In vitro expansion of erythroid cells

Abstract

The present invention relates to a method for in vitro expansion of erythroid cells. The method includes subjecting erythroid cells to 3-dimensional packed cell culture using a porous structure. The use of the composition according to the present invention enables in vitro expansion of erythroid cells in the most efficient manner.

Claims

1. A method for in vitro expansion of erythroid cells comprising subjecting erythroid cells to 3-dimensional packed cell culture with a porous structure, wherein the culture is performed in each pore of the porous structure, and wherein the erythroid cells are mixed and co-cultured with cells selected from the group consisting of mesenchymal stem cells, endothelial cells, monocytes, macrophages, and histiocytes during the 3-dimensional packed cell culture.

2. The method according to claim 1, wherein the pores in the porous structure have a size distribution of 30 to 500 m.

3. The method according to claim 1, wherein the culture is performed in a medium to which shear stress is applied by a continuous flow.

4. The method according to claim 3, wherein the flow is created by stirring.

5. The method according to claim 4, wherein the flow is created by stirring at 1 to 50 rpm.

6. The method according to claim 3, wherein the culture is performed in a filter in the medium to prevent erythrocytes from escaping from the porous structure during the continuous flow.

7. The method according to claim 6, wherein the filter has a mesh size of 1 to 8 m.

8. The method according to claim 1, wherein the erythroid cells are cells that enter the terminal maturation stage.

9. The method according to claim 1, wherein the erythroid cells express at least one adhesion-related gene selected from deleted in liver cancer 1 (DLC 1), intercellular adhesion molecule-4 (ICAM-4), and very late antigen-4 (VLA-4).

10. The method according to claim 1, wherein the erythroid cells are mixed and co-cultured with mesenchymal stem cells, endothelial cells, monocytes, macrophages, and histiocytes during the 3-dimensional packed cell culture.

11. The method according to claim 1, wherein the ratio of the number of cells selected from the group consisting of mesenchymal stem cells, endothelial cells, monocytes, macrophages, and histiocytes to the number of the erythroid cells is from 1:10 to 2:1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows 2D plate culture, 2D high density culture, packed cell culture of erythroid cells using macroporous microcarrier type Cytoline 1, packed cell culture of erythroid cells using Cytopore, packed cell culture of erythroid cells using Biomatrix 3D, packed cell culture of erythroid cells using Honeycomb, and spinner flasks, rotators, and a cylindrical spin filter for medium circulation conditions.

(2) FIG. 2 shows (A) images (200) observed after 2D plate culture and 3D packed cell culture of polychromatophilic erythroblasts/orthochromatic erythroblasts for 1 and 2 days and Wright-Giemsa staining: the asterisks indicate enucleated erythroid cells at the terminal maturation stage and the arrows indicate orthochromatic erythroblasts; and (B) compares the proportions of cells at different maturation stages and the yields of erythrocytes under the conditions shown in (A).

(3) FIG. 3 shows the results obtained after 2D plate of orthochromatic erythroblasts and 3D high density packed cell culture of orthochromatic erythroblasts at different densities for 1 and 2 days: (A) shows the results (200) of Wright-Giemsa staining on orthochromatic erythroblasts, the red asterisks, blue arrows, black arrows, white asterisks, and block asterisks indicate enucleated erythroid cells at the terminal maturation stage, orthochromatic erythroblasts, myeloid cells, cells undergoing enucleation, and cells with dysplasia, respectively; and (B) compares the proportions of cells at different maturation stages and the yields of erythrocytes under the conditions shown in (A).

(4) FIG. 4 shows (a) SEM images of Cytopore and Cytoline 1 as porous microcarriers (left), cell cultures (middle) in the microcarriers, and cell cultures in tubes (right) (the diameter of the upper wider portion was 6.5 mm; and (b) a cylindrical spin filter (left) (diameter 6.5 mm, height 15 mm) and erythroid cells grown at a density of 10.sup.7 cells/ml medium in a 12-well plate.

(5) FIG. 5 shows viable erythroid cells on days 1, 2, and 3 of culture in Cytopore.

(6) FIG. 6 shows viable cells after packed cell culture in Cytoline 1.

(7) FIG. 7 shows the proportions of cells at different maturation stages after culture in Cytoline 1 for 1 and 2 days.

(8) FIGS. 8 and 9 show the results observed after packing culture of cells of Case 2 derived from another cord blood for 1 day. Specifically, FIG. 9 shows the proportions of orthochromatic erythroblasts in a control and experimental group II were 39% and 94.1%, respectively, revealing markedly increased cell maturation. Upon 3D packed cell culture in the presence of Cytoline 1, intercellular contact was increased and cell maturation was effective.

(9) FIG. 10 shows (A) images obtained after erythroblasts was allowed to settle in a spinner flask and cultured without a microcarrier while rotating at RPM 25 and images obtained after 2D plate culture of erythroblasts; and shows (B) the proportions of cells at different maturation stages with increasing culture time.

