COMBINATION OF PHARMACOLOGICAL AND MICROFLUIDIC FEATURES FOR IMPROVED PLATELETS PRODUCTION

Abstract

The present invention relates to an ex vivo method for producing platelets including a combination of use of pharmacological substances and microfluidic device features, for high yield and high quality platelet production from megakaryocytes or their progenitors.

Claims

1.-10. (canceled)

11. An ex vivo method that comprises a perfusion stage for producing platelets from megakaryocytes, the perfusion stage comprising: providing a fluidic device comprising a production chamber including at least one channel, the at least one channel having a width that is delimited by non-porous walls, one inlet opening at one end and one outlet opening at another end; introducing a suspension of cells comprising progenitors of megakaryocytes, mature megakaryocytes, proplatelets, and platelets into the inlet opening of the at least one channel; subjecting the suspension of cells to a flow from the one inlet opening to the one outlet opening with the mature megakaryocytes flowing across the entire width of the at least one channel under a shear rate causing elongation and fragmentation of the mature megakaryocytes into proplatelets and platelets within 2 hours of subjecting the suspension of cells to the flow; adding at least one megakaryocyte modulator compound to the suspension of cells while the mature megakaryocytes of the suspension of cells are in the flow, and wherein the at least one megakaryocyte modulator is an inhibitor of a Rho/ROCK pathway; and collecting the suspension of cells at the one outlet opening of the at least one channel.

12. The method of claim 11, wherein the inhibitor of the Rho/ROCK pathway increases fragmentation of the mature megakaryocytes into the proplatelets and platelets while the mature megakaryocytes are in the flow.

13. The method of claim 11, wherein each channel of the fluidic device is coated with at least one ligand with a binding affinity for megakaryocytes selected from the group consisting of von Willebrand factor (VWF) or its functional variants, polypeptides comprising fragments of von Willebrand factor, fibrinogen, and fibronectin.

14. The method of claim 11, wherein the suspension of cells is obtained by the following steps: providing stem cells selected from the group consisting of CD34+ cells from human umbilical cord blood; culturing the stem cells for expanding the cells and differentiating the expanded cells into mature megakaryocytes; washing the obtained suspension to remove the culture medium; and re-suspending the cells in a perfusion medium.

15. The method of claim 11, wherein each channel of the fluidic device has a substantially square or rectangular section.

16. The method of claim 11, wherein the at least one channel forms a single flow from the one inlet opening to the one outlet opening.

17. The method of claim 11, wherein each channel has a height H between 5 μm and 1 mm and the width is between 100 μm and 5 mm.

18. The method of claim 11, wherein each channel has a height H between 25 μm and 100 μm.

19. The method of claim 11, wherein the width W is between 300 μm and 800 μm.

20. The method of claim 11, wherein the production chamber comprises a plurality of parallel channels.

21. The method of claim 11, wherein there is no fluid communication through the non-porous walls between adjacent channels of the at least one channel for the mature megakaryocytes to be trapped or squeezed.

22. The method of claim 11, wherein each channel of the at least one channel is not divided internally into sub-channels.

23. An ex vivo method that comprises a perfusion stage for producing platelets from megakaryocytes, the perfusion stage comprising: providing a fluidic device comprising a production chamber including at least one channel, the at least one channel having a width that is delimited by non-porous walls, one inlet opening at one end and one outlet opening at another end, the at least one channel forming a single flow from the one inlet opening to the one outlet opening; introducing a suspension of cells comprising progenitors of megakaryocytes, mature megakaryocytes, proplatelets, and platelets into the inlet opening of the at least one channel; subjecting the suspension of cells to a flow from the one inlet opening to the one outlet opening with the mature megakaryocytes flowing across the entire width of the at least one channel under a wall shear rate of at least 300 s-1 causing fragmentation of the mature megakaryocytes into proplatelets and platelets; adding at least one megakaryocyte modulator compound to the suspension of cells while the mature megakaryocytes of the suspension of cells are in the flow, and wherein the at least one megakaryocyte modulator is an inhibitor of the Rho/ROCK pathway; and collecting the suspension of cells at the one outlet opening of the at least one channel.

24. The method of claim 23, wherein the elongation and fragmentation of the megakaryocytes into proplatelets and platelets occurs in the at least one channel without trapping or squeezing the mature megakaryocytes between adjacent channels of the at least one channel.

25. The method of claim 23, wherein the suspension of cells is subjected to a flow in the at least one channel under a wall shear rate between 300 s-1 and 5000 s-1.

26. The method of claim 23, wherein the suspension of cells is subjected to a flow in the at least one channel under a wall shear rate between 600 s-1 and 2400 s-1.

27. An ex vivo method that comprises a perfusion stage for producing platelets from megakaryocytes, the perfusion stage comprising: providing a fluidic device comprising a production chamber including at least one channel, the at least one channel having a width that is delimited by non-porous walls, one inlet opening at one end and one outlet opening at another end, the at least one channel forming a single flow from the one inlet opening to the one outlet opening; introducing a suspension of cells comprising progenitors of megakaryocytes, mature megakaryocytes, proplatelets, and platelets into the inlet opening of the at least one channel; subjecting the suspension of cells to a flow from the one inlet opening to the one outlet opening with the mature megakaryocytes flowing across the entire width of the at least one channel under a shear rate causing elongation and fragmentation of the mature megakaryocytes into proplatelets and platelets; adding at least one megakaryocyte modulator compound to the suspension of cells while the mature megakaryocytes of the suspension of cells are in the flow, and wherein the at least one megakaryocyte modulator is an inhibitor of the Rho/ROCK pathway; collecting the suspension of cells at the one outlet opening of the at least one channel; and reintroducing the collected suspension of cells into the one inlet opening for reperfusion in the at least one channel.

