.3CH.sub.2O); 6.86 (d, 1H, J=2.4 Hz, H ar); 7.06 (dd, 1H, J=9.3 Hz, J=2.7 Hz, H ar); 7.62 (d, 1H, J=8.7 Hz, H ar); 7.78 (d, 1H, J=9.0 Hz, H ar); 7.92 (dd, 1H, J=8.7 Hz, J=1.8 Hz, H ar); 8.31 (br s, 1H, H ar). .sup.13C NMR: δ=14.1; 14.2; 22.7; 24.8; 29.3; 29.5; 29.6; 29.7; 31.9; 38.4; 39.8; 54.4; 61.2; 106.0; 114.6; 115.7; 124.7; 125.6; 126.4; 129.6; 130.8; 131.1; 137.4; 148.6; 170.4; 200.2. MS: m/z (MH+)=426.3. HRMS: m/z (MH+) C.sub.27H.sub.40NO.sub.3 calculated: 426.3003, found: 426.2988.

iii. Synthesis of 6-Dodecanoyl-2-[N-methyl-N-(carboxymethyl)amino] naphthalene (C-Laurdan)

[0261] 211 mg of ethyl 2[(6-dodecanoylnaphthalen-2-yl)methylamino] acetate was dissolved in 7 mL of a freshly prepared solution of potassium hydroxide (0.25 M) in ethanol. The reaction mixture was stirred for 2 days at room temperature. The solution was acidified with HCl (1 M) until pH 1 and was extracted with ethyl acetate (3×20 mL). The ethyl acetate layers were combined, dried over magnesium sulfate and concentrated under reduced pressure to obtain a brown solid which was purified by crystallization from chloroform/petroleum ether to provide 164 mg (83% yield) of C-Laurdan. .sup.1H and .sup.13C NMR and HRMS data of the product are identical with the published data (H. M. Kim, H. J. Choo, S. Y. Jung, Y. G. Ko, W. H. Park, S. J. Jeon, C. H. Kim, T. Joo and B. R. Cho, ChemBioChem, 2007, 8, 553-559).

c. Probe Fluorescence in Organic Solvents

[0262] Fluorescence measurements were performed with a Hitachi F4500 fluorometer. Excitation and emission slits were fixed at 5 nm and spectra were recorded at 25° C. using a 1 cm path length thermostated quartz micro-cuvette. 0.26 mM solutions of the probes were prepared in a dichloromethane: methanol mixture (3:1 volume ratio). Measurements were performed with 2.6 mM probes diluted in the different solvents (see Table 1). The initial solvent of the stock solution thus represents 1% of the final volume.

d. Probe Fluorescence in Liposomes

[0263] Liposomes were prepared in PBS buffer using a classical freeze/though protocol. In brief, lipids dissolved in chloroform were dried under vacuum. The lipid film formed was rehydrated with the PBS solution and agitated until the complete suspension of the lipid film. The vesicle solution was then submitted to 6 freeze/thaw (5 min in liquid nitrogen/10 min in 37° C. water bath) cycles and then extruded 19 times sequentially trough 400 nm and 200 nm polycarbonate membranes (Avanti Polar Lipids Liposome Extruder). The 200 nm liposome suspension was stored at 4° C.

[0264] Fluorescence data was obtained using a Hitachi F4500 fluorometer. Emission spectra were recorded between 420 and 600 nm with an excitation wavelength fixed at 385 nm and 5 nm excitation and emission slits. Ethanol solutions of fluorescent probes were prepared and added to liposomes at a 1:400 (probe: phospholipid) molar ratio. Liposomes were used at a 40 μg.mL.sup.−1 final concentration. Measurements were done at 21° C. for all liposome compositions indicated in Table 2, except for DMPC where a range of temperature from 5 to 65° C. was also applied.

e. Flow Cytometry

[0265] Fluorescence measurements were performed with an LSR Fortessa. Cells passing in front of the laser were exited with a laser of 405 nm and the filters 450/50 and 525/50 were used to respectively observed the rigid state intensity and fluid state intensity. Like the spectrofluorometer, Ethanol solutions of DIOLL were prepared and added to liposomes at a 1:400 molar ratio. Liposomes were used at a 40 μg.mL.sup.−1 final concentration. Measurements were done at 21° C. for all liposome composition indicated in Table 2. Data processing was effectuated on FlowJo.

f. Cell Culture

[0266] Human mammary epithelial cells (HMECs) were provided by Lonza. HMEC were cultured in 1:1 Dulbecco's Modified Eagle's Medium (DMEM)/HAMF12 medium (Gibco) complemented with 10% FBS (Gibco), 91 U.mL.sup.−1 penicillin-streptomycin (Gibco), 10 ng.mL.sup.−1 human epidermal growth factor (EGF) (Promocell), 262 mU.mL.sup.−1 insulin (NovoRapid) and 0,5 μg.ml.sup.−1 hydrocortisone (Sigma) under a humidified atmosphere of 5% CO.sub.2 at 37° C. Cell morphology and growth curve were routinely checked.

g. Confocal Microscopy Experiments

[0267] Cells were cultured in Lab-TEK® at 30% of confluence and DIOLL was added to obtain 0.5 μM 48 hours before observation.

[0268] 48 hours later cells were at 80% of confluence and ready for observation.

[0269] Specific organelle markers (plasma membrane (Dil), endoplasmic reticulum (ER tracker), mitochondria (Mito tracker)) were added 20 minutes before observation. Nuclear stain (Hoechst) was injected automatically in the chamber after DIOLL fluorescence recording.

[0270] The images were taken using a Zeiss LSM 880 AxioObserver inverted spectral confocal microscope. This is a spectral confocal microscope on stand with incubation chamber with environmental control temperature, CO.sub.2 and hygrometry. The light source is a laser diode. The objective is an alpha Plan-Apochromat 63x/1.46 Oil DIC M27.

[0271] For colocalisation experiments, DIOLL excitation wavelength was 405 nm and the emission wavelength was a band from 424 nm to 513 nm. For spectral mode acquisition excitation wavelength was 405 nm and images were taken at 424, 432, 441, 450, 459, 468, 477, 486, 495, 504, 513 nm emission wavelength.

[0272] For ER-tracker and Mito-tracker, the excitation wavelength was 561 nm and detection wavelength was 611 nm.

[0273] For DiI, the excitation wavelength was 561 nm and emission wavelength was 621 nm.

[0274] For Hoechst, the excitation wavelength was 405 nm and emission wavelength was 430 nm

h. Determination of the GP Parameter From the Images Obtained by Confocal Microscopy

[0275] A composite image was created from the image obtained with the DIOLL and the image obtained with one of the specific subcellular markers (Dil, ER-tracker, Mito tracker) to precisely identify areas were the probed colocalized. The emission intensity values in the 440 and 490 nm channels were measured using ImageJ after background correction for the areas identified. GP values were calculated with the same formula as for spectroscopy

[00005]$\mathrm{GP}=\frac{I_{440}-I_{490}}{I_{440}+I_{490}}$

experiments:

[0276] where I.sub.440 and I.sub.490 are recorded fluorescence intensities at wavelengths of 440 nm and 490 nm, respectively.

[0277] GP values were calculated from at least 5 independent measurements.

2. Results

[0278] a. Spectroscopic Properties of DIOLL Probe: A Comparison with Laurdan and C-Laurdan Probes in Organic Solvents of Increasing Polarity

[0279] The spectroscopic properties of DIOLL probe were compared to the one of Laurdan and C-Laurdan in organic solution (FIG. 1). As shown in FIG. 1, Laurdan sensitivity to the polarity of its environment translates itself in a shift of the fluorescence emission maximum from lower wavelength for non-polar solvents (423 nm for toluene, 440 nm for chloroform) to higher wavelength for polar solvents (460 nm for DMSO, 485 nm for ethanol) (Table 1). This effect, named bathochromic effect, is at the basis of the probe's use in the sensing of the order state of the membranes.

[0280] Therefore, the same effect is searched for when alternative fluorophores are proposed: C-Laurdan and DIOLL probe perfectly respond to this requirement as shifts in the maximum emission wavelength are recorded with the increasing solvent polarity (416 nm and 425 nm in toluene, 432 and 441 nm in chloroform, 450 and 460 nm in DMSO and 485 nm and 486 nm in ethanol, for C-Laurdan and DIOLL probe, respectively).