(10) FIG. 11 shows the results obtained after culture of cells using Cytopore in a spinner flask.

(11) FIG. 12 compares the proportions of cells at different maturation stages after 3D culture using Cytoline 1 in a spinner flask.

(12) FIG. 13 shows images (200) observed after culture of cells using Cytoline 1 and a filter under different flow conditions at RPM 0 and RPM 25 and Wright-Giemsa staining, the red asterisks, blue arrows, black asterisks, and black arrows indicate enucleated erythroid cells, orthochromatic erythroblasts, dead cells, and polychromatic erythroblasts, respectively.

(13) FIG. 14 shows the proportions of cells at different maturation stages under the conditions shown in FIG. 13.

(14) FIG. 15 compares the yields of erythrocytes from cells of a cord blood derived case different from that shown in FIG. 13.

(15) FIGS. 16 to 18 show the results observed after 2D high density culture and 3D high density packed cell culture of erythroid cells. Specifically, FIG. 16 shows images (200) observed after Wright-Giemsa staining of cells. The red asterisks, blue asterisks, and black arrows indicate enucleated erythroid cells at the terminal maturation stage, dead cells, and immature cells, respectively, and the remaining cells are orthochromatic erythroblasts. FIG. 17 shows the proportions of cells at different maturation stages under the conditions shown in FIG. 16. FIG. 18 shows the results obtained after culture of erythroid cells using porous structures made of different materials and having different pore sizes: the left images show pellets of mature erythroid cells collected after culture in the porous structures and the right images show cell viabilities.

(16) FIG. 19 shows images (200) observed after 2D high density and 3D packed cell culture of erythroid cells and Wright-Giemsa staining: the red asterisks indicate enucleated erythrocytes; scale bar represents 50 m.

(17) FIG. 20 shows graphs comparing the proportions of erythroid cells at different maturation stages after culture on different porous structures.

(18) FIG. 21 shows the results obtained by qPCR of erythrocyte production markers (ICAM-4, GATA-I, Hb-beta, and Hb-gamma) and adhesion-related signals (DLC-1 and ICAM-4) after 3D packed cell culture for 3 days.

(19) FIG. 22 is a graph showing that in vitro produced erythrocytes were functionally similar to the peripheral blood of a healthy donor.

(20) FIG. 23 shows various structures available for 3-dimensional packed cell culture of erythroid cells.

(21) FIG. 24 shows the distributions of cells obtained after adherent cells (MSCs) and osteoblasts were placed in porous structures, allowed to stand for 3 h until cells were adhered to the porous structures, and added with erythroid cells (upper row 3 h; experimental group A), and the distributions of cells obtained after the three types of cells were added simultaneously and co-cultured for 192 h (lower row mix; experimental group B).

(22) FIG. 25 shows the experimental results obtained after co-culture of myeloid cells including macrophages and monocytes with erythroid cells in 3D porous structures for 72 h. The co-culture induced steric interactions in the porous structures to affect the maturation of erythroid cells and the production of erythroid cells at the terminal maturation stage. The left microscopy image shows the internal morphology of the structure after culture for 48 h and the right images show the stained cells after culture 48 and 72 h (the black arrows, red arrows, and white asterisks indicate immature erythroid cells, mature erythroid cells, and macrophages, respectively, and the remaining cells are monocytes).

DETAILED DESCRIPTION FOR CARRYING OUT THE INVENTION

(23) The present invention will be explained in more detail with reference to the following examples. However, it will be obvious to those skilled in the art that these examples are provided for illustrative purposes only and are not to be construed as limiting the scope of the invention.

EXAMPLES

(24) Experimental Methods

Example 1

Cell Culture and Enumeration

(25) CD34+ cells were isolated from the cord blood of healthy donors by using the immunomagnetic microbead selection method and EasySep CD34 isolation kit (StemCell Technologies). Some of the CD34+ cells were used for culture immediately after isolation or were frozen and stored until use. The cells were cultured under stroma- and serum-free conditions for 17 days [3, 4].

(26) Several cytokines were added to induce the CD34+ cells to differentiate and proliferate. From day 0 to day 7, the cells were cultured in media supplemented with 10 M hydrocortisone (Sigma), 100 ng/ml stem cell factor (SCF; R&D Systems), 10 ng/ml interleukin (IL)-3 (R&D Systems), and 6 IU/ml erythropoietin (EPO; Calbiochem). From day 7 to day 13, basophilic erythroblasts were cultured in media supplemented with 50 ng/ml SCF, 10 ng/ml IL-3, and 3 IU/ml EPO. From day 13 to day 17, 2 IU/ml EPO was added [3, 4]. On day 17, when the proportions of polychromatic erythroblasts and orthochromatic erythroblasts reached 50%, 3D culture experiments were conducted for comparison. On day 17 of culture, 2 IU/ml EPO and 5% cord blood plasma derived serum were added for 3D culture [5].