28. The method of claim 27, wherein prior to reintroduction into the one inlet opening of the at least one channel, the collected suspension of cells is cultured for further maturation of megakaryocyte.

29. The method of claim 27, wherein the mature megakaryocytes are not trapped or squeezed between adjacent channels of the at least one channel.

30. The method of claim 27, wherein the elongation and fragmentation of the megakaryocytes into proplatelets and platelets occurs in the at least one channel without trapping or squeezing the mature megakaryocytes between adjacent channels of the at least one channel.

Description

[0260] This invention will be further understood in light of the following non-limiting examples, which are given for illustration purposes only, and also in connection with the attached drawing in which:

[0261] FIG. 1 illustrates the effect of ROCK inhibitor (Ri) added during the perfusion stage in the absence (left panels) or in the presence (right panels) of Ri during the culture stage: (A) Ratio (no Ri vs addition of Ri during the perfusion stage) of means of surface density of captured megakaryocytes (black line) and elongated megakaryocytes (gray line). Means t SEM is calculated on n=16 channels of the microfluidic device. Statistical analysis was performed using Student t-test (* p<0.05, ** p<0.01, *** p<0.001). The count is made after 35 min of perfusion. Data of one experiment out of 3 are shown. (B) Large-scale images illustrating one embodiment of Ri addition effect in the fluidic suspension during the perfusion stage. Images presented were taken at the position x=1 cm from the beginning of the textured area after 35 min of perfusion of a megakaryocytes suspension cultured in the absence (left panels) or in the presence (right panels) of Ri; and in the absence (top panel) or the presence (bottom panel) Ri in the fluidic suspension during the perfusion stage. The scale bar represents 100 μm.

[0262] FIG. 2 illustrates the effect of VWF on the cell capture and elongation. (A) Ratio (VWF surface coating vs BSA surface coating) of means of surface density of captured megakaryocytes (black line) and elongated megakaryocytes (gray line). Means±SEM is calculated on n=16 channels of the microfluidic device. Statistical analysis was performed using Student t-test (*** p<0.001). The count is made for each experiment after 35 min of perfusion with a megakaryocytes solution containing ROCK inhibitor (Ri) during perfusion stage as well as megakaryocytes culture stage. (B) Typical images of the surface after 35 min of perfusion when using VWF coated microchannels (top panel) and BSA coated microchannels (bottom panel) at the position x=1 cm from the beginning of the textured area. The scale bar represents 100 μm.

[0263] FIG. 3 is an image illustrating the elongation and fragmentation of a mature megakaryocyte from an inlet solution containing ROCK inhibitor, which does not alter the elongation and fragmentation process. The time montage shows a trapped megakaryocyte with a bead-on-a-thread-like structure undergoing one rupture (black arrow) of the elongated structure. White arrows show the extremity of the elongated megakaryocyte. The scale bar represents 20 μm.

[0264] FIG. 4 is a graph representing the ratio (experimental condition vs twenty-four hours static platelet (STAT 24 h) production). Experimental conditions consist in twenty-four hours static platelet production in the presence of ROCK inhibitor (Ri) (STAT 24 h Ri), platelet production from megakaryocyte suspension in the absence (+/+2 h) or in the presence (+/+2 h Ri) of Ri, both after two hours of perfusion through the microfluidic system. No Ri was added during the culture stage on the top graph, whereas Ri was added for the bottom panel. Platelet concentrations ([PLP]) are obtained by counting in a hemocytometer. Means±SEM (top graph: STAT 24 h: n=3, STAT 24 h Ri: n=2, others conditions: n=6; bottom graph: n=2 for each conditions) of the ratio of the platelet production are provided. Statistical analysis was performed using Student t-test (* p<0.05, ** p<0.01).

[0265] FIG. 5 is a picture showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the perfusion stage from a megakaryocytes suspension cultured (A) in the absence and (B) in the presence of ROCK inhibitor (Ri) during the culture stage. For each panel, experimental conditions consist in twenty-four hours static platelet in the absence (STAT 24 h) and in the presence (STAT 24 h Ri) of Ri; platelet production during the perfusion stage from megakaryocyte suspension in the absence (+/+2 h) or in the presence (+/+2 h Ri) of Ri.

[0266] FIG. 6 is a picture showing single color flow cytometry analysis of platelet receptor density, indicating the number of CD61, CD42b and CD49b receptors on the surface of platelets produced from a megakaryocytes suspension cultured in the absence (top graph) and in the presence (bottom graph) of ROCK inhibitor (Ri) during the culture stage. For each, controls consist in platelets prepared from a blood sample of one donor (black bar), as well as a twenty-four hours static platelet production from megakaryocyte in the absence (STAT 24 h) or in the presence (STAT 24 h Ri) of Ri. Experimental conditions consist in platelet production during the perfusion stage from megakaryocyte suspension in the absence (+/+2 h) or in the presence (+/+2 h Ri) of Ri. n=1 each.

