IN VITRO SKIN IMMUNE SYSTEM SIMULATION SYSTEM

Abstract

The present invention relates to a micro-fluid chip for blood vessel formation. The micro-fluid chip of the present invention is constituted by first to fifth channels arranged adjacent to one another on a substrate in sequence, and two or more micro-structures or micro-posts having a gap therebetween are disposed on the interface that each channel forms together with an adjacent channel while contacting the same. Each channel performs a fluidic interaction with a different channel through the gap formed by the micro-structures, and biochemical materials can move therethrough. The micro-fluid chip, according to the present invention, provides a micro-blood vessel having a flat and continuous blood vessel interface outside a body. Furthermore, cancer angiogenesis, cancer intravasation, and cancer extravasation can be modeled using the micro-fluid chip of the present invention. In addition, the micro-fluid chip of the present invention can be used to screen candidate anti-cancer drugs.

Claims

1. A biological tissue chip configured such that blood vessels or lymphatic vessels and cells, co-cultured in vitro, interact with each other, the biological tissue chip comprising: at least one blood vessel channel and blood vessels or lymphatic vessels or a combination of blood vessels or lymphatic vessels, formed in the blood vessel channel; at least one cell channel and cells cultured in the cell channel; and at least one medium channel, wherein the blood vessel channel, the cell channel and the medium channel are disposed adjacent and parallel to one another such that they are in fluidic communication with one another; both sides or one side of the blood vessel channel is adjacent to the medium channel, both sides or one side of the cell channel is adjacent to the other side of the medium channel, and two or more barrier structures or microstructures are disposed at an interface between adjacent two of the channels with a gap; the medium channel is connected with a medium reservoir such that they are in fluidic communication with each other, and each of the blood vessel channel and the cell channel is connected with its inlet such that they are in fluidic communication with each other; each of the channels allows an interaction between biochemical substances contained in the channels through the gap; blood vessels or lymphatic vessels are formed from angiogenic or lymphangiogenic cells in the blood vessel channel, and cells are cultured in the cell channel; and the cultured cells interact with the formed blood vessels or lymphatic vessels.

2. The biological tissue chip of claim 1, wherein the biological tissue is a skin tissue comprising a subcutaneous fat layer, a dermal layer and a horny layer.

3. The biological tissue chip of claim 1 or 2, wherein the cells are one or more selected from the group consisting of pericytes, astrocytes, cancer cells, immune cells, glial cells, mesothelial cells, fibroblasts, smooth muscle cells, pericytes, neuroglial cells, stem cells, stem cell-derived cells, and cells that interact with vascular endothelium.

4. The biological tissue chip of claim 3, wherein the co-cultured cells are mutated cells, transfected cells, or mutated and transfected cells.

5. The biological tissue chip of claim 1 or 2, wherein the angiogenic or lymphangiogenic cells are one or more selected from the group consisting of endothelial cells, epithelial cells, cancer cells, stem cells, stem cell-derived cells, and vascular endothelial progenitor cells.

6. The biological tissue chip of claim 5, wherein the angiogenic or lymphangiogenic cells are mutated cells, transfected cells, or mutated and transfected cells.

7. The biological tissue chip of claim 1, wherein a third channel 130 as the blood vessel channel, a first channel 110 and a fourth channel 140 as the medium channel, and a fifth channel 150 as the cell channel are disposed parallel to one another, wherein: one side of the first channel 110 is adjacent to one side of the second channel 120; the other side of the second channel 120 is adjacent to one side of the third channel 130; the other side of the third channel 130 is adjacent to one side of the fourth channel 140; the other side of the fourth channel is divided into two or more chambers by a barrier extending perpendicular to the other side, and each of the chambers includes a fifth channel 151 or 152 connected to one side of the fourth channel so as to be in fluidic communication with the fourth channel.

8. The biological tissue chip of claim 1, wherein the medium channel comprises: a first channel 210 configured to be in fluidic communication with a first medium reservoir 201; and a second channel 220 configured to be in fluidic communication with a second medium reservoir 202 and disposed parallel to the first channel 210; the blood vessel channel comprises: a third channel 230 configured to be in fluidic communication with a blood vessel channel inlet 203 and disposed between the first channel 210 and the second channel 220 and disposed parallel to one side of each of the first channel 210 and the second channel 220; and the cell channel comprises a fourth channel 240 configured to be in fluidic communication with a cell channel inlet 204 and adjacent to the other side of the second channel 220 and disposed parallel to the second channel 220.

9. The biological tissue chip of claim 1, wherein the medium channel comprises: a first channel 310 configured to be in fluidic communication with a first medium reservoir 301; and a second channel 320 configured to be in fluidic communication with a second medium reservoir 302 and disposed parallel to the first channel 310; the blood vessel channel comprises: a first blood vessel channel 330 configured to be in fluidic communication with a first blood vessel channel inlet 303 and adjacent to one side of the first channel 310; and a second blood vessel channel 340 configured to be in fluidic communication with a second blood vessel channel inlet 304 and adjacent to the other side of the second channel 330; and the cell channel comprises: a first cell channel 350 configured to be in fluidic communication with a first cell channel inlet 305 and adjacent to the other side of the first channel 310 and disposed parallel to the first channel 310; and a second cell channel 360 configured to be in fluidic communication with a second cell channel inlet 306 and adjacent to the other side of the second channel 320 and disposed parallel to the second channel 320.

10. The biological tissue chip of claim 7, wherein endothelial cells and fibrin gel are patterned on the third channel 130, angiogenic cells and fibrin gel are patterned on the fifth channel 151, 152, vascular endothelial cell culture medium is injected into the fourth channel 140, keratinocytes and fibrin gel are patterned on the first channel 110, keratinocyte culture medium is injected into the second channel 120, and the cells are cultured, whereby the endothelial cells in the third channel 130 form perfusable blood vessels opened only toward the fourth channel 140, and form new blood vessels toward the first channel.