TABLE-US-00001 TABLE 1 Maximum emission wavelength of Laurdan, C-Laurdan and DIOLL probe in organic solvents of increasing polarity. Solvent Probe Toluene Chloroform DMSO Ethanol λ emission Laurdan 423 440 460 485 maximum (nm) C-Laurdan 416 432 450 485 DIOLL 424 424 441 486 Maximum Laurdan 3814 4441 5533 4221 fluorescence C-Laurdan 359 3115 2428 1491 intensity (A. U.) DIOLL 5041 5041 9384 7647

[0281] Fluorescence intensity is however very different for the two Laurdan analogues, with DIOLL remarkably showing a high fluorescence intensity, going from 1.5 to 2 fold higher than for Laurdan whatever the solvent used. Inversely, C-Laurdan shows very low fluorescence intensities: 1.3 to 10 times lower than that of Laurdan (hypochromic effect). The good fluorescence yield recorded for DIOLL probe makes it a good candidate for microscopy and spectroscopy applications.

b. Fluorescence of DIOLL Probe in DMPC Liposomes

[0282] To check whether these promising fluorescence properties may serve for membrane state characterization, the inventors compared fluorescence properties of DIOLL probe with those of Laurdan and C-Laurdan in liposomes made of 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC). This lipid is known to present a gel to liquid disordered phase transition at 24° C. that makes it a good model system to evaluate fluorophore capacity to evidence membrane fluidity changes, as PC phosphocholine is the main phospholipid in cell membranes.

[0283] The fluorescence spectrum of Laurdan inserted in DMPC liposomes evolved when temperature increased from 5° C. to 65° C. (FIG. 2A). At temperatures lower than the phase transition temperature the fluorescence spectrum shows a maximum at 440 nm as expected for a gel phase. Above 24° C., the intensity at 440 decreases in favour of an increase in intensity around 490 nm. The decrease in 440 nm emission intensity becomes more pronounced with temperature, i.e. with the amount of lipid molecules in a disordered state. The 490 nm contribution becomes predominant at temperatures above 28° C. This behavior is characteristic for Laurdan and has been described in numerous reports (Parasassi et al., Biophysical Journal Volume 66 January 1994 120-132; Parasassi and Gratton, Journal of Fluorescence, Vol. 5, No. 1, 1995).

[0284] C-Laurdan in DMPC also shifts towards higher wavelength, but an overall decrease in the fluorescence intensity is recorded with the increase in temperature (FIG. 2B).

[0285] DIOLL has a behavior similar to Laurdan, with a shift in the maximum emission wavelength from 440 nm for the gel phase to 490 nm in the liquid crystalline phase (FIG. 2C). Moreover, the decrease in the fluorescence intensity between the gel and the fluid phase is more limited than for Laurdan with a gel over liquid crystalline ratio of 0.8 vs 0.6 for Laurdan.

[0286] Most of the commonly used fluorophores show better fluorescence intensity in the rigid (apolar) conditions, making thus the disordered contribution of the membrane underestimated. The good fluorescence yield of DIOLL probe in the disordered state, comparable with that obtained in the ordered state, opens the possibility of a more accurate quantification of the fluidity state of the membrane with respect to fluorophores commercially available to date.

[0287] The spectral shift described above was quantified by calculation of the Generalized Polarization (GP) parameter. This parameter provides a quantitative index of membrane fluidity and is calculated as the difference between fluorescence emission intensity at a 440 nm (lipids with gel properties) and the intensity at 490 nm (wavelength evidenced in lipids with liquid crystal properties), divided by the sum of these two intensities. The GP parameter was thus calculated from fluorescence emission intensities according to the following formula:

[00006]$\mathrm{GP}=\frac{I_{\mathrm{gel}}-I_{\mathrm{lc}}}{I_{\mathrm{gel}}+I_{\mathrm{lc}}}$

[0288] where I.sub.gel and I.sub.lc are, respectively, intensities of fluorescence at wavelengths characteristic of probe emission, respectively in gel and liquid crystal lipid phases.

[0289] Graphs of GP evolution as a function of temperature are plotted in FIG. 2D. One can notice that Laurdan and DIOLL probes have a similar evolution. At a temperature lower than 20° GP values are rather high, which is expected for lipids in a gel phase. GP values decrease with the increase in temperature, with a rather sharp drop at 21° C., where the liquid crystal phase starts to appear. The decrease in GP is rather sharp as phase transition is a cooperative phenomenon and once lipid chains began to adopt disordered conformation the process is amplified (less energy is required to induce melting of the remaining chains). First derivatives of the GP function vs temperature show an inflexion point at 23° C. for Laurdan and DIOLL , which is in line with previously reported values of melting temperature (Tm). At this point, 50% of lipid chains are in liquid crystalline state. For both DIOLL and Laurdan, above the transition temperature, the GP values continue to decrease, also less sharply than around the phase transition temperature. The two fluorophores are sensitive to the mobility of water molecules in the vicinity of the excited naphthalene moiety and, although lipids are in a liquid crystal phase, the solvent mobility at the polar/non-polar interface continues to increase. This is of interest as, theoretically, GP values would permit to discriminate membrane fluidity far beyond the phase transition temperature. C-Laurdan also shows a decrease in GP values with temperature, but the amplitude of this decrease was lower than for Laurdan and DIOLL and GP did not evolve above the phase transition temperature.

[0290] In conclusion, DIOLL and Laurdan show a large difference in GP values between the ordered/gel and disordered/liquid crystalline states. This property is of importance for microscopy use where a clear-cut separation of fluorescence coming from gel and liquid crystalline phases is required for visualization of membranes in different states.

[0291] The fluorescence intensities of DIOLL were compared with those of Laurdan and C-Laurdan, at 25° C. and a final concentration of 0.2 μM, in liposomes constituted of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), which occurs as one of the main phospholipids in cell membranes (FIG. 2E). The fluorescence intensity of DIOLL obtained in identical conditions was 7 fold higher than that of Laurdan and C-Laurdan.

c. Degree of Order Quantification in Liposomes by Fluorescence Spectroscopy and Flow Cytometry Using DIOLL

[0292] As membrane fluidity state is determined by lipid acyl chain length and unsaturation, the inventors checked whether DIOLL was equally able to render variations in membrane state depending on lipid characteristics. Therefore, DIOLL was inserted in liposomes constituted of phosphatidylcholines with different melting temperatures, dictated by different chain length and unsaturation (ranging from 14 to 18 carbons for the chain length and from 0 to 2 for the chain unsaturation, respectively; Table 2).

[0293] Fluorescence spectra were registered and GP parameter was measured at 21° C.

TABLE-US-00002 TABLE 2 Liposome preparations used and corresponding phase transition temperatures (Tm). Liposome Acyl chain preparation Lipid composition Schematic structure Tm DLPC 1,2-dilinoleoyl- sn-glycero-3- phosphocholine 18:2 (Δ.sup.9,12- Cis) PC [00015] −57° C. DOPC 1,2-dioleoyl-sn- glycero-3- phosphocholine 18:1 (Δ.sup.9- Cis) PC [00016] −17° C. POPC 1-palmitoyl-2- oleoyl-glycero-3- phosphocholine 16:0-18:1 PC [00017]  −4° C. DMPC 1,2-dimyristoyl- sn-glycero-3- phosphocholine 14:0 PC [00018]  24° C. DPPC 1,2-dipalmitoyl- glycero-3- phosphocholine 16:0 PC [00019]  41° C. DSPC 1,2-distearoyl- sn-glycero-3- phosphocholine 18:0 PC [00020]  55° C.

[0294] Results plotted in FIG. 3A show clear differences in the GP values according to the nature of the liposome going from the most rigid, DSPC (C18:0) with a GP value of 0.58 to the most fluid DLPC (C18:2) with a GP value of −0.26. GP values obtained are in correlation with Tm ranging from 55° for DSPC to −57° C. for DLPC (Table 2).

[0295] Moreover, the new fluorophore proved to be an excellent tool to measure liposome fluidity by flow cytometry. This technique allows one to visualize and discriminate events (cells or particles) within a heterogeneous population via differences of size, granularity or fluorescence. Unlike the spectrofluorometer, which gives a bulk fluorescence spectrum and thus a bulk value of GP, by using flow cytometry, one may obtain the GP parameter of individual particles in a heterogeneous population.

[0296] To take advantage of a routine flow cytometer, the inventors replaced the GP parameter by the ratio 525 nm fluorescence emission intensity (corresponding to the fluid phase emission) on 450 nm fluorescence emission intensity (corresponding to the gel phase emission) which is equivalent to a ratio fluid state intensity on rigid state intensity. Results are plotted in FIG. 3B.