[0267] FIG. 7 is a picture showing microphotographs of adhesion of platelets produced in the microfluidic system from a megakaryocytes solution in the presence (+/+2 h Ri) or the absence (+/+2 h) of ROCK inhibitor (Ri) during the two hours of the microfluidic perfusion as well as twenty-four hours static platelet production from megakaryocytes in the absence of Ri (STAT 24 h). Indirect immunofluorescence labeling with an anti-α-tubulin antibodies, revealed by a secondary AlexaFluor488 anti-mouse antibody and AlexaFluor546 phalloidin for F-actin staining is performed in the absence (left panel) or presence (right panel) of PAR-1 thrombin modulator (TRAP6). Image acquisition was performed with an Axio Observer microscope (Zeiss) at 40×1.6-fold magnification with a QIClick-F-CLR-12 Digital CCD Camera (Q Imaging). Platelets are adherent to fibrinogen.

[0268] FIG. 8 is a picture showing two colors flow cytometry analysis of platelets receptors with or without activation of the CD41/CD61 receptor by the agonist peptide of the PAR-1 thrombin receptor (TRAP6). (A) Histogram indicates, within the platelet gate, the % of PAC-1/CD42b positive elements before (black bars) or after (gray bars) TRAP-6 activation of platelets produced in the microfluidic device from megakaryocytes suspension in the absence (+/+2 h) or the presence (+/+2 h Ri) of ROCK inhibitor (Ri). (B) displays the corresponding dot plots within the platelet gate, indicating the population of PAC1.sup.+/CD42b.sup.+ platelets in the non-stimulated (left panels) and TRAP-6 stimulated samples (right panels) produced in the microfluidic device from megakaryocytes suspension in the absence (+/+2 h) or the presence (+/+2 h Ri) of Ri, as well as the control twenty-four hours static platelet production in the presence of Ri (STAT 24 h Ri). Notice the shift of the PAC1.sup.+/CD42b.sup.+ population in the TRAP-6 stimulated platelets produced in the microfluidic device in the absence or the presence of Ri during the perfusion stage. No Ri was added during the culture stage. n=1.

[0269] FIG. 9 is a picture showing two colors flow cytometry analysis of platelets receptors in the absence or presence of activation of the CD41/CD61 receptor by the agonist peptide of the PAR-1 thrombin receptor (TRAP6). Panels A and B depict data as described in the legend of FIG. 8 except that Ri was added during the culture stage. n=1.

[0270] FIG. 10 (A) is a graph representing the concentration of platelet produced after 2 hours of perfusion ([PLP].sub.end). Experimental conditions consist of platelet production from megakaryocyte suspension in the absence (+/+) or in the presence (+/+ NIC) of nicotinamide, both after two hours of perfusion through the microfluidic system. A control without the microfluidic chip was performed in the absence (+/−) or in the presence (+/− NIC) of nicotinamide during the 2 hours of perfusion. No nicotinamide was added during the culture stage for the 4 first experimental conditions, whereas nicotinamide was added for three last experimental conditions. Platelet concentrations ([PLP]) are obtained by counting in a hemocytometer. Means±SEM (+/+ and +/+ NIC: n=3; others conditions: n=2) of platelet concentration are provided. (B-C) are pictures showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the perfusion stage from a megakaryocytes suspension cultured (B) in the absence and (C) in the presence of Nicotinamide during the culture stage. For each panel, experimental conditions consist platelet production during the 2 hours perfusion stage from megakaryocyte suspension in the absence (+/+) or in the presence (+/+ NIC) of Nicotinamide. Additional controls, consisting of 2 hours perfusion stage without microfluidic chip (+/−) and in the absence (+/−) or in the presence (+/− NIC) during these 2 hours of perfusion, were performed.

[0271] FIG. 11 (A) is a graph representing the ratio (experimental condition vs +/+(without any compound introduction) production). Experimental conditions consist of platelet production ([PLP].sub.end) from megakaryocyte suspension in the absence (+/+) or in the presence (+/+ BLEBB) of blebbistatin (BLEBB), and in the presence of DMSO (+/+ DMSO) as a control, during the perfusion, all after two hours of perfusion through the microfluidic system. No blebbistatin was added during the culture stage for the three first experimental conditions, whereas blebbistatin was added during the culture stage for three last experimental conditions. Platelet concentrations ([PLP]) are obtained by counting in a hemocytometer. Means±SEM (n=3) are provided. (B-C) are pictures showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the perfusion stage from megakaryocyte suspension cultured (B) in the absence and (C) in the presence of blebbistatin. For each panel, experimental conditions consist of platelet production during the 2 hours of perfusion stage from megakaryocyte suspension in the absence (+/+) or in the presence (+/+ BLEBB) of blebbistatin. Additional control, consisting of 2 hours perfusion stage with introduction of DMSO (+/+ DMSO) is provided.

[0272] FIG. 12 (A) is a graph representing the ratio (experimental condition vs +/+(without any compound introduction) production). Experimental conditions consist of platelet production ([PLP].sub.end) from megakaryocyte suspension in the absence (+/+) or in the presence (+/+LA) of latrunculin (LA) during the perfusion, both after two hours of perfusion through the microfluidic system. No latrunculin was added during the culture stage for the two first experimental conditions, whereas latrunculin was added for two last experimental conditions. Platelet concentrations ([PLP]) are obtained by counting in a hemocytometer. n=1 of platelet concentration is provided. Means±SEM (+/+ DMSO without LA in cultured MKs n=2; others conditions: n=3) are provided. (B-C) are pictures showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the perfusion stage from megakaryocyte suspension cultured (B) in the absence and (C) in the presence of latrunculin. For each panel, experimental conditions consist of platelet production during the 2 hours of perfusion stage from megakaryocyte suspension in the absence (+/+) or in the presence (+/+LA) of latrunculin during the perfusion.