(27) The culture media were replenished every other day. The cells were cultured in a CO.sub.2 incubator (Sanyo) at 37 C. and 5% CO.sub.2. The numbers of variable cells were calculated by trypan blue staining. Cell integrity and maturation stage were analyzed by counting at least 400 cells per group in a blinded manner. The analysis was performed in a blinded manner by two investigators who were unaware of the cell culture conditions.

Example 2

Imaging of Erythroid Cells

(28) Cells were plated on a slide using a cell centrifuge (Cellspin, Hanil Science Industrial) and subsequent Wright-Giemsa staining (Sigma-Aldrich). The stained cells were analyzed for maturation stage, myelodysplasia, and cell integrity.

Example 3

2D Plate Culture and 3D Packed Cell Culture

(29) On days 15-17 of culture, the erythroid cells (polychromatic/orthochromatic erythroblasts) at the terminal maturation stage were divided and cultured under the conditions shown in FIG. 1. The control cells were cultured at a density of 110.sup.6 cells/mL in a 6-well 2D plate. In the experimental groups, the cells were allowed to be packed at a density of 210.sup.7 cells/mL in tubes and cultured in a state in which the tubes stood upright. Basal media used in the control and the experimental groups were supplemented with 5% plasma-derived serum [5]. After culture for 24 and 48 h, the cell numbers and viabilities were determined and the maturation stages of cells were confirmed using Wright-Giemsa staining.

Example 4

Determination of Stage for 3D Culture

(30) In this example, an optimum maturation stage for packed cell culture of erythroblasts was determined. Packed cell culture started when 50% or more of the basophilic erythroblasts remained (on day 13 of culture) and the proportion of polychromatic/orthochromatic erythroblasts was 50% or more.

Example 5

2D Plate Culture and 3D Packed Cell Culture (100 l, 200 l)

(31) As the cell number increases, it is difficult for nutrients to diffuse into cell packed portions or for toxic metabolites, such as CO.sub.2 and lactate, to exit to the outside. In this example, an optimum packing scale upon packed cell culture of polychromatic/orthochromatic erythroblasts was confirmed. To this end, the yields and enucleation rates of erythrocytes under the respective conditions were compared. The control cells were cultured at a density of 110.sup.6 cells/mL in a 6-well plate. In experimental group I, the cells were allowed to settle and packed at a density of 210.sup.7 cells/mL in a conical tube and cultured. In experimental group II, the cells were cultured at a density of 210.sup.7 cells/2 ml. The cell packing scale of experimental group II was twice in volume as large as that of experimental group I. The tubes had a size such that the cells of experimental groups I and II were sufficiently packed to create a 3D packing environment. The same culture solutions were used. After culture for 24 and 48 h, the cell numbers and viabilities were determined and the maturation stages of cells were confirmed using Wright-Giemsa staining.

Example 6

2D Culture and 3D Packed Cell Culture in Porous Structures

(32) In this example, the packing scale of polychromatic/orthochromatic erythroblasts was increased to reproduce a 3-dimensional environment of the bony trabeculae. To this end, materials that are applicable to humans and have dimensions similar to the bony trabeculae space were selected and compared. The first material was Cytopore (GE healthcare) made of 100% cellulose, which was a porous carrier having an average pore size of 30 m, a density of 1.03 g/ml, a particle diameter of 200-280 m, an effective surface area of 1.1 m.sup.2/g, and a volume of 40 ml/g. The second material was Cytoline (GE healthcare) made of polyethylene and silica and having a pore size of 1-400 m, a density of 1.32 g/ml, a settling velocity of 120-220 cm/min, a length of 1.7-2.5 cm, a thickness of 0.4-1.1 cm, a surface area of 0.3 m.sup.2/g or more, which were similar to those of the bony trabeculae space.

(33) A control for the 3D experimental groups was the same as that for the 2D plate culture conditions. Cytopore was placed in a tube (experimental group I) and Cytoline 1 was placed in a tube (experimental group II) (see FIG. 4), polychromatic/orthochromatic erythroblasts were added thereto, and packed and cultured at a density of 110.sup.7 cells/mi. Culture was continued until the terminal maturation stage was reached. After culture for 24 h and 48 h, the cell numbers and viabilities were determined and the maturation stages of cells were confirmed using Wright-Giemsa staining.

Example 7

3D Packing in Spinner Flasks

(34) In this example, it was confirmed whether cells at the maturation stage were resistant to shear stress caused by a weak flow and were squashed in a 3D packing situation. Control cells were tested by the same procedure as above. For the experimental groups, cells were added at a density of 1.010.sup.7 cells/ml to Cytoline 1 in spinner flasks and cultured. Generally, cells are immobilized in a packed bed bioreactor. In this example, erythroblasts, which are similar to adherent cells rather than to floating cells, are placed in the pores but some of them escape from the microcarriers by a media flow and float outside the microcarriers.