11. The biological tissue chip of claim 8, wherein dermal fibroblasts and fibrin gel are patterned on the third channel 230, keratinocytes and fibrin gel are patterned on the fourth channel 240, vascular endothelial cells are injected into the first channel 210 and attached to the interface between the first channel 210 and the third channel 230, endothelial cell culture medium is injected into the first channel 210, and keratinocyte culture medium is injected into the second channel 220, whereby the attached endothelial cells form new blood vessels toward the fourth channel 240.

12. The biological tissue chip of claim 8, wherein fibrin gel is patterned on the third channel 230, dermal fibroblasts and fibrin gel are patterned on the fourth channel, vascular endothelial cells and pericytes are injected into the first channel 210, and these cells are attached to the interface between the first channel 210 and the third channel 230 and cultured, whereby the attached vascular endothelial cells form new blood vessels toward the fourth channel 240.

13. The biological tissue chip of claim 9, wherein fibroblasts and fibrin gel are patterned on the fifth channel 350 and the sixth channel 360, medium is injected into the first channel 310 and the second channel 320, angiogenic cells and fibrin gel are patterned on the fourth channel 340, and then astrocytes and fibrin gel are patterned on the third channel 330, followed by culture.

14. The biological tissue chip of claim 9, wherein fibroblasts and fibrin gel are patterned on the fifth channel 350 and the sixth channel 360, medium is injected into the first channel 310 and the second channel 320, and a mixture of angiogenic cells and astrocytes together with fibrin gel are patterned on the fourth channel 340, followed by culture.

15. A method of forming microvessels in vitro in the biological tissue chip of claim 7, the method comprising: (i) adding a mixture of fibroblasts and fibrin to the fifth channel, and a mixture of vascular endothelial cells and fibrin to the third channel, followed by culture; and (ii) maintaining the first channel and the second channel in an empty state during the culture.

16. The method of claim 15, wherein a concentration of the endothelial cells is 410.sup.6 to 810.sup.6 cells/ml.

17. A method of generating cancer angiogenesis in vitro in the biological tissue chip of claim 7, the method comprising: (i) adding a mixture of fibroblasts and fibrin to the fifth channel, and adding a mixture of vascular endothelial cells and fibrin to the third channel, followed by culture; (ii) maintaining the first channel and the second channel in an empty channel state during the culture, thereby forming microvessels; and (iii) injecting an angiogenic cell line into the first channel, and injecting fibrin into the second channel, followed by culture.

18. The method of claim 17, wherein the fibroblasts are lung fibroblasts (LF), the endothelial cells are HUVEC, and the angiogenic cell line is U87MG cell line (ATCC HTB-14).

19. A method of generating cancer intravasation in vitro in the biological tissue chip of claim 7, the method comprising: (i) adding a mixture of fibroblasts and fibrin to the fifth channel, and adding a mixture of vascular endothelial cells and fibrin gel to the third channel, followed by culture; (ii) maintaining the first channel and the second channel in an empty state during the culture, thereby forming microvessels; (iii) injecting an angiogenic cell line into the first channel, and injecting fibrin gel into the second channel, followed by culture, thereby generating cancer angiogenesis; (iv) attaching cancer cells to the fibrin gel of the second channel; (v) supplying a medium for cancer cell growth to the first channel, and then adding growth factor-free medium to the first channel.

20. A method of screening an anticancer drug candidate in vitro in the biological tissue chip of claim 7, the method comprising: (i) adding a mixture of fibroblasts and fibrin to the fifth channel, and adding a mixture of vascular endothelial cells and fibrin to the third channel, followed by cultured; (ii) maintaining the first channel and the second channel in an empty channel state during the culture, thereby forming microvessels; (iii) injecting an angiogenic cell line and a sample to be analyzed into the first channel, and injecting fibrin into the second channel, followed by culture; and (iv) determining that the sample is an anticancer drug candidate, when cancer angiogenesis is not generated.

21. A method for generating blood vessels or lymphatic vessels and cells, which interact with each other in vitro, the method comprising: sequentially or simultaneously injecting one or more, selected from the group consisting of angiogenic cells, lymphangiogenic cells, extracellular matrices, cell culture media, angiogenic factors, lymphangiogenic factors and co-culture cells, into one or more independent channels of the biological tissue chip according to claim 1; culturing angiogenic cells; inducing blood vessel formation; and culturing co-culture cells.

22. A method for generating blood vessels or lymphatic vessels and cells, which interact with each other in vitro, the method comprising the steps of: (a) injecting extracellular matrix and angiogenic or lymphangiogenic cells into the blood vessel channel of the biological tissue chip according to claim 1; (b) injecting extracellular matrix or a combination of extracellular matrix and co-culture cells into the cell channel; and (c) injecting cell culture medium, angiogenic or lymphangiogenic factor, or a combination of cell culture medium and angiogenic or lymphangiogenic factor into the medium channel, inducing blood vessel or lymphatic vessel formation in the blood vessel channel, and culturing the co-culture cells in the cell channel.

23. A method for generating blood vessels or lymphatic vessels and cells, which interact with each other in vitro, the method comprising the steps of: (a) injecting extracellular matrix or a combination of extracellular matrix and co-culture cells into the blood vessel channel of the biological tissue chip according to claim 1, and forming a cell adhesion surface for cell adhesion at an interface between the blood vessel channel and the medium channel; (b) injecting angiogenic cells into the medium channel, and attaching the angiogenic cells to the cell adhesion surface; (c) injecting extracellular matrix or a combination of extracellular matrix and co-culture cells into the cell channel; and (d) injecting cell culture medium, angiogenic factor, or a combination of cell culture medium and angiogenic factor into the medium channel, culturing in the angiogenic cells in the blood vessel channel, and inducing blood vessel formation.