[0273] FIG. 13 is a picture showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the perfusion stage from a megakaryocytes suspension cultured (A) in the absence and (B) in the presence of GM6001. For each panel, experimental conditions consist of platelet production during the 2 hours perfusion stage from megakaryocyte suspension in the absence (+/+) or in the presence (+/+GM6001) of GM6001 during the perfusion. Additional control consists of adding DMSO (+/+ DMSO) during the 2 hours perfusion stage.

[0274] FIG. 14 (A) is a graph representing the concentration of platelet produced after 2 hours of perfusion ([PLP].sub.end). Experimental conditions consist of platelet production from megakaryocyte suspension that was introduced in the microfluidic system at Day 12 of maturation (+/+D12) compared to megakaryocyte suspension that was introduced in the microfluidic perfusion at Day 10 of maturation, then back in culture and re-introduced in the microfluidic system at Day 12 (+/+D10+D12). Means±SEM (n=3) of platelet concentration are provided. (B) are pictures showing two colors flow cytometry analysis of platelets receptors, indicating the population of CD41.sup.+/CD42b.sup.+ platelets produced during the 2 experimental conditions described above (D12 compared to D10+D12).

[0275] FIG. 15 schematically represents one embodiment of the geometry of a platelet production with obstacles having a post shape. FIG. 10A is a top view and FIG. 10B is a cross-section view where r represents the radius of a post, p the closest distance between two post centers along the transversal axis, g the closest distance between two post centers along the longitudinal axis, α is the angle defined by the longitudinal direction of the channel and one of the lattice vectors of the primitive cell of the hexagonal periodic structure, H is the height of the microchannel and h refers to the height of the posts.

[0276] FIG. 16 represents two different gradient geometries used in the example: (A) “gradient geometry 1” and (B) “gradient geometry 2”. The numbers inside the device represent the distance p in the corresponding portion of the channel and x the distance from the beginning of the post area.

[0277] FIG. 17 is a stack of images illustrating the effect of the obstacle density on the megakaryocytes capture. The left panel represents a microfluidic device without obstacles density gradient (with p equals to 85 μm) whereas the right panel represents a microfluidic device with an obstacles density gradient (with p varying from 120 m to 70 m along the device for the example “gradient geometry 2”). Images from top to bottom are presented at 0, 2.2, 6.6, 8.8 and 13.2 cm from the beginning of the textured area. The images were recorded 35 min after the beginning of the experiment. Scale bar represents 500 μm.

[0278] FIG. 18 is a graph representing the effect of varying the obstacle density on the megakaryocyte capture for two geometry examples. Black triangles correspond to the density a of captured megakaryocytes in a channel with an increasing post density, while white squares correspond to the density a of captured megakaryocytes in a channel without obstacles density gradient. The dashed lines indicate the change in posts density. The texture corresponds to a geometry defined by r=15 μm, H=50 μm and h=50 μm. The device with constant post density is defined by p=g=85 μm. The upper panel compares this device to the so-called “gradient geometry 1” and defined by g=85 μm and p varying from 85 μm to 55 μm. The lower panel compares a constant p to the so-called “gradient geometry 2” defined by a varying p from 120 μm to 70 μm. The x axis represents the distance from the textured area entrance. The images were recorded 35 min after the beginning of the experiment.

EXAMPLES

[0279] Material and Methods

[0280] CD34.sup.+ Cells Culture and Differentiation

[0281] CD34.sup.+ cells were isolated from human umbilical cord blood (UCB) by an immunomagnetic technique (Miltenyi Biotec, Paris, France) as previously reported (see Poirault-Chassac et al, “Notch/Delta4 signaling inhibits human megakaryocytic terminal differentiation”, Blood, vol. 116, no 25, p. 5670-5678, 2010). These blood samples were obtained after informed consent and approval from our Institute Ethics Committee and in accordance with the Declaration of Helsinki. CD34.sup.+ cells were cultured in a humid atmosphere at 37° C. in 5% CO.sub.2 in complete medium consisting of Iscove modified Dulbecco medium (IMDM; Gibco Life Technologies, Saint-Aubin, France) supplemented with 15% BIT 9500 serum substitute (Stem Cells Technologies, Grenoble, France), α-monothioglycerol (Sigma-Aldrich, Saint-Quentin Fallavier, France) and liposomes (phosphatidyl-choline, cholesterol and oleic acid; SigmaAldrich). Human recombinant stem cell factor (SCF, 20 ng/mL; Miltenyi Biotec) and thrombopoietin peptide agonist AF13948 (TPO, 50 nM) (see Dunois-Lardé et al, “Exposure of human megakaryocytes to high shear rates accelerates platelet production”, Blood, vol. 114, no 9, p. 1875-1883, 2009) were added once at day 0 to the culture medium followed by addition of 20 nM TPO without SCF at day 5. Mature megakaryocytes are obtained after 12-14 days of culture. Removal of platelets formed during culture and immediately prior to shear exposure was performed by means of a BSA gradient according to the methods reported in (Robert A, Cortin V, Gamier A, Pineault N. Megakaryocyte and platelet production from human cord blood stem cells. Methods Mol Biol. 2012; 788: 219-47). The concentration was then adjusted to 200 000 megakaryocytes/mL. 10 μM ROCK inhibitor (Y27632, Sigma Aldrich, Saint Quentin Fallavier, France) may be added to the culture of megakaryocytes at day 8 of culture. Addition of 10 μM ROCK inhibitor was performed before introducing megakaryocytes suspension into the inlet chamber of the microfluidic device. 6.25 μM nicotinamide (N0636, Sigma Aldrich, Saint Quentin Fallavier, France) was added to the culture of megakaryocytes at day 8 of culture. Addition of 6.25 μM nicotinamide was performed before introducing megakaryocyte suspension into the inlet chamber of the microfluidic device. 20 μM blebbistatin (203390, Calbiochem) was added to the culture of megakaryocytes at day 8 of culture. Addition of 20 μM blebbistatin was performed before introducing megakaryocyte suspension into the inlet chamber of the microfluidic device. 10 μM latrunculin-A (sc202691, Santa Cruz Biotechnology) was added to the culture of megakaryocytes at day 8 of culture. Addition of 10 μM latrunculin-A was performed before introducing megakaryocyte suspension into the inlet chamber of the microfluidic device. 25 μM GM6001 (ab120845, Abcam) was added to the culture of megakaryocytes at day 8 of culture. Addition of 25 μM GM6001 was performed before introducing megakaryocyte suspension into the inlet chamber of the microfluidic device

