CLOT RETRIEVAL DEVICES AND THERAPEUTIC USE THEREOF

Abstract

The present invention relates to improved clot retrieval devices, wherein at least the part of the device intended to be in contact with a clot is coated with at least one chromatin binding agent selected from the group consisting of digoxigenin, a digoxigenin derivative, distamycin and a distamycin derivative. The present invention also relates to a method of producing said improved clot retrieval device. The present invention also relates to a method of prevention and/or treatment of stroke in a subject, comprising removing at least one clot using the clot retrieval device coated with said least one chromatin binding agent.

Claims

1. A clot retrieval device, wherein at least the part of said device intended to be in contact with a clot, comprising a coating comprising at least one chromatin binding agent, wherein said chromatin binding agent is selected from the group consisting of digoxigenin, a digoxigenin derivative, distamycin and a distamycin derivative.

2. The clot retrieval device according to claim 1, wherein said clot retrieval device is a stent retriever.

3. The clot retrieval device according to claim 1, wherein said chromatin binding agent is coated on the part of the device intended to be in contact with a clot via at least one coating agent.

4. The clot retrieval device according to claim 3, wherein said coating agent is PDA or PDA-PEG bis amine.

5. The clot retrieval device according to claim 3, wherein at least the part of said device intended to be in contact with a clot is coated with at least one coating agent and said chromatin binding agent is linked to said coating agent.

6. The clot retrieval device according to claim 1, wherein the digoxigenin derivative is a compound of the following formula (III) ##STR00009## wherein L is a linker comprising at least one group selected from the group consisting of an amino, ester, amido, carboxy and hydroxyl group.

7. The clot retrieval device according to claim 1, wherein the digoxigenin derivative is digoxigenin NHS ester and/or the distamycin derivative is distamycin azide.

8. The clot retrieval device according to claim 3, wherein (i) the chromatin binding agent is digoxigenin NHS ester and the coating agent is PDA PEG bis-amine or (ii) the chromatin binding agent is distamycin azide and the coating agent is PDA.

9. A method for producing a clot retrieval device according to claim 1, wherein said method comprises: a) coating at least the part of the device intended to be in contact with a clot with at least one coating agent, and b) contacting the part of the device coated with at least one coating agent obtained in step a) with at least one chromatin binding agent.

10. The method according to claim 9, wherein step a) comprises contacting at least the part of the device intended to be in contact with a clot with a solution comprising dopamine and, optionally, PEG bis-amine.

11. A method for removing a clot in a subject, wherein said method comprises removing said clot with a clot retrieval device according to claim 1 or obtained by the method according to claim 9.

12. A method for preventing and/or treating stroke in a patient in need thereof, wherein said method comprises removing at least one clot with a clot retrieval device according to claim 1 or obtained by the method according to claim 9.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0140] FIG. 1: Experimental plan for assessing chromatin binding

[0141] FIG. 2: Optimization of the copolymer for CFSE-NHS ester binding. Different polymerization time (17 or 40 hours), PDA, and PEGba concentration (mg/ml) were evaluated. Each dot represents an individual sample. Effective binding of the CFSE-NHF ester was measured by the peak±SD fluorescence on each sample in the different conditions, both at 17 h and 40 h of polymerization. *PEGba was used after (not together with) PDA). Ordinate: Fluorescence intensity peak value (A.U.).

[0142] FIGS. 3A-3B: Digital images of fluorescence intensity. FIG. 3A: Fluorescence intensity of DIG NHS ester at 125 μg/mL in PBS 1× and in sodium bicarbonate; FIG. 3B: The higher fluorescence intensities are both for sodium bicarbonate and PBS 1× at 125 μg/mL. Fluorescence intensities show PBS 1× has a higher fluorescence but a wider peak than sodium bicarbonate.

[0143] FIG. 4: Fluorescence intensity of BMC, PDA2, PPba13 and their respective negative controls. PPba13 disks have high fluorescence intensities values with peaks, whereas PDA2 negative control and PDA2 disk are spread along the X-axis. BMC negative control and BMC disks have narrow peaks with fewer fluorescence intensities. Ordinate: Mediane Leverage Residuals

[0144] FIGS. 5A-5C: Analysis of DNA binding by Sytox green. FIG. 5A: analysis of the signal homogeneity, via the comparison of the different fields (“champs”). FIG. 5B: analysis of DNA binding extent among groups. FIG. 5C Intersection of the intersection of the two variables (field and group effect). BMC values are in red, PDA values are in blue, and PPba values are in green. There is no champ effect, but a group effect, and no interdependency between champs and groups.

