DEVELOPMENT OF SPINAL CORD ON A MICROFLUIDIC CHIP

Abstract

The invention relates to culturing brain endothelial cells, and optionally astrocytes and neurons in a fluidic device under conditions whereby the cells mimic the structure and function of the blood brain barrier. Culture of such cells in a microfluidic device, whether alone or in combination with other cells, drives maturation and/or differentiation further than existing systems.

Claims

1-41. (canceled)

42. A method of culturing cells, comprising: a) providing a microfluidic device comprising a top chamber, a bottom chamber, and a membrane disposed in at least one interface region between said top and bottom chambers; b) seeding stem cell derived neurons within at least a portion of said top chamber and brain microvascular endothelial cells within at least a portion of said bottom chamber so as to create seeded cells; c) exposing at least a portion of said seeded cells to a flow of a first fluid for a period of time; and d) culturing said seeded cells under conditions such that a percentage of said neurons exhibit development indicative of the spinal cord.

43. The method of claim 42, wherein said exhibited development indicative of the spinal cord comprises spontaneous bursts of calcium transient activity.

44. The method of claim 43, wherein said spontaneous bursts activity is periodic.

45. The method of claim 42, wherein said exhibited development indicative of the spinal cord comprises at least one marker selected from the list comprising SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

46. The method of claim 42, wherein said top chamber comprises an open region.

47. The method of claim 42, wherein said microfluidic device further comprises a top channel, said top channel fluidically coupled to said top chamber.

48. The method of claim 42, wherein said microfluidic device further comprises a bottom channel, said bottom channel fluidically coupled to said bottom chamber.

49. The method of claim 42, wherein said stem-cell derived neurons seeded within at least a portion of said top chamber are disposed within a gel or gel-precursor.

50. The method of claim 42, wherein said stem-cell derived neurons are seeded on top of a gel present within a portion of said top chamber.

51. The method of claim 42, wherein said stem-cell derived neurons are derived from induced pluripotent stem-cells.

52. The method of claim 42, wherein said stem-cell derived neurons are derived from embryonic stem-cells.

53. The method of claim 42, wherein said neurons comprise one or more cell types selected from the list comprising of motor neurons, upper motor neurons, lower motor neurons, sensory neurons, and interneurons.

54. The method of claim 42, further comprising seeding glial cells on said top surface.

55. The method of claim 54, wherein glial cells comprise one or more cell types selected from the list comprising astrocytes, oligodendrocytes, ependymal cells, Schwann cells, microglia, and satellite cells.

56. The method of claim 42, wherein said first fluid comprises culture media

57. The method of claim 42, wherein said first fluid comprises blood or at least one blood component.

58. A method of culturing cells, comprising: a) providing a microfluidic device comprising a top chamber, a bottom chamber, and a membrane disposed in at least one interface region between said top and bottom chambers; b) seeding neuron progenitors within at least a portion of said top chamber and brain microvascular endothelial cells within at least a portion of said bottom chamber so as to create seeded cells; c) exposing at least a portion of said seeded cells to a flow of a first fluid for a period of time; and d) culturing said seeded cells under conditions such that a portion of said progenitor cells differentiate to neurons and a percentage of said neurons exhibit development indicative of the spinal cord.

59. The method of claim 58, wherein said exhibited development indicative of the spinal cord comprises spontaneous bursts of calcium transient activity.

60. The method of claim 59, wherein said spontaneous bursts activity is periodic.

61. The method of claim 59, wherein said exhibited development indicative of the spinal cord comprises at least one marker selected from the list comprising SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

62. The method of claim 58, wherein said top chamber comprises an open region.

63. The method of claim 58, wherein said microfluidic device further comprises a top channel, said top channel fluidically coupled to said top chamber.

64. The method of claim 58, wherein said microfluidic device further comprises a bottom channel, said bottom channel fluidically coupled to said bottom chamber.

65. The method of claim 58, wherein said neuron progenitors seeded within at least a portion of said top chamber are disposed within a gel or gel-precursor.

66. The method of claim 58, wherein said neuron progenitors are seeded on top of a gel present within a portion of said top chamber.

67. The method of claim 58, wherein said neuron progenitors are derived from induced pluripotent stem-cells.

68. The method of claim 58, wherein said neuron progenitors are derived from embryonic stem-cells.

69. The method of claim 58, wherein said neurons comprise one or more cell types selected from the list comprising of motor neurons, upper motor neurons, lower motor neurons, sensory neurons, and interneurons.

70. The method of claim 58, further comprising seeding glial cells on said top surface.

71. The method of claim 70, wherein glial cells comprise one or more cell types selected from the list comprising astrocytes, oligodendrocytes, ependymal cells, Schwann cells, microglia, and satellite cells.

72. The method of claim 58, wherein said first fluid comprises culture media.

73. The method of claim 58, wherein said first fluid comprises blood or at least one blood component.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0058] FIG. 1 shows a schematic of one embodiment of a workflow for preparing and seeding a microfluidic chip comprising six steps. This embodiment addresses the different surface coating needs/preferences apparent for iBMECs and iMNPs based on experiments such as those illustrated in Tables 1 and 2. In particular, the workflow aims to provide, in one embodiment, different surface coatings for the top fluidic channel and bottom fluidic channel of the device.

[0059] FIG. 2 shows two schematics of microfluidic devices. In one embodiment of a microfluidic device or chip (top), the device comprises top (apical) and bottom (basal) channels (the two Xs indicating that channels are blocked during at least part of the protocol). The other schematic (bottom) shows how the ports of a microfluidic device or chip (16) can be utilized to deposit fluids carrying surface coatings (e.g. dissolved proteins) and/or seed the cells using pipette tips. This image shows a modification to the typical chip ECM coating protocol based on the need in some embodiments to coat the top and bottom channels with different ECM solutions in wet and dry conditions. The procedure developed involved an air dam by which perfusion of ECM1 loaded into the bottom channel was prevented from perfusing through the membrane to the top channel by clamping flexible tubing and trapping air in the top channel. The ports of a second microfluidic channel can be air-filled and plugged up using clips, for example.