[0282] System Architecture

[0283] A suspension of mature megakaryocytes is introduced in a 50 mL tube (Falcon tube, Dutscher, France) fixed on an orbital mixer (IKA MS3 basic), rotating at least at 300 rpm. The orbital mixer is used to maintain the homogeneity of the cell concentration in the suspension. The megakaryocytes concentration range in the tube is at least 100 mL.sup.−1 and cannot exceed 10.sup.12 mL.sup.−1.

[0284] Many methods can be used to control the flow through the different components: a differential pressure controller, a syringe pump and a peristaltic pump for example. When using a peristaltic pump (205U/CA, Watson Marlow, France), both inlet and outlet tubing arrive in the same rotating tube containing the megakaryocyte suspension. The said tube is connected to the inlet of the microfluidic chip with flexible tubing (Tygon ST R-3607, Idex Health and Science, Germany). The suspension is collected at the outlet. The peristaltic pump can be plugged upstream or downstream the microfluidic components.

[0285] Three microfluidic components can be implemented in series: a megakaryocyte sorter and/or a lateral cell mixer upstream, a platelet production channel, and a cell sorter downstream. Cell mixer and cell sorter are optional.

[0286] Devices Fabrication

[0287] Microfluidic components were made following a soft lithography rapid prototyping (Xia et al. 1998. “Soft lithography”. Annual Review of Materials Science. vol. 28, no 1, p. 153-184). First, transparencies were produced from a computer assisted design file containing the design of microchannels. These transparencies were used as masks in transferring the pattern into negative photo resist (SU-8 2000 and 3000 series, Microchem, US) by conventional photolithography, yielding a master with positive relief of micro channels. Both channels were made from molded polydimethylsiloxane (PDMS, Sylgard, Dow Corning, USA), sealed on glass slides. PDMS prepolymer and curing agent were mixed and degassed. The mixture was poured onto the master, cured for 2 h at 70° C., cut into individual chips, and inlet and outlet holes were punched. Glass slides were cleaned with isopropanol and dried. Both PDMS individual structures and glass slides were treated in an oxygen plasma oven and then sealed.

[0288] Platelet Production Channels

[0289] The textured surface is defined by 3D patterns on the channel walls (glass or PDMS). These patterns consist of an array of disks in the (Oxy) plane with a varying density of posts. Herein we present two examples of these possible variations.

[0290] The geometries of those patterns are described in FIG. 15 and FIG. 16. The design of the post array is defined by four parameters: the disk radius r, the distance p between two disk centers along the transverse axis, the distance g between two disk centers along the longitudinal axis and the angle α between the direction of the flow at the inlet and the direction defined by two pillar centers. In the geometries we used for the experiments herein, a was defined by the following criterion: sin α=r/g. This criterion geometrically implies that two neighbor disk centers, once projected on an axis normal to the initial flow direction, are separated by one radius. Experimentally, r varied from 15 μm to 20 μm, g varied from 85 μm to 120 m and p varied from 55 μm to 120 μm. FIG. 15 (B) shows the profile of an obstacle in the (Oxz) plane. The height of the channel and the height of the obstacle are defined respectively by H and h. The parameter h/H was experimentally equal to 1 (full pillars).

[0291] Two different gradients of p were used in this example, illustrated in FIG. 16 (A) and FIG. 16 (B): [0292] The first one, so-called “gradient geometry 1”, was used to increase the megakaryocyte capture in order to increase platelet production. The distance g is fixed at 85 μm and the distance p is varied from 85 μm to 55 μm along the channel. Therefore, the density of posts gradually increases along the channel. [0293] The second one, so-called “gradient geometry 2”, was used in order to homogenize cells density along the device thanks to a starting distance p=120 μm. Moreover, this distance p was varied along the channel from 120 μm to 70 μm. These variations induce a leap in post density and thus in the megakaryocyte capture. In order to allow enough space for megakaryocyte elongation, the distance g was defined as follows: [0294] For p≥85 μm, g=p [0295] For p<85 μm, g=85 μm

[0296] All the devices are composed of 16 parallel channels organized in a serpentine shape (FIG. 16). The channels were textured on the straight parts (13.2 cm long) and patterned with the parameters: r=15 μm, H=50 μm, H/h=1.