[0297] Indeed, the correlation GP and fluidity state observed through spectroscopy is also found with a flow cytometer. The entire fluidity range is discriminated with the DSPC liposomes, theoretically the most rigid at 21° C., which has a lowest ratio 525/450, while the DLPC liposomes which the most fluid has a highest ratio 525/450.

d. Labelling of Membranes with DIOLL Probe as a Function of Their Fluidity State

[0298] To check whether the probe empirically distinguishes differences in membrane fluidity in cell membranes, HMEC cells were incubated with DIOLL probe for 48 h and then observed by confocal microscopy. Fluorescence images were recorded each 10 nm from 425 to 512 nm (excitation 405 nm). DIOLL labeled uniformly all cell membranes, yet the fluorescence profile was different for each emission channel, with membrane structures that were more intensively labeled in the blue edge and others in the red edge of the spectrum. An overlay of images obtained is shown in FIG. 4, where one clearly distinguishes different shades of grey corresponding to the sum of different fluorescence intensities obtained in each channel.

[0299] Different shades of grey corresponding to different fluorescence intensity obtained in each channel leads to the differentiation of the fluidity state of the cell membranes. To precisely identify plasma and organelle membranes, co-localization experiments with specific organelle markers were performed. DIOLL fluorescence co-localized at the plasma membrane level with membrane label Dil. DIOLL also showed a good co-localization with ER-tracker, which is a specific label of the endoplasmic reticulum. It also co-localized with Mito-tracker, a specific label of mitochondria. Of note, DIOLL perfectly labels membrane structures of the nucleus. Nuclear content was labeled with Hoechst.

[0300] From DIOLL membrane labelling, GP values were estimated from the relative intensities at 440 and 490 nm for each type of membrane identified above (FIG. 5). High GP values (0.05±0.03) were obtained for plasma membrane structures (A). This is in line with literature data concerning plasma membrane composition, with a high percentage of sphingomyelin and cholesterol. Indeed, sphingolipids are lipids with high melting temperature and this type of lipids are expected to give stiffness to bilayers. On the other hand, membranes of endoplasmic reticulum (B), mitochondria (C) and nucleus (D) are rather fluid with GP values of −0.14±0.06, −0.18±0.02 and −019±0.04, respectively. This is also in line with the membrane composition of these organelles. Endoplasmic reticulum displays extremely low concentrations of sterols and complex sphingolipids, despite being the main site of synthesis. The loose packing of endoplasmic reticulum membrane lipids is consistent with the function of the organelle in the insertion and transport of newly synthesized lipids and proteins. In mitochondria, cholesterol and sphingolipid content is also low. The major lipid constituents are phosphatidylcholine and phosphatidylethanolamine, which account together for about 80% of total phospholipids. Cardiolipin is present in the range of 10-20% of total mitochondrial phospholipids. In mammalian cells, approximately 50% of mitochondrial fatty acids are unsaturated, linoleic acid (18:1), linolenic acid (18:2) and arachidonic acid (20:4) being the most prominent fatty acids, which gives mitochondrial membranes a very high fluidity also reported with DIOLL probe fluorescence.

Example 3

Quantification of Lipid Membranes in Aqueous Solutions

[0301] Liposomes, various supported or tethered bilayers, monolayers, bicelles, to name only some, are largely used systems to mimic plasma and organelle membranes.

[0302] Among such models, liposomes are widely used to probe protein-lipid interactions, lipid organization and molecule penetration across membranes. They are also versatile assemblies in drug delivery systems, as they are non-toxic, biodegradable, and readily produced on a large scale. They are also extensively used for the reconstitution of transmembrane proteins, to form proteoliposomes, the lipid bilayer offering a native-like medium for membrane-spanning domains. For all these applications, liposomes undergo various fabrication processes: extrusion, centrifugation, protein reconstitution, cell-free protein synthesis, molecule grafting, etc. It is therefore of interest to determine lipid concentrations after various processes and in different production batches.

[0303] To estimate liposome concentration, several methods are popular. Among them, colorimetric methods such as Bartlett. and Rouser. methods are based on the reaction between the inorganic phosphate obtained after liposome treatment with perchloric acid and ammonium molibdate to obtain phosphomolibdate. The addition of a reducing agent such as 1-amino-2-naphtol-4 sulfonic acid (Bartlett method) and ascorbic acid (Rouser method) leads to formation of a violet complex absorbing at 820-830 nm. Both methods are rather sensitive (LOD and LOQ around 5 and 20 μg.ml.sup.−1, respectively). Alternatively, the Stewart method (Stewart, Analytical Biochemistry 104, 10-14 (1980)) probably the most widely used method for phospholipid determination, is based on the formation of a complex between the phosphate group of the lipid polar head and ammonium ferro-thiocyanate, absorbing at 485 nm. It is rather fast and easy to implement, but requires lipid extraction in organic solvent and the use of chloroform-resistant recipes and protection. LOD and LOQ methods are evaluated at 10 μg.ml.sup.−1 and 100 μg.ml.sup.−1, respectively.

[0304] Chromatographic methods of lipid separation, together with mass spectroscopy or NMR are also methods of choice for lipid detection and quantification, as one can obtain not only the lipid amount, but also lipid class distribution. Yet, they are time consuming, rather expensive and not readily used in routine experiments.

[0305] Given the number of studies dwelling on liposomes, proteoliposomes and on cell-derived vesicles, the inventors propose a simple method to estimate the amount of membranes in aqueous solutions.

1. Material and Method

[0306] a. Materials

[0307] DIOLL probe was synthesis as described in Example 1.

[0308] Lipids were purchased from Avanti Polar Lipids (Alabaster, Al.). Hepes, PBS, NaCl, ferric chloride hexahydrate (FeCl.sub.3(H.sub.2O).sub.6) and ammonium thiocyanate (NH.sub.4SCN) were purchased from Sigma Aldrich (St Quentin Falavier, France). Ultrapure water was obtained with a PureLabOption Q7 system (Véolia Water STI, Decines, France).

b. Methods
i. Liposome Preparation

[0309] Liposomes were prepared using a classical freeze/thaw protocol. Briefly, the appropriate lipid mixture solubilized in chloroform was dried under vacuum, then solubilized with 1 ml buffer under strong agitation. Assays were done either in HEPES 20 mM pH 7.4 buffer 100 mM NaCl or in PBS. The liposome emulsion obtained was submitted to 6 freeze/thaw cycles (5 min in liquid nitrogen/10 min in a water bath at 37° C., 1 minute strong agitation) to decrease the number of bilayers and obtain unilamellar vesicles. To obtain homogenous vesicles, the liposome solution was extruded successively through a 400 nm and 100 nm polycarbonate membrane using an extruder (Avanti Polar Lipids). Liposomes were routinely prepared at lipid concentrations between 1 and 10 mg.ml.sup.−1 and subsequently diluted at the desired concentration.

ii. DIOLL Probe Fluorescence-Based Liposome Quantification

[0310] A POPC solution of known concentration (100 μg.ml.sup.−1) was used as standard. Increasing volumes of the standard solution were diluted with buffer in a 96 well plate (Nunc). The final volume per well was 200 μl and lipid concentrations between 0 and 80 μg.ml.sup.−1 were used. Samples were prepared in parallel to a final volume of 200 μl per well.

[0311] A constant volume of DIOLL solution in ethanol (5 μl) was added in each well to obtain a final DIOLL concentration of 2 μM. The volume of ethanol added represented 2.5% of the final volume and had no effect on liposome integrity. The plate was incubated at room temperature for 10 min. Fluorescence emission spectra were recorded between 420 nm and 600 nm for an excitation wavelength of 390 nm with a TECAN InfiniteM200 plate reader. Spectra were recorded at 21° C.

[0312] The surface under the spectrum was calculated for each condition using an Excel complementary function, IntegralP (www.fordom.free.fr) which allows one to calculate the integral of a function when discrete data sets are available. Maximum fluorescence intensities or surfaces under the curve were plotted as a function of the lipid concentration to obtain the standard curve. They both show a hyperbolic variation and the equation of the standard curve was obtained using a hyperbolic fit y=ax/(x+b) (SigmaPlot). The calculation of the a and b coefficients allowed us to determine unknown concentrations in the samples. As various volumes were used for the assays, the concentration of the stock liposome solution was presented taking into account various dilution factors. Measurements were performed in triplicates.