[0060] FIG. 3A-B provides a microscopic analysis of Chip 166 from Table 2, showing neural cells in the top channel of the microfluidic device (FIG. 3A) and BMECs on the bottom channel of the microfluidic device (FIG. 3B)

[0061] FIG. 4A-C provides three images from a microfluidic chip where the cells have been tested for markers to confirm their identity. The top right image (FIG. 4B) shows good staining of BMEC tight junctions indicating BBB formation on chip. On the top left (FIG. 4A), the staining shows neurons and astrocytes. FIG. 4C is a vertical 2D projection of a 3D confocal stack of images slices, which allows for visualization of the neurons and endothelial cells together, even though they are not in the same plane on the microfluidic device.

[0062] FIG. 5 provides an image from a microfluidic chip wherein at least a portion of an apical astrocyte (i.e. the endfoot) has transmigrated the membrane and contacted the BMECs on the other side. Contact or interfacing between astrocytes and endothelial cells is a recognized feature of in vivo blood-brain barriers. To our knowledge, this interface has never been previously observed in in vitro models of the blood-brain barrier. The potential for the formation of astrocyte-endothelial contact observed in some of the embodiments disclosed herein is desired and advantageous, as it is believed that the in vivo contact/junction is related to the tight barrier properties characteristic of the blood-brain barrier.

[0063] FIG. 6A-B shows a first image (FIG. 6A) where iMNs were seeded on a plain (un-patterned) surface, as well as a second image (FIG. 6B) where the same cells were seeded on a nanopatterned surface, resulting in directed neurite growth. Such nanopatterning can be applied to the membrane or any surface of the BBB-on-chip. In particular embodiments, the nanopatterning is applied to the top surface of the membrane to direct neurite growth for neuron seeded on said surface. It is desired in some uses to direct neurite growth, for example, in studying neuron biology or disease (e.g. conditions that disturb neurite growth or its directionality), as a readout of neuron or blood-brain-barrier health (e.g. by monitoring neurite growth or its directionality) or in facilitating electrophysiological measurements (e.g. using a multi-electrode array or patch clamping). The preferred nanopattern is linear valleys and ridges, but alternatives such as circular, curved, or any other desired shape or combination thereof are also contemplated. Linear nanopatterning can include, for example, line spacing ranging from 10 nm to 1 um, 0.5 um to 10 um or 5 um to 50 um, and line depth ranging from 10 nm to 100 nm, 50 nm to 1000 nm, 200 nm to 5 um or 2 um to 50 um.

[0064] FIG. 7A-D show microscopic examination of the morphology of fresh (not frozen) BMECs seeded on a 4:1 mixture of collagen and fibronectin that has either been dried (FIG. 7A, top left) or remained wet (FIG. 7B, top right), as well as an example where the same fresh cells were seeded on laminin (FIGS. 7C and D, the arrow indicating contamination of the cells with neurons).

[0065] FIG. 8 is a schematic showing one embodiment of a standard syringe pump and reservoir setup for perfusion of the chips mediated by flexible tubing for introducing flow into the microfluidic chips. A plurality of fluid reservoirs are in fluidic communication with a corresponding plurality of microfluidic chips via inlet ports, with tubing coming from the exit ports and attached to a plurality of syringes used to draw fluid through the chip at a flow rate. While a convenient method for creating flow conditions, other methods involving different pumping approaches or pressure approaches to drive fluid are contemplated.

[0066] FIG. 9A-D comprises photographs of microscopic examination of cell morphology on the bottom (left-hand side) and top (right-hand side) of the membrane in a microfluidic device where the cells have been exposed to flow (using the system of FIG. 8) for a number of days (7 days). FIGS. 9A and C show the results for Chip 664 where BMECs (on collagen/fibronectin) and iMPs (on wet laminin) were co-cultured. FIGS. 9B and D show the results for Chip 663 where iMPs (on laminin) were cultured alone.

[0067] FIG. 10 is a photograph of fluorescent staining of cells in a microfluidic device where the cells have been exposed to flow (using the system of FIG. 8) for a number of days. The image is a 3D image of the BMEC in the bottom channel showing a complete contiguous BMEC lumen being formed in the chip.

[0068] FIG. 11 is a photograph of fluorescent staining of cells showing the presence of neural stem cells (red) in addition to neural filaments (green), with the nuclei stained with DAPI. In the preferred embodiment, the BBB-on-chip includes endothelial or endothelial-like cells (preferably brain-related endothelial cells) and optionally astrocytes or astrocyte-like cells. However, in some embodiments, the BBB-on-chip contains additional cells type such as, for example, neurons, pericytes and various progenitor cells. Such cells may be included as an intended or unintended bi-product of the stem cell differentiation process from which the astrocytes or endothelial cells are generated (whether on chip or preceding it), as stem cells and progenitor cells are typically capable of differentiating into a plurality of cells types. The presence of neurons is desirable in some embodiments because they can be used as readouts of BBB function (e.g. agents penetrating the barrier may affect the neurons in measurable ways) or because they may interact with other cells types or help generate a local environment that improves the function of the BBB-on-chip. Similarly, pericytes are desirable in some embodiments, because it is believed in the art that they help establish the blood-brain barrier and can be potentially monitored to evaluate BBB health. Neuronal- or endothelial-lineage progenitors are desirable in some embodiments, as they may replenish cell populations and be potentially monitored to evaluate BBB health. Accordingly, in some embodiments, neurons, pericytes, neuronal-lineage progenitors, endothelial-lineage progenitors or combinations thereof or progenitors thereof may be deposited in the BBB-on-chip. In other embodiments, a differentiation process is employed (whether on chip or preceding it) to generate one or more of these cells types.