24. The method of any one of claims 21 to 23, wherein the angiogenic cells are one or more selected from the group consisting of endothelial cells, epithelial cells, cancer cells, stem cells, stem cell-derived cells, and endothelial progenitor cells.

25. The method of claim 24, wherein the angiogenic cells are mutated cells, transfected cells, or mutated and transfected cells.

26. The method of any one of claims 21 to 23, wherein the extracellular matrix is one or more selected from then group consisting of collagen gel, fibrin gel, Matrigel, self-assembled peptide gel, polyethylene glycol gel, and alginate gel.

27. The method of any one of claims 21 to 23, wherein the co-culture cells are one or more selected from the group consisting of astrocytes, glial cells, mesothelial cells, fibroblasts, smooth muscle cells, pericytes, neuroglial cells, stem cells, stem cell-derived cells, and cells that interact with vascular endothelium.

28. The method of claim 27, wherein the co-culture cells are mutated cells, transfected cells, or mutated and transfected cells.

29. The method of any one of claims 21 to 23, wherein the extracellular matrix or the cell culture medium comprises one or more selected from the group consisting of drugs, soluble factors, insoluble factors, biomolecules, proteins, nanomaterials, and siRNA.

30. A biological tissue chip of mimicking a skin immune system in vitro, the chip comprising immune cells co-cultured in a cell channel of a cell tissue chip set forth in claim 1 or 2.

Description

DESCRIPTION OF DRAWINGS

[0095] FIG. 1 shows the structure of a microfluidic chip of the present invention and a design of experimental procedures. Specifically, FIG. 1A is a schematic view showing the structure of the microfluidic chip of the present invention.

[0096] FIG. 1B schematically shows microvessels generated according to the present invention.

[0097] FIG. 1C schematically shows cancer angiogenesis generated according to the present invention.

[0098] FIG. 1D schematically shows cancer intravasation generated according to the present invention.

[0099] FIG. 2 shows the results of optimizing the HUVEC concentration for generating smooth and continuous vascular walls.

[0100] FIG. 2A is a confocal micrograph of microvessels generated depending on the initial HUVEC concentration (7 days after HUVEC inoculation). When the HUVEC concentration was 610.sup.6 cells/ml, smooth and continuous vascular walls were formed.

[0101] FIG. 2B shows the success rate of formation of smooth microvascular walls. When the HUVEC concentration was 610.sup.6 cells/ml, the success rate of formation of smooth microvascular walls was the highest.

[0102] FIG. 2C depicts time-lapse micrographs of microvascular formation at a HUVEC concentration of 610.sup.6 cells/ml.

[0103] FIG. 3 depicts fluorescence micrographs of fully developed microvessels. Specifically, FIG. 3A shows lumens, formed in microvessels, through a three-dimensional (3D) projection and a cross-sectional image of the microvessels.

[0104] FIG. 3B is a micrograph showing before and after FITC-dextran solution is injected into microvessels.

[0105] FIGS. 3C and 3D show microvessels connected to one another. Smooth and clear lines of claudin-5 and ZO-1 suggest that proper connections have been formed. On day 2, dispersed HUVEC cells were elongated and began to differentiate into tubular forms. On day 4, HUVEC cells began to fuse together to form the lumen (interior space). On day 7, HUVEC cells were fully fused to form microvessels having substantial lumens and smooth vessel walls.

[0106] FIG. 4 depicts micrographs showing the results of a cancer angiogenesis experiment and shows the results of quantification. Specifically, FIG. 4A compares microscopic images of microvessel walls before and 3 days after injection of cancer cells. Microvascular sprouts induced by cancer were greatly reduced by treatment with bevacizumab.

[0107] FIGS. 4B and 4C show the results of quantifying the number and coverage area of sprouts under various conditions. Angiogenic sprouts from microvascular walls were formed toward the upper channel and promoted by the secretion of cancer cells. Treatment with bevacizumab significantly reduced the coverage area and number of sprouts, indicating the anti-angiogenic potential of bevacizumab in cancer treatment (***p<0.0005). The error bars represent SEM.

[0108] FIG. 5 depicts micrographs showing the results of a cancer intravasation experiment and shows the results of quantification. Specifically, FIG. 5A shows a three-dimensional micrograph of cancer cells that migrated through the microvascular walls (red: CD31, green: MDA-MB-231, blue: nucleus). Fluorescence and DIC micrographs show that cancer cells have penetrated into the microvascular walls.

[0109] FIG. 5B is a micrograph of VE-cadherin expressed on the cell-cell junctions in microvessels. Compared with microvessels (left) under normal conditions, VE-cadherin in TNF--treated microvascular exhibits a disrupted and pleated form (right), suggesting that the junctions were disrupted by the effect of TNF-. Compared with a control experiment, TNF-a-treated microvessels show a very high rate of cancer intravasation, demonstrating the effect of TNF-a on the connection between microvessels and cancer intravasation (*p<0.0005 compared to control). The error bars represent SEM.

[0110] FIGS. 6, 7 and 8 schematically illustrate several forms of use of FIG. 1A.

[0111] FIGS. 9A, 9B and 9C illustrate a platform that mimics the immune function of the skin's structural layer according to the present invention.

[0112] FIG. 10 shows the structure of a microfluidic chip of the present invention.

[0113] FIG. 11A shows the structure of a microfluidic chip of the present invention, and FIGS. 11B and 11C show the result of co-culturing pericytes by use of the microfluidic chip.