(35) The fact that erythroblasts are very susceptible to shear stress caused by a flow of fluid was confirmed at various RPM conditions. Thus, after the rotating velocity of a bioreactor was adjusted to RPM 25 to minimize the influence of shear stress, cells were cultured until the terminal maturation stage was reached. RPM 25 is the lowest rotating velocity of the bioreactor. The numbers of viable cells in the control and the experimental groups were calculated by trypan blue staining and the maturation stages of cells were analyzed using Wright-Giemsa staining every 24 h.

Example 8

3D Packing Using Cytopore and Cytoline 1 in Spinner Flasks

(36) For scale-up culture considering the bone marrow situation, conditions for optimum medium exchange without cell squash despite packing of large volume and quantity of cells were investigated. To this end, Cytopore or Cytoline 1 was added to a spinner flask and cultured. The same conditions were used for a control. Experimental groups were cultured at a density of 110.sup.7 cells/ml in the spinner flasks. The maturation stages of cells were analyzed using Wright-Giemsa staining every 24 h.

Example 9

3D Culture Using Cytoline 1 in Spin Filter in Spinner Flask (Medium Exchange)

(37) Cytoline 1 was selected as a porous structure for 3D packed cell culture because it was more suitable for the reproduction of an in vitro bone marrow environment. It was found that non-adherent erythroid cells are not immobilized in the porous microcarrier and some of them escape from the carrier and float in the form of suspension. In this case, erythroid cells do not come into contact with each other. In this example, erythroid cells were packed and grown in the pores and they, together with the carrier, were trapped in a cylindrical spin filter for easy medium exchange and immobilized and cultured in an incubator. Generally, erythroid cells belonging to the smallest cell types easily escape through an 8 m filter from a bioreactor due to the large pore diameter of the filter. Thus, in this example, erythroid cells were trapped in a small spin filter whose mesh size was 3 m and were cultured while allowing for slow exchange of culture solution. The same control was used as above. In experimental group I, cells were trapped and cultured at a density of 110.sup.7 cells/ml in Cytoline 1 in a stationary state (at RPM 0). In experimental group II, cells were cultured while shaking at RPM 25. The maturation stages of the cells of the control and the experimental groups were analyzed every 24 h.

Example 10

2D High Density Culture and 3D Packed Cell Culture Using Cytoline 1

(38) From the above experimental results, it was concluded that 3D packed cell culture was more advantageous in cell viability and cell maturation than 20 plate culture. The use of the porous structure lowered the cell viability for 3D packed cell culture but led to a marked increase in cell maturation. These results can be explained by the reproduction of an in vitro bone marrow environment and can be considered innovative for the production of erythroid cells at the terminal maturation stage at a high density.

(39) The present inventors have reported previously that when cells are cultured at a higher density than a general cell density such that they come into contact with each other on the bottom of a 2D plate, high cell viability and better cell maturation are obtained. The present inventors have conducted comparative experiments whether 3D culture produced a larger quantity of erythrocytes than 2D high density plate culture due to the higher frequency of contact and packing interfered with medium exchange. The same culture solution was used. Control cells were cultured at a density of 110.sup.6 cells/mL in a 24-well 2D plate. In experimental group 1, cells were cultured at a density of 110.sup.7 cells/mL with Cytoline 1 in the tube. In experimental group II, cells were cultured at a density of 110.sup.7 cells/ml with Cytoline 1 in the spin filter. The numbers of viable cells in the control and the experimental groups were determined by trypan blue staining and the maturation stages of cells were analyzed using Wright-Giemsa staining every 24 h.

Example 11

Additional Experiment for Selection of Suitable Porous Structure

(40) Erythroid cells at the terminal maturation stage were subjected to packed cell culture at high densities on macroporous microcarriers and scaffolds as 3D porous structures to confirm the effect of the 3D porous structures on the culture of the erythroid cells. 3D porous structures should have a number of pores to stimulate the interactions of cells, reproduce the porosity of the in vivo bony trabeculae, and made of biocompatible materials without damage to cells. To screen suitable porous structures, various pore sizes, skeletal shapes, and constituent components were applied as shown in Table 1. Control cells were cultured at a density of 110.sup.6 cells/mL on a 6-well 2D plate.

(41) TABLE-US-00001 TABLE 1 Pore size Density Type Material (m) Diameter (mm) (g/ml) Macroporous Cytopore 100% 30 0.2-0.28 1.03 microcarriers cellulose matrix Cytoline 1 Polyethylene, 10-400 1.7-2.5 1.3 silica Scaffolds Honeycomb 100% bovine 200-400 6 (diameter) collagen 2 (height) Biomerix 3D Biostable 250-500 5 (diameter) polyurethane 2 (height)

(42) Erythroid cells at the maturation stage were cultured at a density of 110.sup.7 cells/mL in each porous structure in a spin filter for 3 days. Half of the whole medium was replaced with the fresh one daily and the cell morphology was confirmed by Wright-Giemsa staining.