[0069] FIGS. 12A and 12B show graphs with functional measurements performed on BBB-on-chips. FIG. 12A shows the results/readout from transepithelial electrical resistance (TEER) measurements on the microfluidic chip under flow, static, and control conditions. Clearly, flow is important for optimum results. FIG. 12B show TEER measurements on transwells. TEER is a typical measure of in vitro BBB models and is used both for evaluating the model as well as an experimental readout (e.g. after subjecting the BBB model to an experimental condition). Some aspects of the present invention include measuring the TEER of one or more BBB-on-chips. This can be done, for example, to evaluate BBB-on-chip development, maturation or quality, as a readout for experiments involve an introduced agent, as a readout for experiments involving specific cells or cell types (e.g. patient specific, a disease population, or treated to simulate a disease or condition), etc. It is known in the art how to integrate electrodes suitable for TEER measurement into microfluidic devices. Douville et al., Fabrication of Two-Layered Channel System with Embedded Electrodes to Measure Resistance Across Epithelial and Endothelial Barriers Anal Chem. 2010 Mar. 15; 82(6): 2505-2511.

[0070] FIGS. 13A and B show how TEER measurements were made in one embodiment. FIG. 13A is an enlarged schematic view showing how electrodes on the chip were connected, along with pipette tips engaging the chip; FIG. 13B shows the same connected chip to the right of a Epithelial Voltohmmeter.

[0071] FIG. 14A was a follow-up experiment on another round of prototype TEER chips that showed iBMEC barrier function increasing in the presence of flow on a chip followed by a weakening of barrier function with the exposure of the chips to TNFa, a proinflammatory cytokine. Higher TEER values generally indicate a tighter barrier, which is typically desirable for a blood-brain barrier. FIG. 14B also involves TNF alpha exposure, but the readout is membrane permeability as measured by Dextran-FITC.

[0072] FIG. 15 shows permeability results for (and the structure of) fluorescein sodium. Some aspects of the present invention include ascertaining permeability for various additional agents (e.g. drugs, chemicals, hormones, blood components, biomarkers). Such methods can allow qualitative or quantitative estimation of the permeability of the in vitro blood-brain barrier to the one or more agents. Furthermore, according to some aspects of the present invention, the permeability of one agent is measured in response to a second agent, treatment or experimental condition (for example, measuring the effect of a medication on the blood-brain barrier permeability of another medication).

[0073] FIG. 16A shows the user interface and the conditions during the run of human blood across the blood brain barrier. FIG. 16B shows the equipment setup for measuring the transport of solutes from human blood across the blood brain barrier (BBB), a barrier created in vitro in the microfluidic devices described herein using a layer of BMECs. As evidenced, some embodiments include blood or blood components, optionally perfused through one or more fluidic channels within the device. The use of blood of blood components is desired in some embodiments, as the blood or blood components can improve BBB-on-chip function, for example, by providing biochemical cues, or conversely hurt the BBB-on-chip, for example, because the blood may contain a harmful agent that may be under investigation. In some aspects, permeability assays include blood or blood components in order to provide a potentially more in vivo like result. In other aspects, individual-specific blood or blood components are used in order to potentially provide individualized BBB-related measures. This can include, for example, the measurement of the permeability of one or more agents or components from the blood or components, the effect of the blood or components on the permeability of one or more agents that may be added to the blood or another fluid included in the device, the effect of the blood or components on the health of the BBB-on-chip or any of its components (whether positive or negative), etc. This may include diagnostic uses, for example, to identify a disease, biomarker or infectious agent carried by the blood or blood components.

[0074] FIG. 17A-C shows the measurement of thyroid hormone transport by mass spectrometry (FIG. 17A) using the setup shown in FIG. 16A-B, along with the graphed results (FIG. 17B). FIG. 17C shows the ratio of T4 to IS on the Y-axis, and T4 transport across the BBB on the X-axis. After flowing patient blood through the microfluidic chips into the channel under the BMECs, it was possible to measure the transport of compounds from the blood into the neural compartment, i.e. through the BMEC barrier. In this case, the experiment included a control set of BBB-on-chips comprising iPS-derived cells originating from a non-diseased individual, and a second set of BBB-on-chips comprising iPS-derived cells originating from a patient diagnosed with Allan-Herndon-Dudley syndrome (AHDS). The mass-spectrometry data in FIG. 17A is an initial experiment to confirm that the MCT8 transporter defect can be recapitulated on an Organ-Chip.

[0075] FIG. 18A-B shows electrophysiology recordings collected by patch-clamp from neurons in the microfluidic device (BBB-on-Chip). An arrow (FIG. 18A) indicates single action potential. Current recordings (FIG. 18B, right) show negative sodium channel currents (Na.sup.+) and positive potassium channel (K.sup.+). These measurements on-chip can be used, for example, to provide an indication of neuronal maturation or as a readout of neuron health. In turn, neuronal maturation or health can be used as indicators of BBB-on-chip quality (for example, before starting an experiment) or as an experimental endpoint indicating, for example, that an agent as crossed the BBB, a disease condition has emerged, the BBB has been modified or compromised, or conversely, that the BBB or neural function or health have improved. Patch clamping can be performed on the BBB-on-chip by a variety of methods, including for example, by inserting the patch-clamp electrodes through the soft body of an elastomeric BBB-on-chip device. Similarly to patch-clamping, other electrophysiological readouts can be obtained, for example by including one or more electrodes in the device. In particular, a multi-electrode array (MEA) can be integrated on the membrane of embodiments that possess one or similarly in fluidic channels or cavities within the device. Electrophysiological measurements (patch-clamping, MEA) can also be applied to astrocytes, which have been shown in the art to be excitable.

[0076] FIG. 19A-D show the results of calcium flux imaging in the neural channel. The photograph (FIG. 19A, top left) is a single fluorescent image from a movie of such images. The colored circles indicate the positions that correspond to the time traces in the 3 graphs. The traces (FIGS. 19B and C) show that it is possible to observe neuronal function in the microfluidic chips in real-time. The addition of tetrodotoxin (TTX), which is a potent blocker of voltage-gated calcium channels, ablates this activity (FIG. 19D, bottom right). Calcium imaging or imaging using voltage-sensitive dyes or proteins offer similar advantages to electrophysiological readouts but offers the advantage that no electrodes are necessary. Accordingly, some aspects of the present invention include methods of measuring the BBB-on-chip by imaging in the presence of calcium or voltage-sensitive dyes or proteins, to allow the potential recording and optional manipulation of neuronal or astrocyte excitations. These measurements can be used, for example, to provide an indication of neuronal maturation or as a readout of neuron health. In turn, neuronal maturation or health can be used as indicators of BBB-on-chip quality (for example, before starting an experiment) or as an experimental endpoint indicating, for example, that an agent as crossed the BBB, a disease condition has emerged, the BBB has been modified or compromised, or conversely, that the BBB or neural function or health have improved.