[0114] FIGS. 12A to 12F show the results of analyzing the morphological characteristics of blood vessels generated by co-culturing pericytes by use of the microfluidic chip of the present invention.

[0115] FIG. 13 shows the results of observing the responses of blood vessels and pericytes to VEGF-A, TNF-a, and IL-1a, which are typical factors that frequently occur in the peritumoral environment and inflammatory conditions.

[0116] FIGS. 14 and 15 show the structure of a microfluidic chip of the present invention and two methods of co-culturing vascular cells and their peripheral cells.

[0117] FIGS. 16A and 16B shows the results of observing blood-astrocytes co-cultured using the microfluidic chip of the present invention.

[0118] FIGS. 17A and 17B show the results of observing blood vessels and lymphatic vessels cultured using the microfluidic chip of the present invention.

[0119] FIG. 18 shows a skin's structural layer mimicked using the microfluidic chip of the present invention.

[0120] FIG. 19 shows the structure of a microfluidic chip of the present invention.

[0121] FIG. 20 shows the results of observing the passage of immune cells and cancer cells through subcutaneous blood vessels by use of the microfluidic chip of the present invention.

BEST MODE

[0122] Hereinafter, the constituent elements and technical features of the present invention will be described with reference to examples below. However, these examples are only to illustrate the present invention, and the scope of the present invention is not limited by these examples.

Experimental Materials

[0123] Cell Culture, Immunostaining and Reagents

[0124] Human umbilical vein endothelial cells (HUVEC, Lonza) were cultured in endothelial cell growth medium (EGM-2, Lonza). Normal human lung fibroblasts (LF, Lonza) were cultured in fibroblast growth medium (FGM-2, Lonza). Human glioblastoma cells (U87MG, ATCC, Virginia) were cultured in DMEM medium supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/ml) and streptomycin (100 U/ml). MDA-MA-231 cells were purchased from the ATCC (Manassas, Va.). MDA-MB-231 cells were transfected with a pEGFP plasmid, and cells were selected using 1 mg/ml G418 (A.G. scientific, Inc.). MDA-MA-231 GFP cells obtained from monoclones were incubated in RPMI1640 (WELGENE, Korea) supplemented with 10% fetal bovine serum, 1% penicillin/streptomycin (Gibco, BRL) and 250 g/ml G418 (A.G. scientific, Inc.). All cells were incubated in a humidified incubator at 37 C. under 5% CO2. For immunostaining, endothelial cells were imaged using mouse monoclonal antibodies specific for human ZO-1 (Alexa Fluor594, clone ZO1-1A12, molecular probe), CD31 (AlexaFluor1647, clone WM59, Biolegends), VE-cadherin (eBioscience) and claudin-5 (Invitrogen), and the nucleus was stained with Hoechst 33342 (molecular probe). Bevacizumab (Avastin, Genentech) was diluted to 500 g/ml, and microvessels introduced with cancer were treated with the dilution in cancer angiogenesis experiments. Recombinant human TNF-alpha (PetroTech) was diluted to 5 ng/ml, and microvessels were treated with the dilution at 24 hours before introduction of cancer.

[0125] Fabrication of Microfluidic Chip

[0126] A master mold was made by casting photoresist onto a silicon wafer. A 80-m thickness mold was made according to the standard photolithography protocol (Xia, Y. et al. 1998) for SU-8100 (Microchem, US) photoresist. PDMS (Dow Corning, US) was poured onto the prepared master mold and cured in a dry oven at 80 C. PDMS and a cleaned coverslip were bonded together by plasma treatment (Femto Science, KR). To make the surface hydrophobic, the bonded device was kept in an oven at 80 C. for 48 hours or more.

[0127] Hydrogel and Cell Loading

[0128] HUVECs (used at 610.sup.6 cells/ml in most experiments, and used at 3 or 910.sup.6 cells/ml in some experiments) and LFs (710.sup.6/ml) were mixed with a fibrinogen solution (2.5 mg/ml fibrinogen, 0.15 U/ml aprotinin and 0.5 U/ml thrombin) and injected into the third channel 130 (blood vessel channel) and the fifth channel 150, 151 or 152 (LF channel), respectively (FIG. 1A). For fibrin polymerization, incubation was performed for 2 minutes, after which EGM-2 medium was filled in the medium channel. The device was incubated for 7-8 days, thereby forming fully lumenized microvessels having open ends at each medium channel (FIG. 1B). After vascular maturation, U87MG and MDA-MA-231 cells were harvested from tissue culture dishes. For cancer angiogenesis, the U87MG cells treated with fibrinogen solution was injected into the first channel 110 (upper channel), and the second channel 120 (bridge channel) was filled with fibrinogen solution. For intravasation of cancer cells, MDA-MB-231 cells (110.sup.6 cells/ml) together with medium were injected into the second channel 120, and adhered to the fibrin wall between the second channel 120 and the third channel 130 (blood vessel channel) by tilting for 40 minutes. In order to induce chemotactic migration of cancer cells, EBM-2 (medium supplemented with no additional growth factor) was filled in the second channel 120 and the first channel 110, and EGM-2 medium was filled in the fourth channel 140.

[0129] Microscopy

[0130] For microvascular DIC (Differential Interference Contrast Microscope) imaging, the Nikon AE31 microscope was used. For 3D z-stack and cross-sectional imaging, stained samples were imaged using a confocal microscope (Olympus FV1000). Confocal images were analyzed using IMARIS software (Bitplane, Switzerland). For fluorescence imaging, FITC-dextran-injected samples were imaged using the IX81 inverted microscope (Olympus).