Example 12

Measurement of the Ability of Cultured Cells to Transport Oxygen

(43) Oxygen equilibrium curves of erythrocytes produced by packed cell culture in 3D scaffolds were measured using a Hemox analyzer (TCS Scientific) and were compared with those of peripheral blood of a healthy subject as a control.

Example 13

Additional Experiment for Selection of Structure for 3D Culture of Erythroid Cells

(44) In this example, a pore size and a material suitable for culture of erythroid cells were determined. To this end, cell packing experiments were conducted using various 3D structures. To select a material having a pore size of 500 m or less for reproducing an intracellular skeletal structure and sufficient strength to support the constituents of bone marrow and a structure capable of easily collecting cells after culture, experiments were conducted on a 3D-printed scaffold, a microcarrier, a non-porous disk, and metal materials (Ni and stainless steel). The cell densities of a control and experimental groups and the conditions of culture media were the same as the above experimental conditions. Table 2 shows the specifications of the 3D structures.

(45) TABLE-US-00002 TABLE 2 Pore Diameter size (diameter Type Component (m) height) 3D printed 3D printed Polycaprolactone 500 8 mm 2 mm.sup. scaffold scaffold Microcarrier Fibracel- Polyester mesh 6 mm 0.1 mm disk microcarrier having disk polypropylene support Metal Nickel Nickel ~500 8 mm 1.5 mm structures structure Stainless Metal alloy ~500 8 mm 1.5 mm steel structure

Example 14

Identification of Effect of Co-culture with Constituents of Bone Marrow

(46) Actual erythroid islands in bone marrow were reproduced by co-culture of constituents of bone marrow and erythroid cells in 3D structures that mimic the bony trabeculae space where the hematopoietic process occurs actively. The maturation of cells are promoted in the erythroid islands. Mesenchymal stem cells (MSCs), osteoblasts, and mature erythroid cells were co-cultured for 192 h while maintaining their ratio constant (2:1:10). The results were compared to those obtained in 2D plate culture. A mixture of stem line II and Dulbecco Modified Eagle Medium low glucose media in a ratio of 9:1 was used as a basal medium suitable for co-culture of the three kinds of cells. Only a portion of the whole medium was replaced with the fresh one daily. Thereafter, populations of cells (CD45 for myeloid cells, CD71 for erythroid cells, and CD51 and CD90 for MSCs) were confirmed by flow cytometry. Various experiments were conducted to establish optimum culture conditions. To this end, the constituent cells were cultured simultaneously. Alternatively, MSCs and osteoblasts were first cultured and were co-cultured with erythroid cells 3 h later.

(47) Experimental Results

(48) 1. Packing Effect of Basophilic/polychromatic Erythroblasts (2D Plate Culture and 3D Packed Cell Culture)

(49) Hematopoietic stem cells isolated from cord blood were differentiated in a culture medium for 13-17 days. The cells were packed at a high density when the proportions of basophilic/polychromatic erythroblasts reached 50%. Control cells were cultured in a plate. In the experimental group, cells were packed at a density of 210.sup.7 cells/ml in tubes. On days 1 and 2 of culture, the maturation stage and viability of erythroid cells were observed (see FIG. 2A). Unlike in 2D plate culture, erythroid cells were cultured while keeping their intimate contact with each other in 3D tubes, similarly to erythroid islands in bone marrow (see FIG. 2A). After days 1 and 2 of culture, the proportions of orthochromatic erythroblasts in the control were 38.5% and 57.8%, respectively. In contrast, the proportions of orthochromatic erythroblasts in the experimental group having undergone 3D packing were 50.9% and 49.6% (see FIG. 2B). the numbers of enucleated erythrocytes (20.6% and 25.2%) in the experimental group were increased after 3D packed cell culture than in the control (13.7% and 7.5%). On day 2 of culture, the proportions of orthochromatic erythroblasts in the experimental group were smaller than those in the control, which appears to be because more orthochromatic erythroblasts were enucleated into erythrocytes in the experimental group. Myelodysplasia were decreased in the control (5.1% and 6.1%) and the experimental group (2.3% and 3.1%) on days 1 and 2 of culture. Myelodysplasia is a disorder that occurs when nuclei are not sufficiently divided and are split or several nuclei remain in one cytoplasm because of a bad culture environment. Therefore, it can be concluded that the 3D packed cell culture environment is more suitable for cell maturation and erythrocyte production.