[0145] FIG. 6: Median fluorescent intensity of BMC negative control, BMC, PDA2 and PPba13 samples. The median fluorescence intensity of PPba13 sample is higher to the other samples.

[0146] FIGS. 7A-7D: Medians comparison for negative controls. FIG. 7(A): One-Way ANOVA analysis FIGS. 7B, 7C and 7D: Champ effect FIG. 7B, group effect FIG. 7C and champs x groups FIG. 7D for signal/noise ratio (Leverage residuals) in ordinate. BMC are represented in red, PPba13 in green, and PDA2 in blue. The PDA2 background noise is significantly higher to the others sets. There is only a group effect. (Ordinate for A: median)

[0147] FIG. 8: Enhanced binding properties exerted by Distamycin-coated disks on thrombus-associated NETs. Each dot represents an independent test disk.

EXAMPLES

Materials and Methods

Materials and Chemicals

[0148] Cobalt-chromium (L605) disks (4.8 mm diameter, 0.25 mm height, polished on one side) were obtained from Goodfellow (Lille, France). Their composition is described below (Table 1).

TABLE-US-00001 TABLE 1 L605 cobalt-chromium chemical composition Element Co Cr W Ni Fe Mn Chemical 50 20 15 10 3.0 2.0 composition (% m)

[0149] Tris(hydroxymethyl)aminomethane (Tris) buffer (CAS No. 77-86-1, 10 mM, pH adjusted to 8,5) and Sodium bicarbonate buffer (CAS No. 144-55-8, 100 mM, pH=8,3) were filtered with a 250 mL sterile filter system (polyether sulfone membrane, 0,22 μm pore size, Corning) prior to their use. Ethanol (96% pure) and acetone (99,8%, AnalaR NORMAPUR®) were purchased from VWR Prolabo Chemicals. Dopamine powder (Dopamine hypochloride, Alfa Aesar, A11136, 99% pure), ε-(Digoxigenin-3-0-acetamido)caproic acid N-hydroxysuccinimide ester (DIG NHS ester, 55865-5MG-F), polyethylene glycol bis-amine (PEG bis-amine, P9906) were purchased from Sigma-Aldrich. Antifading agent (Pro Long™ Gold), carboxyfluorescein succinimidyl ester (CFSE, C34554), and SYTOX Green (S7020) were purchased from Thermo Fischer Scientific.

Determination of an Optimal Copolymer for L605 Alloy Coating

[0150] (i) Determination of the copolymer optimal initial concentration of dopamine

Disks Preparation

[0151] Each disk was marked with a diamond point to identify the face in contact with the well bottom at each step. The marked disks were ultrasonically polished (VWR, Ultrasonic Cleaner) in three successive baths (acetone, ethanol and finally distilled water), 10 minutes for each bath, in order to remove oxidation and organic residues. Next the disks were separated in 3 groups: untouched (Bare Metal Controls, BMC), coated with polydopamine (PDA Controls, PDAc) and copolymers samples (PDA+PEG bis amine, PPba). After the polishing step, BMC were placed in distilled and filtered water and left under a sterile culture hood, to avoid contamination. The other disks were washed 3 times in Tris during 3 minutes under stirring onto an orbital shaker (Grant-Bio, PMS-1000i) at 400 rpm.

Surface Modification by Coating

[0152] Disks in the PDAc group were immersed in dopamine only solution (1 or 2 mg/ml) in Tris buffer, whereas PPba disks were immersed in a mixture of dopamine (fixed concentration: 1 mg/mL) and PEG bis-amine solution (concentration varying from 1 to 6 mg/mL) in Tris buffer. Polymerization was allowed by incubation of individual disks in the wells of a sterile and transparent polystyrene 48-wells microplate (under stirring at 400 rpm, in the dark, at room temperature) during 17 hours. Polymerization was stopped by repeated washes in clean distilled water under stirring (400 rpm, 3 minutes each), and a brief sonication (limited to 15 seconds to avoid damaging the coatings), after the last wash.