[0297] Shear Rates

[0298] We define a surface element on the channel wall, whatever on glass or PDMS (including PDMS obstacles). On this surface element, we define the unit vector of a plane by the vector acting normal to it, {circumflex over (n)}. A unit vector {circumflex over (m)}, tangential to the surface and in the local direction of the fluid velocity v, is determined so that ({circumflex over (n)}, {circumflex over (m)}) is a planar Cartesian coordinate system. The wall shear rate {dot over (γ)} (in s.sup.−1) is then defined by

[00001]$\stackrel{.}{\gamma }=\frac{\partial \hat{v}.\hat{m}}{\partial n}.$

The wall shear rate is controlled by both the flow rate in the device and by the geometry of the device.

[0299] Videomicroscopy System

[0300] The microfluidic chip was set on the stage of an inverted microscope (DMI6000 B, Leica Microsystems GmbH, Germany). A computer assisted motorized stage control was used to record positions along the channel length. We recorded observation field positions and alternated recording images between them along the experiment time. Differential interference contrast objective was used to record movies and images between 10× and 40×. A CMOS high-speed camera (Fastcam SA3, Photron, USA) was used to record images at frequencies from 0.5 to 1500 Hz.

[0301] Surface captured cells were counted manually from the recorded channel images, and cells in suspension were counted with a hematocytometer.

[0302] Protein Surface Treatments

[0303] Human von Willebrand factor (VWF) was a gift of Laboratoire Français du fractionnement et des Biotechnologies. It was diluted at 40 μg.Math.mL.sup.−1 in phosphate buffered saline phosphate buffered saline (PBS) without calcium and magnesium ions (Lonza, Belgium), and perfused in sealed microchannels. We used this surface coating only in the platelet production channel.

[0304] Bovine serum albumin (Sigma-Aldrich La Verpilleres, France) was diluted at 40 μg.Math.mL.sup.−1 in phosphate-buffered saline (PBS) and perfused in microchannels.

[0305] For both protein treatments, inlets and outlets of the chips were covered by cover slips. The chips were incubated overnight at 4° C. and washed with PBS before the experiment. VWF adsorption on both glass and PDMS was verified by fluorescence labeling with a primary polyclonal rabbit anti-vWF antibody (Dako, 10 μg.Math.mL.sup.−1) and with a secondary Alexa fluor 546 polyclonal goat anti-rabbit antibody.

[0306] Parallelization of Platelet Production Channels

[0307] A high megakaryocyte flow rate into the device is desired to increase the platelet production number. For a given pattern of obstacles and height of the channel, the flow rate can be increased by increasing the channel width. As mechanical constraints of the PDMS channels impose a maximum width over height ratio to avoid channel collapse, we parallelized channels to increase the effective width.

[0308] Megakaryocytes were introduced in the platelet production channel by means of tubing. Cells were distributed in the parallel channel through a triangular shaped entrance, which brings the cells to every channel. For a given channel height, the wall shear rate is inversely proportional to the channel width. Consequently, the wall shear rate is much higher close to the inlet walls than before the entrance of the parallelized channels. To avoid imposing a wall shear rate that could damage the cells, the distribution channel is fabricated using a higher height than the one used in the parallelized shear channels. The ratio of these two heights is typically between 2 and 20.

[0309] Platelet Sorting

[0310] Naked nuclei and intact megakaryocytes can be removed from the platelets by sorting in serial the outflow suspension from the platelet production channel. This can be done with microfluidic techniques. Apheresis techniques can also be considered for large volumes. We give two examples of platelet sorting using the pinched-flow fractionation technique (Takagi et al., Lab Chip, vol. 5, no 7, p. 778, 2005) and the Deterministic Lateral Displacement (L. R. Huang et al. Science, 304, 987, 2004). The Deterministic Lateral Displacement device is composed of one inlet, an array of posts of spherical shape and two outlets (central and lateral). The device is 5 cm long, 3 mm width and 40 μm in height. The post diameter is 85 μm and the spaces between posts are 15 μm. The post row shifting forms an angle of 0.05 radian. The surface is covered by a BSA coating. Cells are deflected and sorted in the central outlet whereas the platelets are not deflected and sorted on the lateral outlets. Another technique to separate platelets from a cells suspension is to use ultrasonic standing wave forces (Antfolk et al, Lab Chip, vol 14, 2791-2799, 2014). By applying an acoustic standing wave field onto a continuously flowing cell suspension, cells can be separated from the surrounding media depending on their physical properties. Then platelets can be sorted.

[0311] Characterization of Collected Platelets

[0312] Platelet production in the microfluidic device of geometry 2 was compared to control samples, consisting of platelet production from twenty-four hours static condition without the microfluidic chip. Expression of CD41 and CD42 antigens was characterized using a flow cytometer BD Fluorescence Accuri C6 ® (BD Biosciences, Le Pont de Claix, France). Platelets were incubated with fluorescein isothiocyanate (FITC)-conjugated anti-human CD41 (αIIb), R-phycoerythrin (PE)-conjugated anti-human CD42b (GPIbα) (both from Beckman Coulter, Villepinte, France) and FITC-conjugated anti-human activated αIIbb3 (BD Biosciences) during 15 minutes at 22° C., Controls were performed using FITC mouse IgG.sub.1 (Beckman Coulter), PE mouse IgG.sub.1 (Beckman Coulter). Single color flow cytometry analysis of platelet receptors was performed using the GP screen assay (Biocytex, Marseille, France). The number of antigenic sites is determined by converting the fluorescence intensity into corresponding numbers of monoclonal antibodies bound per platelet based on a calibrated bead standard curve. Fibrinogen adhesion assay and epifluorescence characterization were performed as reported in as previously reported in the above-cited publication of Dunois-Lardé, except that activation was obtained in the presence of an agonist peptide of the PAR-1 thrombin receptor (TRAP6). Epifluorescence was analyzed at 494 nm and 522 nm (absorption and emission, respectively), using a high-resolution bioimaging platform (QIClick-F-CLR-12 Digital CCD camera, Q-imaging, Microvisions Instruments, Evry, France).