(50) 2. Packing Scale of Orthochromatic Erythroblasts

(51) In the above experiments, 3D high density packing cell culture was confirmed to be more effective in cell maturation than 2D plate cell culture because it reproduced the in vivo bone marrow environment where intercellular contact increases. To find an optimum cell packing scale for 3D high density packed cell culture, erythroblasts at the maturation stage in experimental groups I and II were compared with those in the 2D plate culture control. In experimental group I, cells were subjected to packed cell culture at a density of 110.sup.7 cells/ml in a narrow tube where a 3D packing environment could be created. In experimental group II, cells were subjected to packed cell culture at a density of 210.sup.7 cells/ml. After days 1 and 2 of culture, the proportions of orthochromatic erythroblasts were 40.7% and 22.6% for the control, 43.3% and 23.1% for experimental group I, and 36.0% and 18.2% for experimental group II, respectively. These results demonstrate that the conditions of experimental group I are more effective for cell maturation upon 3D packed cell culture (see (A) and (B) of FIG. 3). Due to the enucleation effect, the proportions of erythroid cells at the terminal maturation stage were higher in experimental group I (23.8% and 46.1%) and experimental group II (23.0%, 30.1%) than in the control (16.1% and 22.6%) on days 1 and 2 of culture. These results are explained by an increase in the number of enucleated erythrocytes upon packing in experimental groups I and II. The packing effect was observed to be more significant in experimental group I than in experimental group II. On day 2, myelodysplasia was decreased to 3.8%, 2.7%, and 2.6% upon packing for the control and experimental groups I and II, respectively. On day 1, cell viability was increased by 4.6% and 5.4% in experimental groups I and II, respectively, compared to in the control. On day 2, cell viability was increased by 45.7% and 5.9% in experimental groups I and II, respectively, compared to in the control (see FIG. 3B).

(52) 3. 3D Packed Cell Culture in Porous Structures

(53) Cells at the maturation stage were cultured in an in vitro bony trabeculae environment. To this end, erythroblasts at the maturation stage cultured in the control were compared with those cultured in Cytopore and Cytoline I (see FIG. 4).

(54) On days 1, 2, and 3 of culture, the viabilities of erythroid cells in Cytopore were lower than those in the control. This is thought to be because the spaces of the pores were too small to induce cell contact and medium circulation, as shown in FIG. 4 (see FIG. 5).

(55) For Cytoline 1, cell maturation was efficient upon packed cell culture (see FIG. 6). As a result, the proportions of orthochromatic erythroblasts in the control were 38.3% and 15.6% and those of Case 1 of Cytoline 1 were 67.5% and 14.1% after days 1 and 2 of culture, respectively. The proportions of enucleated erythrocytes in Cases 1 and 2 of Cytoline 1 were much higher (26.9% and 80.0%) upon packing than those in the control (6.1% and 7.2%). Myelodysplasia was decreased in the control (26.9% and 4.0%) and Cytoline 1 (2.8% and 0.0%) (see FIG. 7). Cells of Case 2 derived from another cord blood was compared after packing for 1 day. The proportions of orthochromatic erythroblasts were 39% for the control and 94.1% for experimental group II. The higher proportion of orthochromatic erythroblasts in experimental group II indicates greatly increased cell maturation. Upon 3D packed cell culture in the presence of Cytoline 1, intercellular contact was increased and cells were matured more effectively (see FIGS. 8 and 9). Since live cells in the control were not erythroid cells but cells of other families, the viability of cells in the control were not compared.

(56) 4. Comparison of Culture of Erythroblasts in Spinner Flasks without Pores

(57) In this example, it was confirmed whether polychromatic/orthochromatic erythroblasts at the maturation stage were resistant to shear stress (hydromechanical damage) caused by a weak flow in spinner flasks (see FIG. 10A). Cells were cultured with stirring at RPM 25, where shear stress was minimized, for 4 days. At this rate, cells were settled down at the bottom without floating. The proportions of orthochromatic erythroblasts in the control (9.1%, 5.9%, 46.2%, and 41.6%) and in experimental group I (35.0%, 12.8%, 36.0%, and 12.8%) were observed every 24 h (see FIG. 10B). The yields of erythrocytes were higher in experimental group I using a spinner flask (6.5%, 9.0%, 26.0%, and 59.3%) than in the control (0.0%, 2.7%, 5.8%, and 3.4%), Therefore, culture in the presence of Cytoline 1 in the spinner flask was also effective in cell maturation and erythrocyte production. However, the cell viability of orthochromatic erythroblasts was lower and the morphology of erythrocytes was not well maintained by shear stress in experimental group I compared to in the control. Therefore, pores protected from shear stress are suitable for culture (see FIG. 10A).

(58) 5. Culture in the Presence of Cytopore in Spinner Flasks

(59) To reduce cell squash and shear stress in large volume, Cytopore and Cytoline 1 were added and cultured in separate spinner flasks for 3D packed cell culture. For Cytopore, the number of mature erythroid cells did not increase even in the spinner flask, as in the results obtained in the previous experiment. Therefore, Cytopore was confirmed to be unsuitable for culture of erythroid cells (see FIG. 11).

(60) 6. Culture in the Presence of Cytoline 1 in Spinner Flask

(61) The proportions of orthochromatic erythroblasts were 11.8%, 3.9%, and 34.7% in the control and 34.1%, 9.7%, and 3.1% in the experimental group. When Cytoline 1 was added and cultured in a spinner flask for 3D packed cell culture, intercellular contact was increased and cells were matured more effectively (see FIG. 12). Due to the enucleation effect, the proportions of enucleated erythrocytes were higher in the experimental group (12.2%, 32.3%, and 90.6%) than in the control (0%, 2.3% and 4.2%) on days 1, 2, and 3 of culture. Therefore, it can be concluded that since cells at the maturation stage in Cytoline 1 are less susceptible to sheer flow while maintaining their packing effect, the presence of Cytoline 1 is effective for the production of erythroid cells at the terminal stage.