[0313] For the estimation of the limit of detection (LOD) and of quantification (LOQ), the fluorescence spectra of 2 μM DIOLL in aqueous buffer were recorded in the absence of liposomes and the surface under the curve was calculated to estimate the blank value on 5 independent measurements. LOD and LOQ were calculated as 3 times or 10 times the standard deviation of the blank, respectively.

[0314] Sensitivity was estimated from the slope of the linear fit for low liposome concentrations. As the fluorescence intensity is dependent on the instrument used, in this study the inventors estimated that an increase of 100 A.U. in the fluorescence intensity obtained with TECAN InfiniteM200 was significant.

[0315] To conclude upon the repeatability of the method, the coefficient of variation (CV) was calculated as the percentage of the standard deviation on the mean of 5 independent measurements. Accuracy was deduced from the difference between the concentrations obtained with the DIOLL-based method and a standard method of phospholipid quantification, in our case the Stewart method.

[0316] For the Stewart method, lipids were extracted from the various samples with chloroform. In brief, aqueous samples were mixed with chloroform (10 times the volume of the sample) under vigorous shaking. The samples were then centrifuged at 300×g for 10 min to accelerate phase separation (Multifuge 3 LR, Heraeus). The chloroform (bottom) phase was recovered with a glass pipette.

[0317] A POPC standard solution was prepared at 100 μg.ml.sup.−1 in chloroform. Increasing volumes of the standard solution were diluted with chloroform in glass centrifuge tubes to a final volume of 2 ml. Samples extracted in chloroform were prepared in parallel and diluted to 2 ml. 2 ml of Ferrothiocyanate solution (obtained by mixing 20.7 g of Ferric chloride hexahydate and 30.4 g of ammonium thiocyanate in 1L of ultrapure water) was added to each tube. The solutions were vigorously mixed and then centrifuged at 300×g for 10 min (Multifuge 3 LR, Heraeus). The chloroform (bottom) phase was recovered with a glass pipette and the absorbance was measured at 485 nm in glass cuvettes with a Jasco V-730 spectrophotometer. For quantification absorbance at 485 nm was plotted against phospholipid concentration.

iii. Purification of exosomes

[0318] The purification of exosomes was realized by a series of differential centrifugations. In brief, 120 ml of HMEC culture medium were collected after two days of culture, and centrifuged at 2 000×g at 4° C. during 10 min (Multifuge 3LR Heraus™). Supernatants were collected and ultracentrifuged at 20 000×g, 4° C. during 30 min (Optima LE-80 Beckman™ with a fixed rotor angle 50.2 Ti Beckman™). A second ultracentrifugation was realized at 100 000×g, 4° C. during 2 h. Pellets were resuspended with 1 mL of PBS buffer, regrouped in the same tube and ultracentrifuged at 100 000×g, 4° C. during 1 h. Finally, the pellet was resuspended in 200 μL of PBS buffer to obtain concentrated exosomes.

iv. Giant Plasma Membrane Vesicles Preparation

[0319] Giant Plasma Membrane Vesicles (GPMV) were prepared according to previously described protocol (Baumgart, T. et al. Large-scale fluid/fluid phase separation of proteins and lipids in giant plasma membrane vesicles. Proc. Natl. Acad. Sci. (2007). doi:10.1073/pnas.0611357104) but adapted to PC-3 cell culture. In brief, PC-3 were cultured in RPMI-1640 10% FBS 100 U/mL penicillin and 100 μg/mL streptomycin under a humidified atmosphere of 5% CO.sub.2 at 37° C. at 80% cell confluence in 25 cm.sup.2 plastic flasks. Growth medium was removed and cells were incubated for 1 h with 25 mM formaldehyde 2 mM DTT in PBS buffer at room temperature under shaking. The supernatant containing GPMVs was recovered and the presence of GPMVs was checked with an optical microscope. For DIOLL-based measurements, 200 μl of the GPMV preparation was incubated with 5 μl DIOLL probe ethanol solution (2 μM final concentration) and fluorescence spectra were recorded as described above. The lipid concentration was determined against a standard curved obtained with POPC liposomes of known concentration.

2. Results

[0320] a. Quantification of Liposomes Using DIOLL Probe

[0321] To probe the dependency of DIOLL probe fluorescence upon liposome amount, increasing concentrations of POPC liposomes were incubated with a constant concentration of 2 μM DIOLL. Fluorescence spectra were recorded and plotted in FIG. 6A. Fluorescence intensity is rather low in the absence of liposomes and increases progressively with lipid amount, but seems to reach saturation at concentrations over 50 μg.ml.sup.−1. The fluorescence intensity emission maximum was obtained at 490 nm, which corresponds to lipid membranes in liquid disordered state, as expected for POPC liposomes at 21° C. (Tm=−4° C.). Fluorescence intensity at 490 nm as a function of the lipid concentration in the well (FIG. 6B) showed a hyperbolic evolution. Yet, for polarity-sensitive probes, such as DIOLL probe, the position of the maximum fluorescence emission largely depends on the order state of the lipid membrane. This characteristic, which makes them valuable for determining the order state of the membrane, is an inconvenient when sample fluidity does not match that of the standard solution. Therefore, to overcome this inconvenient, the fluorescence intensity parameter was replaced by the surface under the peak, which does not depend on the fluidity state of the membrane. The surface under the peak was estimated by integrating the spectra obtained between 420 and 560 nm using an Excel complementary function, IntegralP (www.fordom.free.fr). When plotted as a function of the liposome concentration, the surface under the peak also nicely fits a hyperbolic function (FIG. 6C). Therefore, the hyperbolic dependency between the surface under the experimental fluorescence emission spectra and the liposome concentration in the well can be used as a calibration curve to measure liposome concentration in various samples.

[0322] The limit of detection (LOD), the lowest concentration that can be detected but not quantified, and the limit of quantification (LOQ), the lowest concentration that can be quantified with a reasonable degree of confidence, of the calibration curve were determined from the standard deviation of the blank solution (2 μM DIOLL in buffer) (Table 3). The LOD, which corresponds to the lipid concentration obtained at 3 times the standard deviation of the blank, which was of 0.4 μg.ml.sup.−1, whereas the LOQ corresponded to a lipid concentration of 1.2 μg.ml.sup.−1 obtained for 10 times the standard deviation of the blank. These limits are rather low and make this method interesting for measuring low liposome concentrations, when small or diluted sample volumes are available.

TABLE-US-00003 TABLE 3 Validation of the detection method LOD LOQ Sensitivity Repeatability Accuracy 0.4 μg.ml.sup.−1 1.2 μg.ml.sup.−1 10 000 (A.U) 0.4 μg.ml.sup.−1 1.2% 6.0%

[0323] The sensitivity of the detection method is defined as the dependency between the instrumental response and the variation in the concentration of the compound of interest. It is instrument-dependent. In our conditions, the slope of the linear part was equal to 10.sup.5 A.U (arbitrary unit). per concentration unit (μg.ml.sup.−1). That means that, considering that the sensitivity limit of TECAN InfiniteM200 used in this study is 100 A.U., one can detect variations in liposome concentration of the sample as low as 1 ng.ml.sup.−1.

[0324] The repeatability of the method, which represents the determination of how close values are to each other, under identical experimental conditions, was assessed by determination of the concentration of a POPC liposome solution prepared independently of the standard solution. A concentration of 914±56 μg ml.sup.−1 was obtained with 3 replicates, which gave a repeatability of 6%. To assess the accuracy of the new quantification method, the same POPC liposome solution was measured using the largely used Stewart method. A difference of 1.2% was obtained between mean values obtained with the two methods. Results were therefore consistent and opened the possibility to routinely use the DIOLL as a probe to measure lipid membrane amount in various samples, without lipid extraction and in a rather short time (less than 30 min).

[0325] Robustness of the DIOLL method was checked with respect to various parameters. One major point of interest was whether the membrane fluidity, which largely influences DIOLL fluorescence, had an impact on the DIOLL-based quantification method. To this purpose, liposomes constituted of DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine) and DPPC (1,2-dipalmitoyl-glycero-3-phosphocholine) were prepared (Table 4).