[0313] Megakaryocyte Capture

[0314] We define captured megakaryocytes by surface adherent megakaryocytes, independently of their translocation velocity (including non-moving cells) and elongation process. We define elongated megakaryocytes by surface adherent and elongated megakaryocytes, which include megakaryocytes starting their reorganization in order to form their beads-on-a-thread structure as well as megakaryocytes already presenting the specific necklace structure. We evaluated the megakaryocyte capture of megakaryocyte suspensions introduced in the microfluidic device presenting system geometry 1 or 2 described in FIG. 15 and FIG. 16. Results are shown in FIGS. 17 and 18. Density of captured megakaryocytes is very high at the entrance of the textured area (σ˜300 to 500 mm.sup.−2) but this density decreases exponentially along the device at a fixed p (σ<50 mm.sup.−2 at the end of the textured area). In comparison, when cells experience a gradient of posts, the density of adherent megakaryocytes increases at each transition (corresponding to a decrease of the distance p). The density of cells is maintained and can be higher at the end of the textured area (˜100 mm.sup.−2 for “gradient geometry 1”).

[0315] For a specific megakaryocytes culture and megakaryocyte suspension concentration, we observed that megakaryocyte density was higher at the entrance and lower along the device for a fixed distance p=85 μm, phenomenon called the axial dependency. This density inequality is reduced when a higher distance p is used at the entrance (p=120 μm for “gradient geometry 2” in FIG. 17). The inventors found that increasing the post density along the channel partially compensates the effect of the axial dependency and ensure a more homogeneous megakaryocytes distribution along the device.

[0316] We evaluated the megakaryocytes capture according to the different megakaryocytes suspensions introduced in the microfluidic device. Results are shown in FIG. 1. Density of captured and elongated megakaryocytes has been measured and, in absence of ROCK inhibitor during the culture stage, we observed a sharp density difference between devices receiving megakaryocytes suspension supplemented or not with ROCK inhibitor. The presence of ROCK inhibitor in the suspension during the fluidic process seems to enhance the elongation process since elongated megakaryocytes density is 2.98±0.28 fold more important when ROCK inhibitor is added, as well as the megakaryocytes capture since captured megakaryocytes density is 1.42±0.13 fold more important when ROCK inhibitor is added. Similar effects are observed when ROCK inhibitor is added during the culture stage.

[0317] Effect of the Protein Surface Coating

[0318] On the vascular endothelial cells, VWF allows translocation of circulating platelets when subjected to high shear rates (>1000 s.sup.−1) through binding of their GPIb receptors (Huizinga et al, Science, vol. 297, no 5584, p. 1176-1179, 2002). In vitro, the adsorption of VWF allows megakaryocytes, platelets and proplatelets to translocate on the PDMS and glass surface (Dunois-Lardé et al, Blood, vol. 114, no 9, p. 1875-1883, 2009). We compared the effect of VWF and BSA coating the channel surface on the adhesion of megakaryocytes with ROCK inhibitor added at day 8 of the culture as well as during the two hours of the microfluidic perfusion.

[0319] We performed in parallel one experiment coated with VWF and one experiment coated with BSA. We compared the surface density of megakaryocytes at t=35 min. Results are shown in FIG. 2. When coated with VWF, elongated megakaryocytes density is 7.72±1.32 fold more important that when coated with BSA, while captured megakaryocytes density is 5.86±1 fold more important. VWF sharply increases the density of adherent and elongated cells in channels.

[0320] Megakaryocyte Ruptures and Platelet Release

[0321] We observed platelet shedding from surface adherent megakaryocytes. When megakaryocytes are translocating on the channel walls, they establish transient interactions with VWF on the wall surface, which progressively lead to morphologic changes until platelet shedding from megakaryocytes. Shedding occurs when both elongation and cell body are translocating. This process is described in the above-cited publication of Dunois-Lardé and in the international patent application WO 2010/06382311. After a rupture, both entities continue translocating on the wall surface.

[0322] Shedding also occurs when the cell body translocates until being trapped around or behind a pillar. On the time scale of several minutes, the megakaryocyte undergoes morphological changes leading to the formation of an elongation, which adopted a beads-on-a-thread structure, as previously reported in the above-cited publication of Dunois-Lardé. In addition, instead of full cell contacts with the coated surface, some elongations appeared to be freely moving (dangling) together with the flow, although the cells remained in the same position by at least one point of contact. As the size of the bead-on-a-thread elongation grows, some ruptures occur, releasing platelets and/or proplatelets, as shown in FIG. 3. No differences in the elongation process as well as the beads-on-a-thread structure formation have been observed when ROCK inhibitor is added in the megakaryocytes suspension. However, we observed a significant higher platelet release when ROCK inhibitor was introduced in the megakaryocytes suspension during the perfusion stage and absent during the culture stage. The hemocytometer counting of platelet production from different conditions was measured. Ratio of platelet production (experimental conditions vs twenty-four static without ROCK inhibitor platelet production) was calculated and is shown in FIG. 4. Platelets production is significantly higher with presence of shear stress through the microfluidic device and even higher when ROCK inhibitor is added in megakaryocytes suspension for the two hours of the microfluidic perfusion. Produced platelets in the fluidic device that received a megakaryocytes suspension with ROCK inhibitor displayed platelets production that was significantly more important in comparison to the fluidic device that received a megakaryocytes suspension without ROCK inhibitor. A similar tendency is observed when ROCK inhibitor is added during the culture stage.