[0131] Data Analysis

[0132] To quantify the success rate of the smooth and continuous boundary of microvessels, the length of the vessel boundary was measured, and compared with the straight-line length of the blood vessel channel along the microvessel by use of image J. When the boundary length value was within 10% of the linear length of the vascular channel, the chip was regarded as success. All microvessels having disconnected blood vessel walls were regarded as failure. Calculation of the permeability coefficient was performed using the method disclosed in the prior art (Non-Patent Document 19: Lee et al. 2014b). Specifically, FITC-dextran solution was introduced into microvessels and fluorescently imaged in every 15 seconds using multi-stage time-lapse mode in Metamorph. The acquired time-lapse image was analyzed using Image J, and the permeability coefficient was calculated using Equation 1 below:

P=1/lw(dl/dt)/li

where lw is the length of the vessel wall that separates between perivascular region and microvessel region, li is the mean intensity in the microvessel region, I is the total intensity in the perivascular region.

[0133] For the quantification of cancer angiogenesis, sprout area and the number of sprouts were manually quantified by using Image J. For quantification of cancer intravasation, the microvessels were stained with CD31 and fluorescence imaged, and cancer cells at the apical side of the microvessel were manually counted using the IMARIS software.

EXAMPLE

Example 1: Fabrication of Microfluidic Chip

[0134] In previous works, the present inventors have described in detail the formation of a perfusable microvessel network in a microfluidic device using a co-culture system of HUVEC and LF. The HUVEC sprouts were stimulated by LF, which opened their lumens to both sides of the channel, allowing fluid passage through the vessels. However, the structure of microvessel in the previous model was unpredictable, and other cell types could not be introduced into the perivascular region, as the perivascular regions were filled with gel or PDMS (polydimethylsiloxane) wall. Therefore, the present inventors generated a microvessel with more predictable geometric characteristics and with perivascular regions that could be filled with other cell types after the generation of the microvessels with vasculogenic process. However, in this Example, the present inventors positioned two openings of the third channel 130 (blood vessel channel) on the same side (lower side), and interfaced the upper portion of the third channel 130 (blood vessel channel) with an empty channel (FIG. 1A). This generates a microvessel having two openings on the same side, while the other side of the microvessel is interfaced with the second channel 120, which is named the bridge channel in FIG. 1A.

[0135] The first channel 110 (upper channel) and the second channel 120 (bridge channel) were empty during microvessel growth. The interface with the empty channel prevented HUVECs in the upper region from migrating or generating sprouts toward the upper direction of the third channel 130 (blood vessel channel), and resulted in the generation of a smooth vessel wall, parallel to the interface.

[0136] FIGS. 1B to 1D show a schematic view of the generation of a microvessel, and the performance of the cancer angiogenesis and intravasation assays using the microfluidic chip 10. The present inventors first injected an LF-fibrin mixture into the fifth LF channel 150, 151 or 152 (LF channel), and a HUVEC-fibrin mixture into the third channel (blood vessel channel) to generate the microvessel. Medium was added to connect the cell-loaded channels. A microvessel with two openings toward the fourth channel 140 (medium channel) was generated after 7-8 days of incubation (FIG. 1B). Next, cancer angiogenesis was modeled using the U87MG cell line (obtained from the Korean Cell Line Bank) known to have high angiogenic potential as a cancer sprout inducer from the microvessel. The present inventors injected an U87MG-fibrin mixture into the first channel 110 (upper channel), followed by fibrin injection into the second channel 120 (bridge channel) (FIG. 1). The present inventors observed the formation of cancer sprouts from the pre-existing microvessels toward the cancer site, which was promoted by the secretion of pro-angiogenic factors from U87MG cells. The present inventors attached MDA-MB-231 cells (obtained from the Korean Cell Line Bank) to the fibrin gel exposed to the second channel 120 (bridge channel), and supplied medium to the first channel 110 (upper channel) for the cancer intravasation assay (FIG. 1D). Growth factor-free medium (EBM-2) was supplied to the first channel 110 (upper channel) to induce chemotactic migration of the cancer cells toward the microvessel wall. After 2-3 days of incubation, intravasated cancer cells were observed inside the microvessels, indicating successful modeling of the cancer intravasation process.

Example 2: Formation of Perfusable Microvessels Having Smooth and Continuous Boundaries

[0137] Formation of smooth and continuous vessel boundaries is important in reproducible data analysis and the use of microvessels in further studies. Thus, the present inventors have optimized the HUVEC cell concentration to form microvessels having smooth and continuous boundaries. The present inventors tested three conditions (3, 6, and 910.sup.6 HUVECs/ml and 7106 LF/ml) using 4-5 chips per condition, and results are expressed as the mean of three independent experiments. FIG. 2A shows the microvessel morphology on day 7 under each condition. The microvessel boundary was strongly dependent on the HUVEC concentration. Microvessels formed with 310.sup.6 HUVECs/ml showed a rough and discontinuous boundary having scattered small sprouts. Microvessels formed with 910.sup.6 HUVECs/ml had a continuous boundary with fewer, scattered, small sprouts. However, the microvessels covered the majority of the area between the microposts in the upper region and exhibited a rough boundary. In contrast, microvessels formed with 610.sup.6 HUVECs/ml showed not only a continuous boundary without scattered small sprouts but also the smoothest boundary morphology with few sprouts protruding toward the upper region of the third channel 130 (blood vessel channel). FIG. 2B shows the success rate for microvessel formation for different HUVEC concentrations. Using 610.sup.6 HUVECs/ml resulted in the highest success rate (>77%) in terms of obtaining smooth and continuous boundary morphology. Therefore, the microvessels formed using 610.sup.6 HUVECs/ml showed the most pertinent boundary morphology according to the present invention, and this condition was used in all further experiments. FIG. 2C shows a series of micrographs during microvessel formation in the third channel 130. The HUVECs began to exhibit an elongated morphology with the small vacuoles 2 days after the seeding. At this time, the microvessels were not fully connected and no patent lumen was observed. HUVEC sprouts began to form a continuous lumen on day 4, and showed an elongated shape. Then, the HUVEC sprouts began to merge to form a microvessel in the third channel 130 (vessel channel). Furthermore, HUVECs in the upper region between the microposts 610 began to migrate toward the lower direction, forming a flat and smooth microvessel boundary. The HUVEC sprouts began to be fused to yield a fully lumenized microvessel on day 7, and HUVECs in the upper region between the microposts 610 had migrated lower compared to those on day 4 to form the smooth microvessel boundary. The microvessel maintained its morphological characteristics for 14 days or more.