TABLE-US-00004 TABLE 4 Liposome preparations used and corresponding phase transition temperatures (Tm) Liposome Acyl chain preparation Lipid composition Schematic structure Tm POPC 1-palmitoyl-2- oleoyl-glycero-3- phosphocholine 16:0-18:1 PC [00021] −4° C. DMPC 1,2-dimyristoyl- sn-glycero-3- phosphocholine 14:0 PC [00022] 24° C. DPPC 1,2-dipalmitoyl- glycero-3- phosphocholine 16:0 PC [00023] 41° C.

[0326] In our experimental conditions (21° C.), DMPC (gel to liquid crystalline transition temperature (Tm) of 23° C.) is expected to be in an ordered state, but close to the liquid ordered-liquid disordered transition, and DPPC (Tm 41° C.) in an ordered configuration. Independently, a standard liposome solution was prepared for each type of liposomes to obtain the corresponding calibration curves. DMPC-inserted DIOLL fluorescence spectra (FIG. 7A) showed a maximum fluorescence emission at 440 nm consistent with lipid membranes in an ordered state, with a contribution at 490 nm corresponding to the fluid phase. Although it is difficult to calculate the proportion of each state, the fluorescence emission spectra indicate that, in our experimental conditions, DMPC liposomes make a good model of membranes in which two phases coexist. When inserted in DPPC liposomes, DIOLL maximum fluorescence emission was obtained at 440 nm, which is consistent with a membrane in gel/ordered state (FIG. 7B). For both DMPC and DPPC liposomes, the surface under the peak estimated as described above gave hyperbolic evolutions (FIGS. 7C and D). The equations obtained by fitting the experimental data to a hyperbola allowed us to estimate the concentration of liposomes in various samples.

[0327] To check the influence of the membrane fluidity on the quantification method, POPC, DMPC and DPPC liposomes were prepared and their concentration was determined using either POPC, DMPC or DPPC calibration curves or by the Stewart method with POPC in chloroform solution as standard. Results are presented in FIG. 7E. The concentration of the POPC solution measured using either POPC, DMPC or DPPC calibration curves was consistent with that obtained by the Stewart method. Similar results were obtained for DMPC and DPPC liposomes quantified using DIOLL-based calibration curves or by Stewart method. Therefore, the surface under the peak parameter as a function of the liposome concentration can successfully replace the fluorescence intensity to measure sample concentration, whatever the membrane fluidity state. This is particularly interesting as membrane fluidity in samples of interest is usually unknown.

[0328] Cholesterol is one major component of membranes and also a major modulator of membrane order. Depending on membranes, the proportion of cholesterol varies between 10 and 30% of lipid total mass. Liposomes containing a mixture of POPC and cholesterol where prepared and the concentration was measured using DIOLL method and POPC liposomes of known concentration as a standard curve. The concentration obtained for 10, 20 and 30% cholesterol correlated well with the concentration of phospholipids obtained using the Stewart method (FIG. 8, white and black bars), but not with the total amount of lipids (if one takes into account 10, 20 or 30 mass % of cholesterol in the sample) (FIG. 8, dotted bars).

[0329] Membranes contain non negligible amounts of negatively charged lipids with precise and fundamental functions. It is thus of interest to check the robustness of the DIOLL-based quantification method towards negatively charged lipids. As an example, the inventors have chosen to prepare liposomes containing 50% POPC and 50% POPS (1-palmitoyl-2-oleoyl-glycero-3-phosphoserine). The concentration of the prepared liposome solution was estimated either by the DIOLL-based method, with POPC liposomes of known concentration as standard curve, either by the Stewart method with POPC chloroform solution as standard (FIG. 9). The two methods very nicely correlated.

[0330] To sum up, by using the surface under the curve parameter instead of the fluorescence emission intensity, the new fluorescence-based quantification method of membrane proved to be robust with respect to the order state of the membrane. By using a simple standard curve such as plain liposomes of increasing concentration and a constant fluorophore concentration, one can determine membrane amount in samples of different compositions: DMPC, DPPC, cholesterol-containing liposomes or negatively charged liposomes.

b. Quantification of Exosomes Using DIOLL Probe

[0331] Exosomes are small vesicles (50 to 150 nm) secreted by cells, specific of their origin cell and which can be extracted from numerous body fluids (blood, urine . . . ). They are therefore of interest for studies of pathologies, but also for the development of non-invasive diagnostic tools, notably in oncology. Measuring exosome concentration rapidly, with small sample volumes and in a non-destructive manner is of growing interest. Therefore, the inventors adapted the DIOLL-based quantification method to this type of vesicles. A standard curve based on POPC liposomes of known concentration in the presence of a constant DIOLLconcentration (2 μM) was recorded as the surface under the curve vs. liposome concentration in the well, as described above. Exosome samples were incubated with the same DIOLL concentration as the standard solution and the surface under the curve was calculated. Vesicle concentration was obtained using a hyperbolic fit of the standard curve and correlated with phospholipid determination using Stewart method (98±10 μg.ml.sup.−1 vs 102±10 μg.ml.sup.−1) (FIG. 10).

c. Quantification of GPMVs Vesicles Using DIOLL Probe

[0332] A second type of vesicles used was obtained from living cells by plasma membrane budding. GPMVs are derived from biological cells, by chemically induced (here with 25 mM formaldehyde, 2 mM DTT in PBS) plasma membrane vesiculation or “blebbing”.

[0333] This type of vesicles is largely used to analyze phase coexistence in systems with realistic biological membrane compositions or peptide penetration across membranes. Given the raising interest in the study of plasma membrane heterogeneities, the inventors determined the amount of membranes in the GPMVs samples. As expected from literature data, GPMVs obtained by this method show simple, low-curvature geometries of giant unilamellar vesicles and appear free of cytoskeletal constraints. Using DIOLL fluorescence a lipid concentration of 84±1 μg.ml.sup.−1 was obtained from an estimated PC-3 cell population of 5 million. Using the Stewart method, a good correlation was observed with an estimated phospholipid sample concentration of 87±1 μg.ml.sup.−1 (FIG. 11).

3. Discussion

[0334] Quantification methods of membrane content in various samples are scarce and require either lengthy protocols involving lipid extraction and costly instruments (GC, MS, NMR), organic solvents or energetic conditions.

[0335] The inventors present a new, rapid and easy to use method of lipid membrane quantification in artificial and natural vesicles based on the fluorescence of DIOLL probe. This fluorophore has a very low fluorescence in aqueous buffers, but when it spontaneously inserts into lipid bilayers its fluorescence drastically increases. the inventors checked therefore whether this was a regular increase, by incubating a constant amount of fluorophore with increasing amounts of liposomes. As shown in FIG. 6, there was a hyperbolic dependency not only between the fluorescence intensity and the lipid amount in the sample, but also between the surface under the fluorescence emission spectra and the lipid amount. This parameter, unlike the fluorescence intensity, does not depend on the membrane fluidity and may serve as a universal parameter to quantify membranes.

[0336] To check this hypothesis, 3 types of liposomes were prepared with lipids with different ordering degrees at room temperature (23° C.): POPC (fluid state), DMPC (phase transition) and DPPC (gel state). The amount of liposomes was quantified using as standards either liposome solutions of the same nature, either liposome solutions of different fluidity. Quantification results (FIG. 7) show that similar results were obtained whatever the composition of the standard solution. Moreover, the lipid concentrations in the samples correlated well with those obtained by Stewart method after lipid extraction in chloroform. Thus, a plain solution of POPC liposomes of known concentration may serve as a standard curve for samples of diverse lipid composition and diverse membrane states.

[0337] Detection and quantification limits of the POPC standard curve attained values as low as 0.4 and 1.2 μg.ml.sup.−1, comparable with those of Bartlett and Stewart methods, which allows one to quantify low amounts of biological samples. Although a fluorophore is inserted into the bilayer, this method is not destructive and liposomes, proteoliposomes or diverse vesicles may be used for further applications. The sensitivity of the method was good up to 40 μg.ml.sup.−1. Above this value, the intensity of fluorescence reaches a plateau and samples require dilutions. The sensitivity of the method depends on the instrumental response to the variation in the analyte concentration, therefore it is instrument-dependent. In our conditions, by using an instrument such as TECAN InfiniteM200, the inventors detect variations in the liposome concentrations the range of 1 ng.ml.sup.−1.