Analysis of NHS-Ester Binding by CFSE

[0153] For each group (BMC, PDAc, PPba), 3 disks were immersed in a CFSE solution (50 μg/mL in sodium boicarbonate) for 30 minutes at room temperature under stirring (400 rpm), in the dark. The reaction was stopped by washing two times with distilled water (5 minutes each, under stirring at 400 rpm) and the disks were placed, marked face up, onto 20 μL of antifading mounting medium on the bottom of individual wells of 12-wells-Ibidi® slides (Ibidi, Germany) and left untouched for at least 30 minutes in the dark (to allow the antifade mounting medium to polymerize and render the observation by inversed fluorescent microscope possible.

Fluorescence Microscope Analysis

[0154] Digital images were acquired on the coated face of each disk in the green fluorescent channel (ex 475/40 em 530/50 nm) of an Axio Observer® microscope (Zeiss, Germany). The time of exposure was set on the BMC sample that had not been contacted with CFSE (background fluorescence). A second background noise control was obtained by the emission in the near infrared light channel (ex 640/30 em 690/50). The fluorescence intensity histograms from both channels were calculated using the proprietary software (ZEN®, Zeiss, Germany) and raw data were exported and further analyzed using the software Microsoft® Excel. [0155] (ii) Evaluation of chromatin adhesive properties of DIG-coated disks by fluorescence microscopy and SYTOX Green

Determination of Optimal Conditions for Maximal DIG NHS Ester Coating

[0156] DIG NHS ester was diluted in cascade by a 1:2 factor, in either PBS 1× or sodium bicarbonate (10 mM, pH 8.3), from 500 μg/mL to 16,625 μg/mL. PPba disks were immersed individual solutions during 30 minutes, under stirring (400 rpm) at room temperature and in the dark, as for the CFSE ester. The reaction was stopped by washing three times with distilled water (5 minutes each) under stirring. DIG that had effectively onto the disks was detected by indirect immunofluorescence, using a biotynilated anti-DIG NHS ester antibody (NBP2-31191B, Novusbio, USA) and streptavidin PE (554061BD Biosciences, USA) as well as by chromogenic development and absorbance detection, using streptavidin-HRP (557630, BD Biosciences, USA) and TMB (T0440, Sigma). Background signal was set on disks that had been contacted with the buffer alone, in the absence of DIG-NHS). Absorbance was analyzed on the supernatant, on a Infinite® 200 Pro plate reader (TECAN, Switzerland) whereas the amount and distribution of PE at the surface of the washed disks was analyzed by fluorescence microscopy on Ibidi 8 plates, as previously described, in the orange channel (ex 545/25, em 605/710 nm).

Genetic Material Extraction

[0157] Genetic material (genomic DNA and associated proteins) derived from laboratory mouse tails and was obtained using the “F-355L DNA Release” kit (Thermo Scientific). Since the binding of digoxigenin to chromatin is thought to occur via its interaction with topoisomerase, we assessed the presence of proteins associated with the extracted DNA, using the kit “Pierce™ BCA Protein Assay Kit” from Thermo Scientific, prior to using this material.

Functional Analysis of DNA Binding onto Coated Disks

[0158] Disks from the three groups were coated in optimal DIG-NHS conditions (125 μg/mL in sodium bicarbonate buffer), washed three times in distilled water and immersed in a solution of mouse tail-derived genetic material (diluted 1:10 in PBS 1× corresponding to a final concentration of 87,3 μg/mL total protein, as determined using the Pierce™ BCA protein assay kit, Thermo Scientific). The incubation with genetic material was carried out at 37° C. and lasted 5 minutes (to mimic the time of contact between the stent retrievers and the clot in stroke patients). The disks were then thoroughly rinsed and immersed in SYTOX Green (167 nM in buffer) for 30 minutes and washed three time with distilled water. Finally, the disks were placed, marked face up, onto anti-fading mounting medium at the bottom of individual wells in a 8-wells Ibidi® slide and analyzed in the green channel.

[0159] Artificial thrombi were generated using peripheral human whole blood, collected in dry tubes and stimulated with PMA to generate Nets in order to obtain thrombi containing extracellular Chromatin;

[0160] Fresh thrombi were retrieved from patients with stroke.