(62) 7. Comparison of Effect of Spin Filter and Cytoline 1 in Spinner Flask and Shaking RPM on 3D Culture

(63) To prevent cells from escaping from the microcarriers and from receiving shear stress caused by a medium flow without being packed, microcarriers and cells were trapped in a spin filter (3 m) to allow for slow exchange of culture solution while maintaining the packing effect of cells unchanged. In experimental group I, cells were trapped in a spin filter and cultured with Cytoline 1. Considering shear stress, cells were stirred at RPM 0. In experimental group II, cells were cultured with shaking at RPM 25 for better medium exchange. After days 1 and 2 of culture, the proportions of orthochromatic erythroblasts were 51.4% and 64.2% in the control, 66.4% and 70.0% experimental group I, and 63.2% and 59.4% experimental group II. The numbers of enucleated erythrocytes were 15.0% and 25.6% in experimental group I and 25.5% and 32.3% experimental group II, which were larger than that in the control (8.6% and 10.7%). Particularly, larger numbers of enucleated erythrocytes were observed in experimental group II. On day 1 of culture, less myelodysplasia was observed in experimental groups I and II (5.7%, 1.0%, and 2.0%) than in the control (FIGS. 13-15). No cell debris was observed in the spin filter. That is, the use of the porous structure maintained the packing effect of cells, and the use of the spin filter reduced shear stress caused by a medium flow and allowed easy culture solution exchange. Therefore, the combination of the porous structure and the spin filter is the most effective for the maturation of cells and the production of erythroid cells.

(64) 8. 2D High Density Culture and 3D Packed Cell Culture Using Cytoline 1

(65) To again demonstrate the above experimental results, the present inventors reproduced the experiments with better results. Control cells were cultured at a high density on a 2D plate, cells of experimental group I were subjected to 3D packed cell culture in the presence of Cytoline 1 in a tube, and cells of experimental group II was subjected to 3D packed cell culture with Cytoline 1 in a spin filter. After days 1 and 2 of culture, the proportions of orthochromatic erythroblasts were 60.0% and 66.7% in the control, 45.1% and 56.5% in experimental group I, and 31.7% and 44.3% in experimental group II, respectively. The conditions of the experimental groups were confirmed to be more effective in cell maturation upon 3D packed cell culture (FIGS. 16-18). Due to the enucleation effect, the proportions of enucleated erythrocytes were higher in experimental group I (22.5% and 43.5%) and experimental group II (24.4% and 50.4%) than in the control (7.2% and 24.2%). In the experimental groups, intercellular contact was increased, contributing to effective packing. Myelodysplasia was decreased in the control (2.0% and 1.0%), experimental group I (4.2% and 0.0%), and experimental group II (0.0% and 1.7%) (see FIGS. 16-18).

(66) 9. Selection of Porous Structure

(67) During culture for 3 days, fast maturation to erythrocytes was observed in the 3D porous structure compared to in the control. Hemoglobin was accumulated, rendering the color of cell pellets deep red. On day 1 of culture, noticeably red cell pellets were observed in Cytoline 1 than in the other scaffolds. As the days of culture passed, pellets of cultured cells were colored red also in the scaffolds Bioemrix 3D and Honeycomb. Although the maturation rates of cells may vary depending on the kind of the structures, all 3D packed cell culture environments were confirmed to be efficient and suitable for erythrocyte production. FIG. 19 shows images of stained cells (left) and FIG. 20 shows the proportions of cells at different maturation stages. Particularly, as the days of culture increased, the viabilities of cells cultured in the scaffolds Biomerix 3D and Honeycomb were higher than those in the control (see FIG. 18, right) and the number of stained erythrocytes having undergone enucleation increased in Biomerix 3D on day 4 of culture (see FIG. 19).

(68) Cell signaling was analyzed in cells after high density packed cell culture (for 3 days) and were compared with that in cells before culture (Oh). The mRNA expression levels of deletion in liver cancer 1 (DLC-I) and intercellular adhesion molecule 4 (ICAM-4) as erythroblast adhesion-related markers, GATA-1 as a transcription factor, and Hb-beta and Hb-gamma as erythroid cell maturation markers were measured (see FIG. 21). On day 1 of culture, the mRNA expression levels of DLC-1 in Biomerix 3D and ICAM-4 were 253.2-fold and 104.9-fold higher than that in the Oh control, respectively. On day 2 of culture The mRNA expression level of GATA-1 increased by 9.1 times in erythroid cells cultured in Biomerix 3D. On day 3 of culture, the mRNA expression levels of Hb-beta increased by 313.0 times and 298.2 times in Biomerix 3D and Honeycomb. On day 3 of culture, the mRNA expression level of Hb-gamma increased by 203.3 times in Biomerix 3D. These results demonstrate that packed cell culture of mature erythroid cells in the porous structures leads to an increase intercellular contact, resulting in greatly enhanced cell adhesion and cell maturation.