[0077] FIG. 20 shows both a protocol for generating, and staining results confirming the generation of, neural cells from neural progenitors. Such techniques allow one to make multipotent neural stem cells and motor neuron precursor directly from iPSC, allowing differentiation into many neural cell types (neurons, astrocytes, etc.).

[0078] FIG. 21 shows the corrected T3 concentration in the top channel of seven different chips, i.e. chips populated with normal cells (2280, 2289 and 2284) as compared to chips populated with cells from an MCT8 cell line or patient (2285-2288).

[0079] FIG. 22A is a schematic showing one embodiment of the microfluidic device or chip (16), comprising two microchannels (1), each with an inlet and outlet port (2), as well as (optional) vacuum ports. FIG. 22B is a topside schematic of an embodiment of the perfusion disposable or pod (10) featuring the transparent (or translucent) cover (11) over the reservoirs, with the chip (16) inserted in the carrier (17). The chip can be seeded with cells and then placed in a carrier for insertion into the perfusion disposable or pod, whereupon culture media in the reservoirs flows into the microchannels and perfuses the cells (e.g. both BMECs and MNs).

[0080] FIG. 23A-B shows a schematic of an illustrative microfluidic device or organ-on-chip (16) device. The assembled device is schematically shown in FIG. 23A with the top surface (21) indicated. FIG. 23B shows an exploded view of the device of FIG. 23A, showing a bottom piece (97) having channels (98) in a parallel configuration, and a top piece (99) with a plurality of ports (2), with a tissue-tissue interface simulation region comprising a membrane (101) between the top (99) and bottom (97) pieces, where (in one embodiment) cell behavior and/or passage of gases, chemicals, molecules, particulates and cells are monitored. In an embodiment, an inlet fluid port and an outlet fluid port are in communication with the first central microchannel such that fluid can dynamically travel from the inlet fluid port to the outlet fluid port via the first central microchannel, independently of the second central microchannel. It is also contemplated that the fluid passing between the inlet and outlet fluid ports may be shared between the central microchannels. In either embodiment, characteristics of the fluid flow, such as flow rate and the like, passing through the first central microchannel is controllable independently of fluid flow characteristics through the second central microchannel and vice versa.

[0081] FIG. 24 is a print out of electrophysiological data from neurons cultured in a microfluidic device or chip, showing highly complex spontaneous activity in a chip.

[0082] FIG. 25A-B shows print outs of electrophysiological data from neurons cultured alone (FIG. 25A, top panel) and co-cultured with BMECs (FIG. 25B, bottom panel) in a microfluidic device or chip, showing that neural tissue have more mature electrophysiological properties in the chip, and in co-culture with BMECs.

[0083] FIG. 26A-B shows print outs of electrophysiological data from neurons cultured alone (FIG. 26A, top panel) and co-cultured with BMECs (FIG. 26B, bottom panel) in a microfluidic device or chip, showing that neural tissue have more mature electrophysiological properties in the chip when in co-culture with BMECs.

[0084] FIG. 27A-D provides neural calcium measurement read-outs comparing neurons (MN) co-cultured with BMECs (FIG. 27D, bottom panel), cultured alone (FIG. 27C, first panel up from the bottom panel), cultured in endothelial cell conditioned medium or ECCM in a (96-well) static culture (FIG. 27B, second panel up from the bottom panel), along with an unconditioned media (96-well) static control (FIG. 27A, top panel). Each neuron's activity is simultaneously tracked and analyzed (calcium influx is an indirect measure for neuronal activity that can be observed live in the chip). The results show that co-culture increases diMN neural calcium transient activity, i.e. a significant increase in transient frequency is observed upon contact of MNs with iBMECs.

[0085] FIG. 28 is a bar graph of neural calcium measurements (average frequency events per cell) comparing neurons (MN) co-cultured with BMECs (far right), cultured alone (next bar to the left), cultured in endothelial cell conditioned medium or ECCM in a static culture (next bar to the left), along with an unconditioned media static control (far left). The results show that co-culture increases diMN neural calcium transient activity, i.e. a significant increase in transient frequency is observed upon contact of MNs with iBMECs.

[0086] FIG. 29A-B shows the results of a transcriptomic study of iMNs in a microfluidic chip. Neurons were either cultured alone (FIG. 29A, top box) on the chip or in a co-culture with BMECs (FIG. 29B, bottom box), and this was compared with a 96-well static culture. The MNs were sorted on a FACS and RNA was sequenced (i.e. gene expression was detected). RNA-Seq from FACS sorted MNs show that neural development gene pathways (PC1) are upregulated in chip. Vascular interaction genes (PC3) are recreated in co-culture with iBMECs. In addition, there are chip induced genes (PC2), i.e. gene activity induced in the cells simply from being cultured on the chip.

[0087] FIG. 30 shows the detailed results from which FIG. 29A-B was prepared, showing the names of various neural development genes (PC1), chip induced genes (PC2) and vascular interaction genes (PC3). The colored bars on the right in FIG. 30 represent the expression of each gene (row) in each of the 5 conditions (columns). Column order is MN Only, BMEC/MN, 96-well control, 96 well ECCM, MN progenitor. Red=high and blue=low. These vascular gene pathways have not been shown to be induced in any other culture system and may be inducing the observed increase in maturity and activity.

[0088] FIG. 31 shows, at 26 days, periodic spontaneous bursts of calcium transient activity that involves 30% of all active cells.