Results and Discussion

[0161] The presence of chromatin was evaluated at the thrombus surface and in its inside (after its sectioning), by epifluorescence microscopy. The analyzed whole thrombus were retrieved from a cerebral artery in stroke patient.

[0162] A dense mesh of chromatin was present in all the analyzed fresh thrombi (n=5) (data not shown).

[0163] It was first used a two-step dip-coating using PDA in the first bath and PEGba in a subsequent bath. This strategy was intended to orient the grafting of PEGba by one of its amine arms and leave the other amine group available for the subsequent reaction with the NHS ester, in order to drive a unidirectional binding with DIG instead of having disordered bindings. The concentration of PDA at 2 mg/ml was based on the most widely used. The concentration of PEGba was set to 6 mg/mL, to reproduce the ratio 3:1 that had previously been reported as ideal in combination with another self-assembling polymer. Contrary to what expected, the grafting of the CFSE NHS ester was higher onto PDA alone than on PDA+PEGba (see FIG. 2), suggesting that the both of the amines of the PEGba had reacted with the cathecols of the PDA polymer and none was left free and available to react with the NHS ester in the subsequent step.

[0164] The strategy was therefore changes and PEGba and PDA were used in the same polymerization bath: the excess in amine groups was supposed to allow the formation of a copolymer with free amines functions at its surface, available for the subsequent reaction with the NHS ester. A copolymer made using the same concentration of the two reagents (2 mg/mL of initial dopamine and 6 mg/mL of PEGba) was compared to PDA alone at 2 mg/ml, BMC, and various concentrations of PEGba used in combination with a fixed concentration of PDA at 1 mg/mL (see FIG. 2). The results show that the use of PDA 1 mg/ml in combination with PEGba 3 mg/ml during 40 h gives the best copolymer condition for subsequent grafting of a NHS ester (CFSE in the case of FIG. 2). This condition has therefore been chosen as the working copolymer solution for the subsequent studies.

[0165] Five different concentrations (cascade 1:2 dilutions) of DIG-NHS ester, ranging from 31.25 to 500 μg/ml, prepared in either sodium bicarbonate (100 mM, pH 8.3) or PBS (pH 7.4) were evaluated. The effective grafting of DIG NHS ester onto the PPba copolymer was evaluated using indirect immunofluorescence and an antibody specifically directed against the DIG NHS ester.

[0166] As shown in FIG. 3, the highest disk coverage with DIG, as detected by immunofluorescence intensity, was obtained at the concentration of 125 μg/mL. Both sodium bicarbonate and PBS buffer were effective, but the fluorescence curve obtained with PBS 1×, although yielding a higher peak, was wider than the one obtained with sodium bicarbonate, meaning the density of DIG NHS ester onto the copolymer was more heterogeneous. Sodium bicarbonate and a concentration of 125 μg/ml of DIG NHS was therefore retained for the subsequent functional testing with genetic material.

[0167] PPba (1:3) disks were immersed in 125 μg/ml DIG NHS ester solution in bicarbonate sodium buffer, rinsed and contacted during 5 minutes at 37° C. with the genetic material (obtained by extraction form the tail of laboratory mice, diluted in PBS 1×; final protein concentration set to 87,3 μg/mL). The experimental samples were then rinsed and the binding of DNA revealed by SYTOX Green. BMC and PDA-coated disks served as control for specific binding to DIG (vs passive adsorption on the surface).

[0168] The fluorescence signal obtained at 5× magnification from the different samples was very large (the curves do not display a narrow “peak” but rather a spread along the X-axis) suggesting that the presence of DNA was globally inhomogeneous (see FIG. 4).

[0169] The fluorescence of SYTOX green was assessed at higher magnification (×20) in 5 random fields, on each disk. As shown in FIG. 5, one-way ANOVA analysis ruled out the inhomogeneity of DNA binding as there was no “champs” (field) effect (A, champs leverage, p=NS) in any of the groups (C, champs*group Leverage, P═NS). In contrast, the higher magnification analysis, possibly thanks to the more consistent fluorescence signal, showed that the binding of DNA, as detected by SYTOX green signal, was highest on the disks coated with PPba (green) as compared to those coated with PDA (Blue) or the bare metal controls (BMS Red). These data demonstrate that the affinity binding of DNA through DIG-coated surfaces is consistently superior to what can be achieved with the passive adsorption of genetic material onto the stent-compatible alloy surfaces (panel B, group leverage, p<0.0001).