(69) 10. Measurement of the Ability of Cultured Cells to Transport Oxygen

(70) FIG. 22 shows pellets of mature erythrocytes after culture (left). The ability of the mature erythroid cells to transport O.sub.2 was confirmed to be similar to that of the control (p50=20.8 and p50=25.8 for produced erythrocytes and control, respectively) (see FIG. 22).

(71) 11. Confirmation of Effect of Co-culture with Constituents of Bone Marrow Cells

(72) Erythroid cells were co-cultured with the constituents of bone marrow. As a result, the cell populations were different under the two conditions. Co-culture of all cells in the 3D structure was confirmed to be helpful in the differentiation of erythroid cells (see FIG. 24).

(73) FIG. 24 shows the results obtained after co-culture of mature erythroid cells with mesenchymal stem cells (MSCs) and osteoblasts as the cells of bone marrow in the porous structures. In experimental group A, adherent cells MSCs and osteoblasts were placed in the porous structures, allowed to stand for 3 h until cells were adhered to the porous structures, added with erythroid cells, and cultured. In experimental group B, the three types of cells were added simultaneously and co-cultured for 192 h. In the experimental group B where cells were cultured simultaneously, after 192 h, the percent of CD51+CD90+ representing MSCs was increased by 3.1%, which was maintained constant, in 3D culture compared to in 2D culture. The percent of CD45+CD71+ representing immature erythroid cells was increased by 16.0% and the percent of CD45CD71+ representing mature erythroid cells was similar to that in the control. In the experimental group A where erythroid cells were added later than adherent cells, after 192 h, the percent of cell markers CD51+CD90+ cells was lower by about 0.7% but the percent of CD45+CD71+ was higher by 21.4% in 3D culture than in 2D culture, revealing that the conditions of experimental group A are effective in maintaining and proliferating erythroid cells in the 3D porous structure. Erythroid cells in bone marrow are attached together 3-dimensionally to create spaces called erythroblastic islands, where they are mature and proliferate. The erythroblastic islands in bone marrow are important spaces where immature erythroid cells closely interact with histiocytes for their maturation. Likewise, the maturation of erythroid cells and the production of final red blood cells could be maximized through co-culture of erythroid cells with myeloid cells including histiocytes and monocytes in the 3D porous structure, which were empirically determined. To this end, bone marrow mononuclear cells (BM MNC) were added at a density of 110.sup.8 cells/mL to a 3D stainless steel porous structure, surrounded with a 3 m spin filter, and cultured for 72 h (experimental group). The conditions of 2D control, the culture medium, and medium exchange conditions were the same as described above. Cell adhesion and 3D packing in the porous structure were observed using a microscope at 24-h intervals (see FIG. 25, left). The maturation of erythroid cells by steric intercellular interaction in the porous structure was confirmed using Wright-Giemsa staining at the same time intervals (see FIG. 25, right). After 48 h of culture, immature erythroid cells were noticeably observed in the 3D experimental group compared to in the 2D control (the black arrows indicate immature erythroid cells and the remaining cells are mononuclear cells and histiocytes). Also after 72 h of culture, mature erythroid cells and histiocytes were clearly observed in the experimental group compared to in the control (the black arrows indicate immature erythroid cells, the red arrows indicate mature erythrocytes, and the white asterisks indicate histiocytes). This reveals that steric interactions between myeloid cells in the 3D porous structure play an important in the maturation of erythroid cells.

(74) While specific embodiments of the invention have been described in detail, it will be appreciated by those skilled in the art that such detailed descriptions are merely preferred embodiments and the scope of the present invention is not limited thereto. Therefore, the true scope of the present invention should be defined by the appended claims and their equivalents.

REFERENCES

(75) 1. Lowe K. Blood substitutes: from chemistry to clinic, Journal of materials chemistry 2006; 16: 4189-4196.

(76) 2. Timmins N E, Nielsen L K. Blood cell manufacture: current methods and future challenges. Trends in Biotechnology 2009; 27: 415-422.

(77) 3. Choi H S, Lee E M, Kim H O et al. Autonomous control of terminal erythropoiesis via physical interact ions among erythroid cells. Stem Cell Res 2013; 10: 442-453.

(78) 4. Kim H O, Baek E J. Red blood cell engineering in stroma and serum/plasma-free conditions and long term storage. Tissue Eng Part A 2012; 18: 117-126.

(79) 5. Baek E. J., Kim H. S., Kim J. H. et al. Stroma-free mass production of clinical grade red blood cells by using poloxamer 188 as a RBC survival enhancer. Transfusion 2009.