[0089] FIG. 32A-G Exemplary Spinal Neural Progenitor (SNP) Neurons Survive And Mature In The Chip Microenvironment. FIG. 32a Schematic of Spinal neural progenitor (SNP) differentiation from induced pluripotent stem cell (iPSC) cultures. Cells were fated to neural ectoderm (NE) using WNT agonist CHIR99021 and SMAD inhibitors LDN193189 and SB431542 for 6 days then patterned to ventral spinal neurons using Retinoic Acid (RA) and Sonic hedgehog agonist (SAG) in 6-well plates. At day 12, SNPs were frozen, banked and thawed for experiments (Cryo-bank). SNPs were seeded into the top channel of the SC-chip (dotted line) and incubated for 6 days. FIG. 32b Schematic of dual-channel organ chip seeded with SNPs in top channel. FIG. 32c Phosphorylated neurofilament heavy chain (SMI32) is enriched in spinal motor neurons (spMNs) and expressed in cells populating entire top channel. Scale bar=200 microns. FIG. 32d Immunostaining of main channel of the chip of markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP), scale=200 microns (left); 40 microns (right). FIG. 32e Representative image of SC-chip neurons treated with Fluo-4 calcium activated dye and acquired live in FITC channel. FIG. 32f Florescence of individual active neurons normalized to baseline florescence and plotted over time (dF/F). FIG. 32g Raster plot showing detected events of corresponding traces.

[0090] FIG. 33A-H Exemplary Co-culture of iPSC derived MNs and BMECs in SC-Chip. FIG. 33a Schematic of dual-differentiation and seeding paradigm in EC/SC-Chip. Both SNPs and BMECs are generated from human iPSCs and seeded into top and bottom channels respectively. Transverse section of the EC/SC-Chip (right) shows two compartments separated by porous membrane. FIG. 33b Immunostaining of whole SC-Chip at 6-days incubation expressing SMI32 in top channel and ISL1 expressed by BMECs on bottom channel. FIG. 33c Maximum projection images cropped along Z-axis to show top and bottom compartments of seeding end of EC/SC-Chip immunostained with MN markers SMI32 and islet 1 (ISL1), and tight junction marker zona occludens 1 (ZO-1), Scale bar=400 microns (left) 40 microns (right). FIG. 33d Confocal optical reconstruction at along Z axis (top) of SC-Chip exhibits distinct separation of cultures at 6 days separated by porous PDMS membrane. Perspective view (bottom) exhibits confluent layer of BMECs surrounding entire bottom channel, Scale bar=100 microns. FIG. 33e Representative images of cells attached to porous membrane in the main channel (top and bottom planes). SC-Chip was seeded either with SNPs alone in the top channel (SC-Chip) or with the addition of non-GFP BMECs into the bottom compartment (EC/SC-Chip). FIG. 33f High magnification of top compartment show GFP negative, ISL1 positive clusters indicating BMEC infiltration. Quantification of FIG. 33g GFP and FIG. 33h ISL1 populations in top compartment. Error bars represent standard deviation. Plots were determined from 6 individual chips from three culture rounds (dots). No significant difference cultures determined by t-test (ns).

[0091] FIG. 34A-H Exemplary SC-Chip environment increases spontaneous neural activity. FIG. 34a Individual Representative calcium transient activity of SNP (top) and BMEC (bottom) cultures acquired at the seeding compartments plotted by change in florescence intensity over background (dF/F) with respect to time totaling 180 seconds. FIG. 34b Calcium transient activity plots of 30 representative neurons from each culture condition. Transient frequency of 194-297 neurons spanning three separate replicates per condition calculated in two individual lines FIG. 34c 83GFP and FIG. 34d 00iCTR. Neuron activity completely ablated with the administration of tetrodotoxin (TTX). FIG. 34e Immunocytochemistry staining of islet1 (ISL1) and smi32 (right) of site previously acquired for live calcium transients using fluro-4 dye (left). ISL1.sup.+ neurons (arrows) superimposed to determine MN firing specificity. FIG. 34f Mean transient frequency of representative active neurons compared to ISL1 overlaid MNs. Charts showing representative transient events from FIG. 34g ISL1 SMI32 double-positive neurons quantified and displayed as raster plot and plotted in FIG. 34h as frequency in hertz (Hz). Means represented by black bars and error bars represent standard error of the mean. Significance calculated by ANOVA, ** denotes P<0.01, ***=P<0.005, and ****=P<0.001.

[0092] FIG. 35A-I Exemplary SC-Chip induces neural differentiation and vascular interaction gene expression. FIG. 35a GFP-MN isolation schematic: nuclear GFP expressing SNPs seeded on top channel with isogenic non-GFP expressing BMECs on bottom. Fluorescence-assisted cell sorting (FACS) conducted on FITC to purify spMNs after BMEC co-culture. FIG. 35b Single population histogram of chips seeded only with non-GFP BMECs (black) or GFP-MNs (green) that defined negative and positive fractions respectively. Number of events displayed on y-axis and FITC intensity on x-axis FIG. 35c Mean RPKM data of purified SC-Chip and EC/SC-Chip cultures normalized to RPKM data from iPSC derived BMECs cultured alone. Canonical markers claudin 5 (CLDN5), occluding (OCLN), Tight junction protein 1 (TJP1), Glut-1 (SLC2A1), von willebrand factor (VWF), Tie2 receptor (TEK), endoglin (ENG), and melanoma cell adhesion molecule (MCAM). FIG. 35d Principle component analysis plots of first two principle components (PC) (top) and PCs 2, and 3 (bottom). Arrows indicate weighting along the axis of each respective PC. FIG. 35e Top 200 ranked genes from each PC displayed as Z-score calculated across all conditions for each row and indicated by colorimetric scale. Each PC was entered into DAVID ontology pathway analysis and top 7 categories listed for each PC. The number of genes (count) in each category is displayed with corrected significance values from DAVID analysis (Bonferonni). FIG. 35f FIG. 35g Differentially expressed genes contributing to the Response to Protein Stimulus and Neural Differentiation DAVID terms enriched in PC2. RPKM data averaged across sample replicates and normalized to 96 W control. (FIG. 35h, FIG. 35i) Differentially expressed genes contributing to Vascular Development and Extracellular Matrix DAVID terms enriched in PC3.