[0138] After formation of the microvessel (days 7-8), nuclei (blue) and CD31 (red) were immunostained and imaged with a confocal microscope (FIG. 3A). Cross-sectional images of the microvessel at various positions revealed patent three-dimensional lumens. The microvessel was composed of a clear lumen that connected both openings toward the fourth channel 140 (medium channel). The present inventors introduced microbeads (Sigma) through the microvessels, and as a result, confirmed that about 94% (101 of 108 chips) of the microvessels opened toward the medium channel and the medium could be perfused through the vascular lumens.

[0139] The present inventors introduced 20 kDa FITC-dextran solution into the microvessel to verify the patency of the lumen, and visualized the behavior of the lumen using fluorescence microscopy. The FITC-dextran filled the microvessel lumen without severe focal leakage through the wall (FIG. 3B). The permeability coefficient of the microvessel was 1.580.3210.sup.6 cm/s (meanstandard error, n=5) as determined by a time-lapse acquisition of the FITC-dextran intensity profile. The permeability coefficient was in the range of those for other in vitro models of blood vessels, indicating that the microvessel includes appropriate cell-cell junctions. The present inventors analyzed the junctions by fluorescent immunostaining for claudin-5 and ZO-1 (FIGS. 3C and 3D). The fluorescence images of two cell-cell junctions showed a smooth, clear elongated morphology without any disrupted or wrinkled positions, which are the basic characteristics of cell-cell junctions in normal blood vessels in vivo.

Example 3: Cancer Angiogenesis Assay

[0140] The present inventors analyzed the angiogenic potential of cancer cells by using the microvessels having a smooth vessel wall. After formation of the microvessels using the microfluidic chip 10 according to Example 2 (day 7-8), the present inventors introduced U87MG, a glioblastoma cell line having high angiogenic potential, and a fibrin mixture into the first channel 110 (upper channel), while the second channel 120 (bridge channel) was filled with fibrin gel only. The microvessel was regarded as pre-existing at the cancer site, whereas the U87MG cells in the perivascular region secrete angiogenic factors to induce production of angiogenic sprouts toward the microvessels. Three different concentrations (control 0/ml, 2.510.sup.6 and 510.sup.6) of U87MG cells with or without bevacizumab (bev) were used, and an anti-VEGF antibody used widely for targeting angiogenesis was used. The angiogenic sprouting from the microvessels was analyzed, and the mean values of six chips per condition were calculated.

[0141] Three days after introduction of U87MG cells into the chip, the microvessels were imaged, and the number and coverage area of angiogenic sprouts under each condition were quantified. As shown in FIG. 4A, most of the microvessels without U87MG cells showed few angiogenic sprouts, but the microvessels with U87MG cells showed a considerable number of angiogenic sprouts. Furthermore, most microvessels treated with bevacizumab had a flat boundary without angiogenic sprouts. In some bevacizumab-treated samples, sprout regression which existed prior to introduction of cancer cells and bevacizumab was observed (FIG. 4A). It was postulated that this phenomenon was due to the anti-angiogenic effect of bevacizumab.

[0142] Next, the number and coverage area of sprouts. The microvessels with cancer cells exhibited significantly increased sprout numbers and coverage areas compared to the control (FIGS. 4B and 4C). Furthermore, bevacizumab treatment attenuated the angiogenic potential of the U87MG cells, drastically decreasing the number and coverage area of the microvessel sprouts. These trends of the sprout induction by cancer cells and sprout attenuation by anti-VEGF treatment are in agreement with previous in vivo reports, and demonstrate that this microvessel model is appropriate for the assessment of drugs targeting cancer angiogenesis.

Example 4: Cancer Intravasation Assay

[0143] As a vessel to assess the transendothelial migration of MDA-MB-231 cancer cells, the microvessel obtained in the present invention was used. The cells are conventionally used for the assessment of transendothelial migration and have aggressive metastatic potential. As shown in FIG. 1D, cancer cells were attached to the interface between the fibrin gel and the second channel (bridge channel), and were incubated for 2-3 days to allow them to migrate toward and penetrate into the microvessel walls. FIG. 5A shows fluorescence and differential interference contrast (DIC) micrographs of a MDA-MB-231 cancer cell (green) in the process of intravasation through the microvessel wall (red). The number of cancer cells intravasated into the microvessels was manually counted. In this Example, the microvessels were treated with 5 ng/ml TNF- 24 hours before the introduction of the cancer cells, and the rate of cancer intravasation was compared with that of the control. The adheren junction VE-cadherin in the TNF- treated microvessels exhibited disrupted continuity and wrinkled morphology compared to control microvessels (FIG. 5B). The permeability of TNF--treated microvessels over 24 h was 2.220.6710.sup.6 cm/s (meanstandard error, n=5 chips), which was 1.4-fold higher than that under the normal condition. These results confirm the effect of TNF- on junctional disruption of endothelial cells, and are in agreement with previous reports. In addition, the effect of TNF- on cancer cell intravasation was analyzed. the rate of intravasation with or without TNF- treatment was compared (n=7 chips per condition). The rate of cancer intravasation of TNF--treated microvessels was 3-fold higher than that of the control (FIG. 5C). This result suggests that TNF- treatment exerted marked effects on cancer cell intravasation, in agreement with previous reports.