[0338] Robustness of the detection method was checked with respect to the presence of cholesterol (FIG. 8) and negatively charged lipids (FIG. 9) and a good correlation was obtained with phospholipid quantification by Stewart method. Therefore, this method was used for quantification of more complex vesicles, and natural vesicles, such as exosomes and GPMVs. If the inventors consider the insertion mode of its analogue, Laurdan, DIOLL is expected to fit between lipids at the level of the glycerol backbone. Therefore, the fluorescence intensity is dependent on the amount of phospholipid membranes. Cholesterol and other components, like proteins, or glycolipids, in natural vesicles, largely influence membrane fluidity and, consequently, the shape of the spectrum, but did not influence fluorescence yield in terms of surface under the curve. Therefore, systematic correlation was obtained between membrane content determined by DIOLL and phospholipid concentration determined by Stewart method.

[0339] To conclude, DIOLL fluorescence can be used as a parameter to measure the amount of phospholipid membranes in complex samples without solvent extraction and that by using simple liposome compositions as standards. The detection method, based on fluorescence, is fast (less than 30 min with a plate reader, less than 1h with a classical fluorimeter), simple to use, sensitive and realizable with standard lab equipment.

Example 4

Fluidity Characterisation of Nanoemulsions Preparation with DIOLL Probe

1. Materials and Methods

[0340] a. Material

[0341] DIOLL probe was synthesis as described in Example 1.

[0342] Medium chain triglycerides, MCT (Miglyol®812) purchased from CREMER OLEO GmbH & Co. KG (Hamburg, Germany) was used as the oil phase. Polyoxyethylene (40) stearate (Myrj®52) from Sigma-Aldrich (St Quentin-Fallavier, France) and oleoyl polyoxyl-6 glycerides (Labrafil®M1944CS) from Gattefossé (Saint-Priest, France) were used as non-ionic surfactants. The aqueous phase used to prepare emulsions was sodium phosphate buffer solution (5 mM; pH 7.4). Tacrolimus (FK-506) was purchased from LC Laboratories (Woburn, Mass., USA). Hydrochloric acid 37%, AnalaR NORMAPUR® Reag. Ph. Eur. was from VWR International (Fontenay-sous-Bois, France). Dichloromethane, methanol, acetonitrile (HPLC grade), dodecyl sulfate sodium salt pure, sodium taurocholate hydrate 96% and sodium hydroxide were purchased from Fisher Scientific (Illkirch, France). Egg phospholipids with 70% phosphatidylcholine (Lipoid E80S) were from Lipoid GmbH (Ludwigshafen am Rhein, Germany). Acetic acid was obtained from Chem-Lab NV (Zedelgem, Belgium). Maltodextrin MD (Glucidex® 12D) from Roquette Freres (Lestrem, France) and trehalose 100 (TR) as a gift of Hayashibara Co. Ltd (Okayama, Japan) were used as cryoprotectants in the drying studies. Milli-Q water, used as the aqueous phase to prepare all solutions and emulsions, was obtained using a Milli-Q Academic System (Millipore, Saint Quentin, Yvelines France).

b. Methods
i. Nanoemulsion Preparation

[0343] Nanoemulsions (NE) were prepared by emulsion phase inversion (EPI) technique. The oil phase was prepared by mixing the oil (MCT) and surfactants (Polyoxyethylene (40) stearate: S1; and oleoyl polyoxyl-6 glycerides: S2) under magnetic stirring (750 rpm) at 80° C. Then, the aqueous phase (PBS 5 mM), heated up at 80° C. as well, was added all at once into the organic melt phase. Stirring was performed using a rotor-stator disperser (T25 digital Ultra-Turrax® equipped with a S25N-8G shaft (IKA®-Werke GmbH & Co. KG, Staufen, Germany) at 11,000 rpm. The process was designed so that the final emulsion always had a total mass of 5 g. The resulting colloidal system was cooled to room temperature under magnetic stirring during 30 min.

ii. Determination of NE Fluidity

[0344] The fluidity of the NE shell was determined by means of DIOLL probe. The DIOLL probe was dissolved in ethanol (263 μM). NE was diluted 1:1000 (0.027% w/v). 10 pL of DIOLL solution was added to 1 mL NE to reach a 2.6 μM final concentration of DIOLL, corresponding to a DIOLL-NE ratio of 1:290 (w/w) The sample was incubated for 20 min at room temperature. As controls, the fluorescence of DIOLL solutions (2.6 μM) in MCT and in water, and the fluorescence of blank NE (without DIOLL) were recorded. In this case, 10 μL of ethanol was added to 1 mL NE. Fluorescence data were obtained using a FP-8500 spectrofluorometer (JASCO applied science, Halifax, Canada). The excitation wavelength was 390 nm and the emission spectra were recorded between 400 and 600 nm at several temperatures ranging from 5 to 27° C. in steps of 2° C. and at 37° C. (2.5 nm bandwidth). The generalized polarization (GP) parameter was calculated from emission intensities according to:

[00007]$\mathrm{GP}=\frac{I_{440}-I_{490}}{I_{440}+I_{490}}$

[0345] where I.sub.440 and I.sub.400 are recorded fluorescence intensities at wavelengths of 440 nm and 490 nm, respectively.

2. Results

[0346] Fluorescence measurements using polarity sensitive fluorophores were performed. Laurdan (6-dodecanoyl-2-dimethylaminonaphthalene) is a polarity sensitive fluorophore commonly used for the assessment of liposomes rigidity and the study of biological membranes. Yet, when incubated with NE solution, its fluorescence emission spectrum was characteristic of a highly hydrophobic environment, single peak at 440 nm, suggesting that the fluorophore was inserted in the hydrophobic core of the NE.

[0347] DIOLL probe was post-inserted in the NE shell by simple mixing the probe with the NE suspension (DIOLL-NE). Insertion of DIOLL in the NE shell was checked by measurements of emission spectra of controls. Due to the lack of fluorescent activity in hydrophilic environment, no fluorescence emission was detected when the probe was dissolved in water. Neither blank NE emitted fluorescence. A single peak at 440 nm was observed when DIOLL was dissolved in MCT related to the radiative de-excitation of a “locally excited” state of DIOLL in oil. Emission spectra of DIOLL-NE at temperatures ranging from 5° C. to 37° C. showed two peaks at 440 nm and 490 nm, (FIG. 12), which, based on literature data on Laurdan, corresponded to fluorophore inserted into a rigid or a fluid environment, respectively, each band depending on the physical state of the medium and on the capacity of the probe to undergo interactions with the surrounding water molecules. So, the inventors could conclude that DIOLL was emitting from NE surfactant shell and not from the MCT core of NE. Indeed, the DIOLL hydrophobic tail could be oriented towards the NE oil core, while the hydrophilic and fluorescent naphthalene moiety could be aligned with the PEG components of S1, towards the surrounding water phase. NE size, Pdl and surface charge did not vary upon insertion of the probe (size: 105±1 nm, Pdl: 0.21, zeta potential: −18±1 mV).

[0348] Decreasing the temperature from 37° C. to 5° C. led to the progressive increase of peak intensity at 440 nm (FIG. 12A). To quantify membrane fluidity state at each temperature, the GP values were calculated as described in Materials and methods section. GP increased as the temperature was lowered (FIG. 12B), proving decreased fluidity of the NE shell. However, GP values remained negative (fluid state) and no significant phase transition from fluid to rigid state occurred over the full temperature range. The NE shell was in its fluid state, even at low temperature (5° C.).

Example 5

Dynamic Following of Cell Membrane Fluidity through a Nutritive Stimulus with DIOLL Probe

1. Material and Method

[0349] a. Chemicals

[0350] DIOLL probe was synthesis as described in Example 1.

[0351] Fetal bovine serum (FBS) and Horse serum were purchased from Eurobio (ref: CVFSVF 00-01, S090H-500), Dulbecco's Modified Eagle's Medium/Nutrient Mixture F-12 (DMEM/F-12) nutrient mixture (Ham) (ref: 31331-028) and penicillin-streptomycin were purchased from Gibco. Human epidermal growth factor (EGF) was purchased from Promocell and insulin was purchased from NovoRapid. Palmitate and BSA were purchased from Merck (Sigma).

b. Confocal Microscopy Experiments

[0352] Cells were cultured in Lab-TEK® at 30% of confluence and DIOLL was added to obtain 0.5 μM 48 hours before observation. During this time, cells were either incubated, 16 h just before the observation, in control medium with 10% SVF or in medium containing Horse serum (10%) or in medium containing 10% SVF and 500 μM of palmitate complexed with BSA (ratio 4/1) to maintain its solubility in acqueous solution.

[0353] 48 hours later, cells were at 80% of confluence and ready for observation.