[0093] FIG. 36A-E Exemplary BMECs Induce Vascular Interaction Gene Expression In Chip. FIG. 36a PCA comparing expression data of differentially expressed genes from PC2 and PC3 and including in vivo adult laser captured MNs (green), in vivo fetal spinal cord (purple), and in vitro experimental data. FIG. 36b Heatmap of top ranking genes from Dev-PC2 that describe variance in fetal gene expression. Z-score calculated from Log.sub.2 RPKM data per gene row and displayed by colorimetric scale. FIG. 36c Mean Somatostatin (SST) expression plotted for each condition, error bars represent standard deviation. FIG. 36d Representative images of 96 well (96 W) and SC-Chips co-cultured with BMECs (EC/SC-Chip) immunostained with SST and MN markers SMI32 and ISL1, scale bars=40 microns. FIG. 36e Whole mount image of day 67 human fetal spinal cord (top) immunostained with SST, SMI32 and ISL1. Higher magnification of ventral horn ISL1 positive MN pool (box) co-expressing SST (bottom).

[0094] FIG. 37A-D Shows an illustrative comparison of fetal developmental stages (week 2 through birth) to differentiation of specialized cell types, including Motor Neurons, Endothelial Cells, Skeletal Muscle Cells and Astrocytes and examples of cell staining. FIG. 37A Schematic of relative emergence of cell types that directly interact with developing human MNs in vivo. FIG. 37B Days 53 and 67 human fetal spinal cords immunostained with neuronal marker neuofilament heavy chain (NFH), BMEC marker glucose transporter 1 (GLUT1), and transcription factor ISL1. High magnification of ventral horns (box) show ISL1 positive MN pools (yellow dotted line). FIG. 37C Images of iPSC-derived BMECs cultured in 96-well plate for 6 days and immunostained with GLUT1, ISL1, and tight junction markers zona occludens 1 (ZO-1) and Occludin (OCLDN). FIG. 37D Images of SNPs cultured in 96 W plate and immunostained with markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

[0095] FIG. 38A-C An Exemplary Comparison Of Gene Expression Showing Density Plots And Comparison Of Datasets. FIG. 38A Density plots of 10,001 genes for each sample for transcriptomic analysis. Quantile normalized data (right). FIG. 38B Pierson correlation coefficient analysis of normalized RPKM datasets displayed by colorimetric scale. FIG. 38C Notch pathway genes differentially expressed in the SC-Chip conditions.

[0096] FIG. 39A-B An Exemplary Comparison Of Gene Expression Between Density And Principle Component Analysis (PCA). FIG. 39A Density plots of 10,001 genes of developmental comparison containing adult laser captured MNs (green: peaks on the left), fetal Spinal cord (purple: peaks on the right) and sorted iPSC-derived neurons (black: middle peaks); upper graph. Quantile normalized data (bottom). FIG. 39B PCA analysis of variance across all samples and all 10,001 genes. Adult laser captured MNs (green: upper left dots), fetal Spinal cord (purple: cluster of dots on the left) and sorted iPSC-derived neurons (black: lower left dots). Quantile normalized data (bottom).

[0097] FIG. 40A-B Exemplary schematic illustrations of a brain blood vessel in vivo and a microfluidic chip in vitro. FIG. 40A shows a schematic illustration of the histology of a cross section of a brain blood vessel in vivo depicting a blood vessel lumen formed by and surrounded by endothelial cells (EC), a vascular basement membrane separating ECs from pericytes (PC), which in turn are overlain with astrocytes. Kim, et. al., Journal of Cerebral Blood Flow and Metabolism. 2006. FIG. 40B shows one embodiment of a chip in vitro: 1. Upper channel, where the arrow depicts directional fluid flow. 2. Lower channel, where arrow depicts directional fluid flow. 3. Cell layer in upper channel. 4. Enodothelial cell layer in lower channel. 5. Porous Membrane. 6. Optional vacuum chambers for providing membrane stretch.

[0098] FIG. 41A-D Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating exemplary micrographs of immunofluorescently labeled cells in neuronal and vascular channels. FIG. 41A shows an exemplary overview (looking up through the lower channel (dark-purple staining of cells) into the upper channel (light-green staining of cells)) demonstrating cell coverage of both channels along with cell morphology. FIG. 41B shows an exemplary micrograph of immunofluorescently labeled cell markers in an enlarged view from FIG. 41A where the two channels begin overlapping. The cells on the upper right in the upper channel are labeled light-green for a GFAP marker. The cells on the upper left in the lower channel are stained for a Phalloidin marker (dark-purple) for cells restricted to the vascular channel. FIG. 41C shows an exemplary micrograph of an immunofluorescently labeled cell marker showing ZO-1 tight junctions (outlines) of endothelial cells in the vascular channel. Nuclei are stained with DAPI (blue). FIG. 41D shows an exemplary micrograph of an immunofluorescently labeled cell marker showing light-green for a GFAP marker in the neuronal channel, as an enlarged view from FIG. 41A. Nuclei are stained with DAPI (blue).

[0099] FIG. 42A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary endothelial cells in the Vascular Compartment (lower channel). FIG. 42A schematic illustration of one embodiment of an On-Chip System showing neuronal cells in a neural compartment (upper channel) and cells lining a vasculature forming a simulated blood vessel in the lower channel as one embodiment for a chip as shown in FIG. 42B. FIG. 42B shows a micrograph demonstrating a cross section of one embodiment of a blood brain barrier on-chip as a 3D-Reconstruction of a Confocal Micrograph Image of fluorescently labeled cells, i.e. neuronal progenitors in the upper channel and HBMECs in the lower channel. As shown in the insert, endothelial cell markers (left to right) show immunofluorescent staining of Glut-1, PECAM-1, Occludin-5, and ZO-1.