Example 5: Blood Vessel-Pericyte Co-Culture

[0144] Using the microfluidic chip 20 (FIG. 11A) of the present invention, blood vessels and pericytes were co-cultured. 2.5 mg/ml fibrin gel was patterned on the third channel 230 of the microfluidic chip 20, and a mixture of dermal fibroblasts and fibrin gel was patterned on the fourth channel 240 at a concentration of 5 to 10 million/ml, thereby forming a three-dimensional environment. HUVECs (blood vessel cells) and pericytes (pericytes were obtained from Procell, Germany) were mixed at a ratio of 5:1, and a cell suspension having a cell concentration of a total of 6 million/ml was obtained and injected into the first channel 210. Next, the chip device was tilted by an angle of 90 for 30 minutes such that the cells were attached to the fibrin gel at the interface between the first channel 210 and the third channel 230. Skin fibroblasts injected into the fourth channel 240 formed an asymmetric gradient of growth factor between the first channel 210 to the third channel 230, and thus the attached vascular cells showed characteristic collective migration similar to in vivo angiogenesis within 6 days. At this time, pericytes were also formed while surrounding the blood vessels. As the fibroblasts injected into the fourth channel 240 and formed the growth factor gradient, dermal fibroblasts were used. In this case, it was confirmed that the contact between the blood vessels and the pericytes became closer to each other. The morphological characteristics of the blood vessels formed in the microfluidic chip according to this Example were analyzed, and as a result, it was shown that (a) when blood vessels were co-cultured with pericytes, the vessels became much thinner and the number of vascular branches increased, compared to when the blood vessels were cultured alone. This suggests that the pericytes have the effect of inhibiting the expansion of blood vessels by interaction with the blood vessels. Fluorescence micrographs were analyzed by a computer, thereby quantifying (b) the width of blood vessel, (c) the number of blood vessel branches; and (d) the number of junctions. As a result, it was confirmed statistically and quantitatively that when blood vessels were co-cultured with pericytes, the width of the blood vessels decreased and the number of branches and junctions increased. These results are shown in FIGS. 12A to 12F.

Example 6: Mimicking of Skin Immune SystemObservation of the Response of Blood Vessels and Pericytes to Immune Cells

[0145] FIG. 13 shows the results obtained by adding VEGF-A (b), TNF-a (c) and IL-1a (d), which are typical factors that frequently occur in the peritumoral environment and inflammatory conditions, to the blood vessel-pericytes co-cultured in Example 5 above, and observing the responses of the blood vessels and the pericytes to these factors in comparison with a control (a). In FIG. 13, light pink indicates nuclei, dark pink indicates CD31 that stains the blood vessels, and green indicates -SMA that stains the pericytes. In the test groups (c and d) mimicking the situation where the immune system was activated, it can be observed that the pericytes sprouted out from the blood vessels and formed many filopodia, compared to the control group. This mimics the situation where blood vessels become unstable and angiogenesis increases.

Example 7: Blood Vessel-Astrocyte Co-Culture

[0146] For a microfluidic chip 30 (FIGS. 14 and 15), two hydrogels were patterned adjacent to each other (third channel 330 and fourth channel 340), unlike a conventional chip. In order to perform patterning such that air bubbles do not enter the hydrogel channel formed in the third channel 330 and the fourth channel 340, hydrogel was filled in the first channel 340 and cured for 2-5 minutes, and then the third channel 330 was filled with hydrogel. On the fifth channel 350 and the sixth channel 360, fibrin hydrogel mixed with 5-10 million/ml of lung fibroblasts (Lonza, 3D ECM; obtained from Sigma Aldrich Korea) was patterned. Medium was added to the first channel 310 and the second channel 320. To see the effect of the physical junction of astrocytes to blood vessel cells, 5 million/ml of blood vessel cells-astrocytes were added to each of the third channel 330 and the fourth channel 340, which were divided so as to be close to move toward each other but away from each other. FIGS. 16A and 16B are confocal micrographs showing the results. When the blood vessel cells and the astrocytes were cultured separately in the third channel 330 and the fourth channel 340, angiogenesis in the channel containing the astrocytes was inhibited (FIG. 16A). In addition, blood vessel cells and astrocytes were mixed at a ratio of 1:1 and co-cultured in the third channel 330 or the fourth channel 340. In this case, the mixture of the two types of cells was added at a concentration of 10 million/ml. When the blood vessel cells and the astrocytes were co-cultured in the same channel, it was observed that the endfoot of the astrocytes was formed toward the blood vessels (FIG. 16B). In FIG. 16B, green fluorescence indicates GFAP that selectively stains the astrocytes, and white indicates CD31 that stains the blood vessel cells. FIGS. 14 and 15 show the structure of the microfluidic chip used in this Example.

[0147] In addition, two fibrin hydrogels were patterned adjacent to each other, and blood vessel cells or lymphatic vessel cells were cultured alone in the fourth channel 340, and in vivo blood vessel/lymphatic vessel formation processes were observed. Over 2-3 days, a blood vessel or lymphatic vessel network was first formed in the fourth channel 340, and then over 2-3 days, angiogenesis could be induced from the generated blood vessel network. This confirms that it is possible to sequentially and accurately mimic in vivo blood vessel formation processes in which a blood vessel network is formed through angiogenesis following vasculogenesis. In addition, it is also possible to mimic the process of lymphatic vessel formation and reproduce lymphatic vessels in this context. The results are shown in FIG. 17. In FIG. 17A, green fluorescence indicates F-actin, and red indicates CD31 that selectively stains blood vessels. It is a confocal micrograph showing an intermediate process in which angiogenesis was generated in the left empty half of the hydrogen after angiogenesis in the right half of the hydrogel was completed. FIG. 17B is a micrograph of lymphatic vessel cells which were grown and stained in the same manner. In FIG. 17B, green indicates Podoplanin known to stain only lymphatic vessels, and red indicates F-actin.