[0354] The images were taken using a Zeiss LSM 880 AxioObserver inverted spectral confocal microscope. This is a spectral confocal microscope on stand with incubation chamber with environmental control temperature, CO.sub.2 and hygrometry. The light source is a laser diode. The objective is an alpha Plan-Apochromat 63x/1.46 Oil DIC M27.

[0355] For spectral mode acquisition excitation wavelength was 405 nm and images were taken at 424, 432, 441, 450, 459, 468, 477, 486, 495, 504, 513 nm emission wavelength.

c. Determination of the GP Parameter from the Images Obtained by Confocal Microscopy

[0356] A mask was created from the image obtained with the DIOLL by selecting alighted pixels from the background. From this mask, the emission intensity values in the 440 and 490 nm channels were measured using ImageJ. GP values were calculated with the same formula as for spectroscopy experiments:

[00008]$\mathrm{GP}=\frac{I_{440}-I_{490}}{I_{440}+I_{490}}$

[0357] where I.sub.440 and I.sub.490 are recorded fluorescence intensities at wavelengths of 440 nm and 490 nm, respectively.

[0358] GP values were calculated from at least 5 independent measurements.

2. Results and Discussion

Measuring Fluidity Change Dynamically After a Nutritive Stimulus

[0359] To check whether the probe is able to characterize a modification of membrane fluidity after a brief natural stimulus, the inventors verified the ability of the probe to detect and mesure the changes associated to a metabolic perturbation through the nutritive environement of the cells.

[0360] For that, the inventors treated HMEC cells with different media containing different spectrum of natural lipids. Horse serum has been well characterized for its high content in polyunsaturated lipid, at the difference of calf serum containing more saturated and monounsaturated lipids. Therefore, by taking these lipids from their extracellular environement, cells modify dynamically their lipids content into their membranes leading to rapid modification of their fluidity, not related in this case to a complicated genetic regulation. To obtain a counterpart to the Horse serum stimulus, the inventors also treated cells with palmitate, the most commun saturated fatty acid, to induce an opposite rapid rigidification of cell membranes. HMEC cells loaded with DIOLL 36 h before the start of the stimulus were thus briefly (16 h) incubated with these different conditions, and then observed by confocal microscopy. Fluorescence images were recorded each 10 nm from 425 to 512 nm (excitation 405 nm). Different shades of grey corresponding to different fluorescence intensity obtained in each channel leads to the differentiation of the fluidity state of the cell membranes.

[0361] As demonstrated in FIG. 13, Horse serum and palmitate induced indeed a strong modification of membrane fluidity that we could detect and quantify with DIOLL insertion into their membrane. Palmitate strongly increased the overall rigidity of cell endomembranes and induced the appearance of numerous bundles of hyper-rigid plasma membrane. At the opposite, after horse serum treatment, extra-fluid endomembranes occupied almost all the available intracellular space. Therefore, DIOLL was clearly able to capture the rapid change of membrane fluidity in all directions. In consequence, DIOLL can measure a fluidity state responding to a combination of upstream regulation (genetic, epigenetic, metabolic), normally relatively stable, but DIOLL can also follow a dynamic and sudden change of membrane fluidity after a strong cell perturbation involving different kind of stimuli (nutritive perturbation, genetic or epigenetic lesion, infection by a pathogen.

Example 6

Distinguishing Cancer Cells from Healthy Cells with DIOLL Probe

[0362] A large number of pathologies originate in the deregulation of cellular differentiation processes leading to the appearance of non-adapted phenotypes, a phenomenon summarized as abnormal cellular genetic plasticity such as pulmonary, hepatic or cardiac. The mechanisms driving these processes are complex and rely heavily on epigenetic control of the genome. These epigenetic controls can be influenced by environmental stresses such as infections, toxicological agents or poor nutrition, but also by intrinsic genetic alterations such as those found during tumour development.

[0363] In general, cancer cells have greater genetic plasticity than healthy cells. However, there are different expressions of this plasticity which are difficult to quantify since only in vitro functional tests can measure the ability of cancer cells to modulate their differentiation program according to the signals imposed on them. On the other hand, a large amount of data from the literature suggests that the fluidity of cancer cell membranes is increased. the inventors thus assumed that, in general, the fluidity of the biological membranes could represent the degree of plasticity of the cell, thus allowing the measurement of a plasticity index based on a biophysical parameter that can easily be quantified by DIOLL probe.

[0364] In order to demonstrate and validate this concept, the inventors used cancer cells from two different tissues, but with common characteristics allowing their comparison: the mammary gland and the prostate. Among these characteristics, it is possible to highlight the existence of different forms of tumours with varying and well-characterized degrees of dedifferentiation. On one hand, little aggressive tumours can retain their glandular structure and on another hand, cells can retain their dependence on hormones, while conversely, the most aggressive tumours are hormone-independent and totally unstructured. Finally, there are pathologies in which cancer cells are no longer only dedifferentiated, but also acquire new phenotypes that are outliers in relation to their original tissue. These tumours testify that after having seen a sharp increase in their genetic plasticity, the phenotype of cancer cells is fixed in an abnormal state and different from their original state, which is then a priori characterized by a reduction in their plasticity.

[0365] DIOLL probe was synthesis as described in Example 1.

1. Distinguishing Mammary Gland Cell Lines with Known Various Degree of Genetic Plasticity as Function to their Membrane Fluidity

[0366] In order to validate the use of the DIOLL probe to measure the plasticity of cancer cells, the inventors first generated a cell line with a strong increase of its genetic plasticity. Non cancerous human mammary epithelial cells (HMEC) have been transduced by a genetic construct encoding ZEB1, under the control of a promoter repressed in the absence of doxycycline in the cell medium. ZEB1 is an embryonic transcription factor normally expressed in mammary adult stem cell. In this system, its induction can be controlled through cell incubation with doxycycline at 2 μg/ml, leading to a sharp dedifferentiation process and a lost of epithelial identity, a strong marker of cell plasticity. The inventors first verified that ZEB1 activation induced this transient phenotype correlated to an increase of membrane fluidity. Spectrofluorimetry and flow cytometry after simultaneous DIOLL incubation both confirmed the higher fluidity of cells loosing their identity(FIGS. 14A, 14B).

[0367] Then, since this phenotype is normally very unstable and transient, the inventors designed an home-made protocol to fix this unstable state through simultaneous cell incubation in controlled nutrient medium (Glutamine 0.5 mM, Glucose 6 mM, and Pyruvate 2 mM) with a combination of growth factor (TGFb at 10 μg/ml) and hormones (Hydrocortisone 500 nM and Dexamethanose 200 nM). The cells are thus in a metastable phenotype, and can be derived in multiple cell types representing different population normally found in mammary gland but also aberrant fibroblast-like cells. An important characteristic of plastic cells is their higher capacity of pliancy, a phenotype associated with their abiliity to support the stress induced by an oncogenic lesion. The inventors thus verified that metastable cells were indeed completely insensitive to an oncogenic stress produced by the genetic activation of a vector encoding a constitutive active form of the protein Ras. At the opposite, all more differentiated cell line derived from this metastable HMECs were not resistant anymore to this oncogene induction (FIG. 15).

[0368] Secondly, to show the correlation existing between the level of genetic plasticity and the global membrane fluidity, original HMEC cell line, metastable-derived and redifferentiated cells were incubated 48 h with the DIOLL probe. The generalized polarization (GP) parameter was calculated from emission intensities after excitation at 390 nm according to:

[00009]$\mathrm{GP}=\frac{I_{440}-I_{490}}{I_{440}+I_{490}}$

where I.sub.440 and I.sub.490 correspond to measured fluorescence intensities of the whole cells at wavelengths of 440 nm and 490 nm, respectively.

[0369] Highly plastic metastable cells clearly displayed the highest membrane fluidity, while other cell types had lower membrane fluidity (Former HMECs and myoepithelial-like cells presented in the FIG. 16)

[0370] Then, an RNAseq was performed to obtain a genetic signature associated with this genetic plasticity and it used to analyse well characterized breast cancer cell lines (FIG. 17).