[0100] FIG. 43A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary pericytes and astrocytes as part of a neural compartment (upper channel). FIG. 43A shows exemplary micrographs of fluorescently labeled cells using markers for demonstrating exemplary pericytes (upper micrograph) labeled for alpha-SMA (green) and astrocytes (lower micrograph) labeled for GFAP (red) on-chips. FIG. 43B shows an exemplary micrograph of a lower power confocal micrograph image of fluorescently labeled cells as in FIG. 43A. Nuclei are stained with DAPI (blue). As shown in the insert, these labeled pericytes and astrocytes are an example of cells illustrated schematically in the neuronal compartment (upper channel) above a vascular channel in one embodiment of a blood brain barrier (BBB) on-chip.

[0101] FIG. 44A-C Exemplary micrographs of differentiating iPSC-derived endothelial cells on-chip and an exemplary chart showing that on-chip iPSC-derived endothelial cells form a tighter cell layer simulating a blood brain endothelial barrier. FIG. 44A An exemplary timeline for providing iPSC-derived endothelial cells on-chip. Day 0: exemplary morphology of an iPSC colony stained with GFP showing nuclei (blue). After expansion of iPSc colonies, Day 8-12 Start HBMECs Differentiation. Day 18-22 HBMECs selection for providing HBMECs. FIG. 44B shows a phase contrast micrograph of Human Brain Microvascular Endothelial Cells (HBMECs). FIG. 44C iPS-derived HBMECs (right blue column) demonstrates higher and more physiologically-relevant TEER values as compared to a brain endothelial cell-line (left grey column). .Math.cm.sup.2 vs. Average TEER (ohm.Math.cm.sup.2).

[0102] FIG. 45 Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cells expressing typical markers of Astrocytes, Pericytes and HMBECs after 5 days on-chip. Middle fluorescent confocal micrograph shows a cross section of cells on a chip. Upper fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary pericytes (SMAred) and astrocytes (GFAPgreen) as part of a neural compartment (upper neuronal channel). Nuclei are stained with DAPI (blue). Lower fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary endothelial cells (iHBMEC (Glut1red) as part of a vascular compartment (lower vascular channel). Nuclei are stained with DAPI (blue).

[0103] FIG. 46A-B A Barrier Function test using Fluorescent Dextran (3 kDa) demonstrates exemplary low diffusion of Dextran indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 46A An exemplary chart showing Dextran Diffusion as % Dextran (fluorescence). The majority of dextrin is in the vascular channel. FIG. 46B An exemplary Barrier Function Schematic illustrating the movement of dextrin from the Vascular Channel to the Neuronal Channel.

[0104] FIG. 47A-F A Barrier Function test demonstrates exemplary low diffusion of human IgG and albumin but not transferrin indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 47A-B Low diffusion of IgG indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47C-D Low diffusion of Albumin indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47E-F High levels of transferrin in the neuronal channel indicates presence of transferrin transporters functionally active on iPS-derived brain endothelial cells. FIGS. 47A, C and E Exemplary charts showing human IgG measured by ELISA, human albumin diffusion as albumin concentration ug/ml, and human transferrn (ELISA), respectively. The majority of human IgG and albumin is in the vascular channel. FIGS. 47B, D and F Illustrate exemplary barrier function schematics illustrating the movement of human IgG, albumin or transferrin, respectively, from the Vascular Channel to the Neuronal Channel.

[0105] FIG. 48 illustrates exemplary materials and methods for one embodiment including factors used in the protocol for the generation of motor neurons (using iPSCs as the starting material).

[0106] FIG. 49 illustrates exemplary materials and methods for one embodiment including various factors used in the protocol for the generation of motor neurons are provided (using iPSCs as the starting material).

DESCRIPTION OF THE INVENTION

[0107] In one embodiment, the invention relates to culturing brain cells and particularly astrocytes together with endothelial cells in a fluidic device under conditions whereby the cells mimic the structure and function of the blood brain barrier and/or spinal cord. Good viability and function allow for measurements of barrier integrity and physiology, whether by trans-epithelial electrical resistance (TEER), patch clamp or other testing measures.

[0108] In one embodiment, the invention relates to culturing endothelial cells (preferably brain-related endothelial cells) and optionally astrocytes, optionally neurons, and optionally pericytes in a fluidic device under conditions whereby the cells mimic one or more structural or functional features (e.g. tight junctions) of the blood brain barrier and/or the spinal cord. Culture of these cells in a microfluidic device, such as a microfluidic chip with flow as herein described, whether alone or in combination with other cells, drives maturation and/or differentiation further than existing systems. For example, a mature electrophysiology of the neurons includes negative sodium channel current, positive potassium channel current, and/or action potential spikes of amplitude, duration and frequency similar to neurons in a physiological environment or when compared to static culture neurons, static culture neurons lack one or more of the aforementioned features. The evidence also supports improved maturation of the astrocytes and BMECs.

[0109] As described herein, astrocytes were observed to send out of processes to contact the endothelial cells. As described herein, improved and sustained barrier function indicates maturation of the BMECs. Good viability and function allow for measurements of barrier integrity and physiology, whether by TEER, permeability assays, patch clamp (or other electrophysiological methods), calcium or voltage imaging, or other testing measures. Observed characteristics of the in vitro BBB-on-chip of the present invention include: (1) tight junctions between endothelial cells (which creates selective permeability to substances); (2) optional cell-to-cell communication exemplified by contact of the endothelial cells with astrocytes (e.g. endfoot contact by partial transmigration of the membrane separating these cells); (3) optional extended neurite projections, (4) optional fluid flow that influences cell differentiation and tight junction formation; and (5) high electrical resistance representing the maturity and integrity of the BBB components. With respect to neurite projections, in one embodiment, the present invention contemplates seeding on nanopatterned surfaces which promote extended and direct (e.g. along a relatively linear path) neurite growth. The preferred nanopattern is linear valleys and ridges, but alternatives such as circular, curved, or any other desired shape or combination thereof are also contemplated. With respect to endothelial cells, in one embodiment, the present invention contemplates BMECs which form a lumen on the chip (for example, completely lining a flow channel, at least for a portion of its length). Among other advantage (e.g. endothelial layer stability) this potentially enables the use of the device with blood or blood components. With respect to selective permeability, the present invention contemplates, in one embodiment, introducing substances in a channel under the BMECs such that at least one substance passes through the BMEC barrier (e.g. BMEC cells on the bottom side of the membrane) and into a channel above the membrane, and detecting said at least one substance (e.g. with antibodies, mass spec, etc.).