[0170] The superiority of functionalized surfaces vs bare metal and control coated surfaces is clear in FIG. 6, which shows a representative fluorescent curve in the SYTOX green channel to compare each group

[0171] However, the use of fluorescence as a reading out for the functional studies could be biased by the background noise (fluorescence of the negative control in the SYTOX green channel) yielded by the aromatic ring of dopamine, as reflected by the higher background detected in the PDA-coated samples (Median Fluorescence Intensity=38.6 in PDA vs 13 in BMC, FIG. 7 A, p<0.001). However, PDA was also contained in the copolymer but its concentration was reduced by half and this might explain why the median background fluorescence intensity observed in PPba samples is very low and not different from BMC controls (see FIG. 7A). When using the signal/noise ratio (fluorescence in the experimental disks/fluorescence in the control disks) for comparing data between groups, the statistical analysis indicates a group effect (see FIG. 7C, p<0.001) at least as strong as the one observed by comparing absolute signal values from experimental disks (FIG. 5) where the median fluorescence intensity in PDA was likely biased by the background noise (contrary to BMC and PPba disks).

[0172] The protocol used for cobalt-chromimum disked was used for the coating of disks made of nitinol (another stent-suitable alloy) and the performance of digoxigenin-coated surface in terms of capture of soluble DNA was confirmed (data not shown).

[0173] The coating with distamicyn azide has been produced by a copper-free chemistry protocol.

[0174] The ability of fragments of clot retrieval devices coated with digoxigenin-NHS vs those coated with Distamicin-Azide vs bare metal disks to capture DNA and protein material from artificial (in vitro) and from patient's (ex-vivo) thrombi was then evaluated.

[0175] To this aim, the tested devices (n=>3/group) were inserted manually in fragments of the same thrombus and hold still within a 3D printed recipient immersed in human plasma (5 minutes, at 37° C.). The tested devices were then removed manually, incubated with DAPI 1 μg/ml and Evan's Blue 1% in PBS for 5 minutes. After careful washing with PBS, the tested devices were mounted using Prolong Gold anti-fade mounting medium in glass bottom dishes and the presence of thrombus captured at their surface was examined by fluorescence microscopy.

[0176] Digoxigenin and Distamycin coated devices appeared to be effective for improving the capture of thrombus material as compared with uncoated (bare metal) devices (data not shown). The experiments have been repeated at least three times, yielding similar results (qualitative analysis only).

[0177] Nitinol disks were dip-coated in three serial baths composed of [0178] 1. Polydopamine (1 mg/ml in tris buffer, pH 8.5, 22±2 hours) [0179] 2. DBCO-PEG 4-Amine [0180] 3. Azide-derivatized Distamycin or a control irrelevant compound (Control azide).

[0181] A fresh thrombus was prepared with human whole blood added to PMA 100 nM and ionomycin 1 μM to promote strong leukocyte activation leading to the release of extracellular chromatin within the thrombus.

[0182] Slice of the same thrombus were layered onto coated and bare metal disks in the wells of a 48-well plate, covered with human plasma and incubated at 37° C. on an orbital shaker for 5 minutes.

[0183] The thrombus slices were mechanically removed using tweezers, the disk washed in PBS and stained with a solution made of Hoechst 33342 (1 μg/ml, to stain DNA) in PBS for 5 minutes. After careful washing, the presence of thrombus associated—extracellular DNA (NETs) was revealed by fluorescence microscopy in the blue channel.

[0184] As shown in FIG. 8, distamycin-coated disks have enhanced binding properties on thrombus-associated NETs, by comparison to control conditions (bare metal or control azide).

[0185] A coating comprising digoxigenin or distamycin specifically confers the ability to bind genetic material onto stent-compatible metal surfaces. Distamycin and Digoxigenin coated struts both improve the capture of patient's thrombus material as compared with uncoated struts. This bioactive coating is thus suitable for improving the performances of clot retrieval devices.

[0186] Besides, the material used for the coating is natural, biocompatible and easily available, and is thus suitable for production at the industrial scale.