[0371] An intrinsic molecular subtype has been associated to these cell lines through the analyse of their transcriptome and this subtype is associated to several phenotypic features used classically to sort mammary tumours. Her2 subtype present for instance a very strong dependence to this oncogene and are mostly epithelial. Basal-like and Claudin-Low tumour are considered as the most aggressive even if, an important heterogeneity exists between the neoplasia belonging to a same molecular subtype. Claudin-Low, very often presenting the acquisition of new and aberrant mesenchymal differentiated traits, have been recently separated in three different groups, some of them being very less aggressive. At the opposite, basal-like tumours are strongly undifferentiated, hormone independent and more homogeneously aggressive.ln contrast, luminal cell lines show some conservation of their original mammary characteristics such as their ability to respond to hormones, their lower capacity to form tumour and metastasis, and the maintenance in the produced tumour of a glandular shape.

[0372] By using their new transcriptomic signature generated from the metastable cell line, the inventors performed a principal component analysis based on epithelial identity and cell plasticity. As expected, most basal cell line were sorted together in an highly plastic group with a partial lost of their epithelial identity. Interestingly, a much more heterogeneous pattern was observed for Claudin-Low and Luminal-b cell lines with some of them being closer of the basal cell lines. Therefore, this analyse capture the known heterogeneity of Luminal-b and Claudin-Low tumour with a significant part of them being more aggressive and resistant.Then from this characterisation the inventors verified that GP coudl similarly classify these cell lines, this procedure being much more simple to perform than a complete genetic analysis.

[0373] Table 5 below summarizes the different lines studied and their known characteristics.

TABLE-US-00005 TABLE 5 Known Mammary gland tumour cell lines and their specifications Human tumour cell lines Specifications Mammary gland MCF-7 Luminal cell lines responding to ZR-75 hormones, which are very close to epithelial cells BT20 Basal cell lines, dedifferentiated HCC1954 HCC1937 HS578T “Claudin-low” cell lines, aberrant MDA-MB-231 mesenchymal differentiation MDA-MB-436

[0374] One control line were also used, HMEC cells representing normal mammary epithelial luminal progenitors

[0375] The results are summarized in FIG. 18. Inventors validated that basal cell lines genetically defined as the most plastic have the most fluid membranes (HCC1954, BT20 and HCC 1937). On the other hand, luminal and “Claudin-Low” cell lines, regardless of the conservation of their epithelial identity, have globally larger GP showing the greater rigidity of their membrane in accordance with their supposed lower plasticity. However, in line with the previous genetic analysis, the inventors can also note the existence of Claudin-Low cell lines such as MDA-MB231 with a membrane fluidity similar to that of the basal lines (FIG. 19). These results, therefore, confirm the existence of a very good correlation between the epigenetic plasticity of the cell and membrane fluidity.

[0376] It is worth noting that the cells lines were distinguished with DIOLL probe regardless of the temperature at which the I.sub.440 and I.sub.490 were recorded.

[0377] Finally, to complete the previous result, the inventors verified the DIOLL capacity to follow the cell plasticity dynamically in cancerous cell line like in the previous metastable model during its differentiation. HCC1954 was subject during three weeks to a differentiating protocol based on nutritive deprivation and a combination of growth factor (TGFb at 10 μg/ml) and hormones (hydrocortisone 500 nM). The fluidity of their membranes were clearly reduced confirming ability of the DIOLL to measure a dynamic reduction of cell plasticity in a cancerous cell line (FIG. 20).

2. Variation of the Membrane Fluidity in Prostate Cell Lines

[0378] Then the inventors verified that the relationship between the degree of genetic plasticity and the fluidity of the membranes was not a particularity of mammary cells but could extend to other neoplasia such as that affecting prostate tissue. 4 cell lines were tested: LNCaP which is a basal cell line, PC-3 which an aggressive metastatic cell line, C4-2B cell line which are mesenchymal cell lines derived from LNCap and WPMY-1 which is an epithelial and mesenchymal control cell line.

[0379] The results are presented in FIG. 21 and confirm the modulation of membrane fluidity as a function of the level of plasticity attributed to these cell lines. It can be noted that the normal mesenchymal WPMY-1 cell line conserves relatively rigid membranes, while the PC-3 cell line, genetically close to basal mammary gland cell lines, has a very high membrane fluidity. Besides, it can be noted that the derived hormone-independent C4-2B cell line, which reflects its de-differentiated state compared to its original LNCaP cell line, has a much greater fluidity than the latter.

[0380] It is worth noting here again that the cells lines were distinguished with DIOLL probe regardless of the temperature at which the I.sub.440 and I.sub.490 were recorded.

3. Variation of the Membrane Fluidity in Exosomes

[0381] The analysis of the membranes of exosomes is of twofold interest. Indeed, on one hand, these vesicles are released into the circulation and are therefore easily accessible, making it possible to consider a less invasive diagnostic tool than a tumour biopsy and longitudinal patient follow-up, as the sampling can be repeated several times. On the other hand, the membranes of these vesicles are not subjected to any mechanical stress related to tissue architecture, unlike the cells recovered by biopsy, which eliminates a factor that strongly constrains membrane fluidity and overlaps with the intrinsic properties of these membranes.

1. Exosomes from Differentiated and Dedifferentiated HMEC Cells

[0382] The exosomes from the HMEC cells described above were purified, as described in Example 3, in order to analyse the GP of their membrane. Regardless of the method used, spectrofluorometry (FIG. 22A) or flow cytometry (FIG. 22B), the inventors demonstrate that the fluidity of the membranes of these exosomes was perfectly correlated with that of the cells from which they originated, thus making them excellent witnesses of the degree of genetic plasticity of these cells.

2. Exosomes from Dedifferentiated HMEC Cells at Different Times of Induction of Dedifferentiation

[0383] Here is studied the kinetic analysis of exosomes from HMEC cells during the lost of their epithelial identity, purified as described in Example 3 (FIG. 23A in spectrofluorometry and FIG. 23B in cytometry). The HMEC cells were mediated from 0 to 12 days by the ZEB1 transcription factor, leading to an evolutive phenotype from differentiated HMEC cells (0 days) to dedifferentiated HMEC cells (12 days).

[0384] The inventors observe that the progressive increase in the fluidity of the membranes of these vesicles is well related to the increase in the level of genetic plasticity of the cells from which they are derived. Therefore, these experiments confirm that the membrane fluidity of exosomes reflects an index of the genetic plasticity of the cells from which their originate.

4. Validation of the Membrane Fluidity in Mouse Tumor

[0385] Finally, inventors verified the ability of membrane fluidity to capture the most plastic cell in an in vivo tumour model. Transgenic mice expressing the c-myc oncogene under a promoter system specifically active in mammary gland and inducible by doxycycline were used and developed epithelial mammary tumour. These tumours can be further made resistant to myc deprivation, acquiring simultaneously a strong plasticity and mesenchymal features. A syngenic model was developed from these tumours recapitulating this course of tumour development in 10 weeks. Importantly, several normal and cancerous populations then co-exist in these tumours, some of them bein characterised as highly plastic from their genetic analysis and a pattern of surface marker (CD29high/CD24high) specifically associated to this model

[0386] These different cell populations from the tumor can be quantified and discriminated by flow cytometry. Of note, when membrane fluidity was analysed simultaneously, highly plastic cells (CD29high/CD24high) had the highest membrane fluidity, whereas normal fibroblasts and erythrocytes presented the strongest rigidity (FIG. 24).

Example 7

Labeling of Viral Membranes and Calculation of GP Parameter to Estimate Membrane Fluidity with DIOLL Probe

[0387] 1. Materials and methods
a. Chemicals

[0388] DIOLL probe was synthesis as described in Example 1.

b. Methods

[0389] Viral samples of purified influenza virus (ref Durous et al. Vaccine, 2019) were incubated with 0.2 μM DIOLL under sterile conditions for 1 h. Fluorescence spectra were recorded using a Jasco fluorimeter in quartz sterile cuvettes. Emission spectra were recorded between 410 and 600 nm with an excitation wavelength fixed at 390 nm and 1 nm excitation and emission slits, at 21° C. and 37° C. GP parameter was calculated as described in Example 1.

2. Results

[0390] DIOLL efficiently labeled viral membranes with a maximum fluorescence intensity at 440 nm (FIG. 25A, full lines) at 21° C., caracteristic of rigid membranes. The shape of the spectrum changed with increasing temperature as the 440 nm contribution decreased when the sample was heated at 37° C. in favour of the 490 nm contribution (FIG. 25A, dotted lines). This indicated that DIOLL fluorescence was sensitive to viral membrane fluidity state. The GP parameter was calculated (FIG. 25B) and confirmed a more rigid membrane state at 21° C. (GP=0.315) and a more fluid membrane at 37° C. (GP=0.096).