Example 8: Horizontal Culture of Skin's Structural Layers

[0148] In this Example, the skin's structural layer was cultured horizontally. In this Example, the microfluidic chip 20 having the structure shown in FIG. 11A was used. The height of the chip was adjusted to 100 m, and the height of each channel was adjusted to 500 m to 800 m, and triangular or hexagonal posts were disposed between the channels. These posts could surface tension, and thus hydrogel could be patterned sequentially on the third channel 230 and the fourth channel 240 without mixing. On the third channel 230, 2.5 mg/ml fibrin gel mixed with 2 million/ml of dermal fibroblasts was patterned, and on the fourth channel 240, fibrin gel mixed with 5-10 million/ml of keratinocytes was patterned. After the concentration of the cell supernatant was adjusted such that HDMECs (blood vessel cells) had a cell concentration of 7 million cells/ml, the cell supernatant was injected into the first channel 210, and the chip was tilted by an angle of 90 for 30 minutes such that the cells were attached to the fibrin gel at the interface between the first channel 210 and the third channel 230 by gravity. Next, EGM2-MV (HDMEC culture medium) was added to the first channel 210, and Epilife (keratinocyte culture medium, Thermo Fisher Scientific) was injected into the second channel 220, followed by culture of the cells. The keratinocytes injected into the fourth channel 240 formed an asymmetric concentration gradient of growth factor between the first channel 210 to the third channel 230, and thus the attached blood vessel cells exhibited characteristic collective migration similar to in vivo angiogenesis within about 6 days. In this case, the dermal fibroblasts in the third channel 230 helped the growth of blood vessels. FIG. 18 is a confocal micrograph showing the results of observation. In FIG. 18, the green fluorescence (HDMEC marker CD31) portion on the left, in which the blood vessel cells sprout out, is the interface between the first channel 210 and the third channel 230. The blue points in the third channel 230 are the nuclei of the dermal fibroblasts. In this Example, the first layer 210 mimicked the skin's subcutaneous fat layer having blood vessels; the third channel 230 mimicked the dermal layer having blood vessels and fibrous cells; and the fourth channel 240 mimicked the epidermal layer having keratinocytes.

Example 9: Mimicking of Immune Function of Skin's Structural Layers

[0149] In this Example, the immune function of the skin's structural layers was mimicked. In this Example, the microfluidic chip 10 having the structure shown in FIG. 19A was used. The height of the chip device was adjusted to 100 m, and the height of each channel was adjusted to 500-800 m. Triangular or hexagonal posts were disposed between the channels, and thus hydrogel could be patterned sequentially on the first channel 110, the third channel 130 and the fifth channel 150 without mixing. 2.5 mg/ml fibrin gel mixed with 5-10 million/ml of endothelial cells was patterned on the third channel 130, and fibrin gel mixed with 5-10 million/ml of angiogenic cells (fibroblasts, msc, cancer, etc.) was patterned on the fifth channel 151 or 152. When blood vessel culture medium was injected into the fourth channel 140, the angiogenic cells injected into the fifth channel 151 or 152 formed an asymmetric concentration gradient of growth factor between the third channel 130 and the fourth channel 140, and thus the blood vessel cells formed a perfusable blood vessel that opened only toward the fourth channel 140, without about 6 days. Next, 2.5 mg/ml fibrin gel mixed with 5-10 million/ml of keratinocytes was patterned on the first channel 110, and the keratinocyte culture medium Epilife was injected into the second channel 120. In the third channel 130, a blood vessel network that opened on one side formed new blood vessels by the influence of the keratinocyte growth factor present in the first channel 110. Next, immune cells were injected into the fourth channel 140, and then the immune cells entered the blood vessels through the opened portion, and then extravasated from the blood vessel walls, and were positioned between the second channel 120 and the third channel 130. Thus, this Example suggested a platform capable of observing in vivo blood vessels and tissue structures, and the migration and response of immune cells therein (see FIG. 20). It reproduced the process in which cancer cells and immune cells extravasate through blood cells, thereby mimicking the process of formation of the skin immune system, for example, the migration of progenitor cells of the skin immune system through blood vessel walls and the differentiation of the progenitor cells in skin tissue. Thus, it was confirmed that the present invention makes it possible to mimic the skin immune system in vitro more efficiently than when mature immune cells are used.

DESCRIPTION OF REFERENCE NUMERALS

[0150] 10, 20, 30: microfluidic chip; [0151] 110, 210, 310: first channel; [0152] 120, 220, 320: second channel; [0153] 130: third channel; 140: fourth channel; [0154] 150, 151, 152: fifth channel; [0155] 160: barrier; [0156] 201, 301: first medium reservoir; [0157] 202, 302: second medium reservoir; [0158] 203: blood vessel channel inlet; [0159] 204: cell channel inlet; [0160] 230: blood vessel channel; 240: cell channel; [0161] 303: first blood vessel channel inlet; [0162] 304: second blood vessel channel inlet; [0163] 305: first cell channel inlet; [0164] 306: second cell channel inlet; [0165] 330: first blood vessel channel; [0166] 340: second blood vessel channel; [0167] 350: first cell channel; [0168] 360: second cell channel; [0169] 610: microstructures; [0170] 620: gap; [0171] 660: barrier structures.