[0323] Other megakaryocyte modulator compounds have been tested for their effect on the quantity and quality of the produced platelets, when specifically added at the perfusion stage in an ex vivo method for producing platelet according to the present invention.

[0324] Nicotinamide, an inhibitor of the sirtuin pathway, was introduced in the megakaryocyte suspension during the 2 hours of perfusion in the microfluidic device. No differences in the elongation process as well as the beads-on-a-thread structure formation have been observed. However, we observed higher platelet release when nicotinamide was introduced in the megakaryocyte suspension during the perfusion stage and absent during the culture stage (FIG. 10 (A)). The hemocytometer counting of platelet production from different conditions was measured. Produced platelets in the fluidic device that received a megakaryocyte suspension with nicotinamide displayed platelet production that was more important in comparison to the fluidic device that received a megakaryocyte suspension without nicotinamide. A similar tendency is observed when nicotinamide is also added during the culture stage.

[0325] Similar results were observed when blebbistatin (FIG. 11) or latrunculin (FIG. 12) were introduced in the megakaryocyte suspension during the perfusion stage and without or with previous introduction during the culture stage.

[0326] FIG. 14 shows the effect of the shear stress, at an earlier day than the platelet production day, on the final platelet production. In other words, megakaryocyte suspension was introduced in the VWF-coated microfluidic device at day 10 of culture. After 2 hours of circulation in this device, remaining megakaryocytes, meaning non-captured megakaryocytes, were back in culture for two additional days. Then at day 12, they were re-introduced in a fresh VWF-coated microfluidic device and platelet production was compared with production from megakaryocyte suspension introduced once in the VWF-coated microfluidic device, at day 12. As shown in FIG. 14 (A), higher platelet release was quantified when megakaryocyte suspension had already circulated into a VWF-coated microfluidic device. This production also showed a higher proportion of CD41.sup.+/CD42.sup.+ platelets (FIG. 14 (B)).

[0327] Characterization of Platelet Produced in the Fluidic Device

[0328] After two hours of perfusion, produced platelets in the fluidic device that received a megakaryocytes suspension with ROCK inhibitor displayed several characteristics that were better in comparison to those of platelets produced in the fluidic device that received a megakaryocytes suspension without ROCK inhibitor. As shown on FIG. 5, proportion of platelets expressing CD42b receptor was higher when produced by the fluidic device that received a megakaryocytes suspension with ROCK inhibitor. They also showed similar characteristics that were comparable to those of natural platelets, i.e. platelets isolated from blood (FIG. 6). Interestingly, in FIG. 7, upon TRAP-6 stimulation, platelets from megakaryocytes suspension with ROCK inhibitor show a tubulin conformation typical of a transient activation, whereas actin filaments are reorganized as thrombin-activated platelets isolated from blood. They seem to be able to go back in a resting conformation, which is not observed in samples from the fluidic control without addition of the ROCK inhibitor. Without any stimulation, FIG. 7 shows that the tubulin ring characteristic of platelets is present in the samples collected from the fluidic device but not from the static samples. In fact, static samples show presence of filopodia and lamellipodia in TRAP-6 presence and absence. Upon TRAP-6 stimulation, platelets produced in microfluidic device were able to undergo activation features similar to those known to characterize blood platelets, such as to increase their levels of binding of the PAC1 monoclonal antibody specific of the activated conformation of the αIIbβ3 receptor. According to FIGS. 8 and 9, platelets produced from megakaryocytes suspension with or without ROCK inhibitor both show an activated conformation of the αIIbβ3 receptors upon TRAP-6 stimulation.

[0329] Other megakaryocyte modulator compounds have been tested such as nicotinamide (FIG. 10 (B-C)), blebbistatin (FIG. 11 (B-C)) or latrunculin (FIG. 12 (B-C)), and showed a higher proportion of CD41.sup.+/CD42b.sup.+ platelet when produced by the fluidic device that received a megakaryocyte suspension with nicotinamide, blebbistatin or latrunculin during the 2 hours of perfusion, without or with previous introduction during the culture stage. Similar tendency was observed for two additional experiments for each tested compound (data not shown).

[0330] A metalloproteinase inhibitor, GM6001, has been tested for its effect on improving the final yield of CD42b.sup.+ platelet. CD42b receptor has been shown to be unstable on the surface of platelets produced ex vivo. It may be due to the cleavage by metalloproteinase. Addition of the broad-range metalloproteinase inhibitor GM6001 during the culture as well as during the 2 hours perfusion stage increased the proportion of platelet CD42b.sup.+. As shown in FIG. 13, proportion of platelets expressing CD42b receptor was higher when produced in the fluidic device that received a megakaryocyte suspension supplemented with GM6001 for 2 hours. A similar effect was detected when the megakaryocyte suspension was cultured, at the later stage, with GM6001. Similar tendency was observed for two additional experiments (data not shown).