[0110] More specifically, cells for use on-chip include but are not limited to brain microvascular endothelial cells, astrocytes, and pericytes. Brain microvascular endothelial cells, for example, may be iPSc derived (e.g. 83i cells from a patient used for some of the experiments as described herein). Astrocytes and pericytes for use on-chips may be obtained as primary cells. In some embodiments, primary astrocytes and primary pericytes may be obtained from explants, i.e. biopsies, of patients, purchased commercially, e.g. from ScienCell (6076 Corte Del Cedro Carlsbad, Calif. 92011). As exemplary primary cells offered by ScienCell, astrocytes are isolated from human brain (cerebral cortex) then cryopreserved at passage one for delivery as a frozen stock. As another example, pericytes are isolated from Human Brain Vascular areas then offered for sale by ScienCell as a frozen cell pellet. One or more of these cell types are added to microfluidic chips. See, FIG. 40A-B Exemplary schematic illustrations of a brain blood vessel in vivo and a microfluidic chip in vitro. FIG. 40A shows a schematic illustration of the histology of a cross section of a brain blood vessel in vivo depicting a blood vessel lumen formed by and surrounded by endothelial cells (EC), a vascular basement membrane separating ECs from pericytes (PC), which in turn are overlain with astrocytes. Kim, et. al., Journal of Cerebral Blood Flow and Metabolism. 2006. FIG. 40B shows one embodiment of a chip in vitro: 1. Upper channel, where the arrow depicts directional fluid flow. 2. Lower channel, where arrow depicts directional fluid flow. 3. Cell layer in upper channel. 4. Enodothelial cell layer in lower channel. 5. Porous Membrane. 6. Optional vacuum chambers for providing membrane stretch. In some embodiments, membrane stretch may be used to mimic movement of blood vessels within a living organism.

[0111] In some embodiments, methods for seeding cells on-chip are provided, as follows for one example: Day1: Activate and Coat Chips, Day 0: Seed HBMECs, Astrocytes and Pericytes; Day 1: Initiate flow of media @60 l/hr. Bottom channel only or Top and bottom channels; Day 2: Initiate permeability assay: Dextran+Serum proteins+Compounds of interest; Day 3: End permeability assay. Collect media and freeze; Day 4: Test permeability modulators/experimental stimuli Day 5: Repeat permeability assay; Day 6: End permeability assayCollect media and freeze. Fix or lyse chips.

[0112] Exemplary embodiments of cells on-chip, e.g. HBMECs, astrocytes and pericytes, are provided in the following figures.

[0113] FIG. 41A-D Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating exemplary micrographs of immunofluorescently labeled cells in neuronal and vascular channels. FIG. 41A shows an exemplary overview (looking up through the lower channel (dark-purple staining of cells) into the upper channel (light-green staining of cells)) demonstrating cell coverage of both channels along with cell morphology. FIG. 41B shows an exemplary micrograph of immunofluorescently labeled cell markers in an enlarged view from FIG. 41A where the two channels begin overlapping. The cells on the upper right in the upper channel are labeled light-green for a GFAP marker. The cells on the upper left in the lower channel are stained for a Phalloidin marker (dark-purple) for cells restricted to the vascular channel. FIG. 41C shows an exemplary micrograph of an immunofluorescently labeled cell marker showing ZO-1 tight junctions (outlines) of endothelial cells in the vascular channel. Nuclei are stained with DAPI (blue). FIG. 41D shows an exemplary micrograph of an immunofluorescently labeled cell marker showing light-green for a GFAP marker in the neuronal channel, as an enlarged view from FIG. 41A. Nuclei are stained with DAPI (blue).

[0114] FIG. 42A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary endothelial cells in the Vascular Compartment (lower channel). FIG. 42A schematic illustration of one embodiment of an On-Chip System showing neuronal cells in a neural compartment (upper channel) and cells lining a vasculature forming a simulated blood vessel in the lower channel as one embodiment for a chip as shown in FIG. 42B. FIG. 42B shows a micrograph demonstrating a cross section of one embodiment of a blood brain barrier on-chip as a 3D-Reconstruction of a Confocal Micrograph Image of fluorescently labeled cells, i.e. neuronal progenitors in the upper channel and HBMECs in the lower channel. As shown in the insert, endothelial cell markers (left to right) show immunofluorescent staining of Glut-1, PECAM-1, Occludin-5, and ZO-1.

[0115] FIG. 43A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary pericytes and astrocytes as part of a neural compartment (upper channel). FIG. 43A shows exemplary micrographs of fluorescently labeled cells using markers for demonstrating exemplary pericytes (upper micrograph) labeled for alpha-SMA (green) and astrocytes (lower micrograph) labeled for GFAP (red) on-chips. FIG. 43B shows an exemplary micrograph of a lower power confocal micrograph image of fluorescently labeled cells as in FIG. 43A. Nuclei are stained with DAPI (blue). As shown in the insert, these labeled pericytes and astrocytes are an example of cells illustrated schematically in the neuronal compartment (upper channel) above a vascular channel in one embodiment of a blood brain barrier (BBB) on-chip.

[0116] FIG. 44A-C Exemplary micrographs of differentiating iPSC-derived endothelial cells on-chip and an exemplary chart showing that on-chip iPSC-derived endothelial cells form a tighter cell layer simulating a blood brain endothelial barrier. FIG. 44A An exemplary timeline for providing iPSC-derived endothelial cells on-chip. Day 0: exemplary morphology of an iPSC colony stained with GFP showing nuclei (blue). After expansion of iPSc colonies, Day 8-12 Start HBMECs Differentiation. Day 18-22 HBMECs selection for providing HBMECs. FIG. 44B shows a phase contrast micrograph of Human Brain Microvascular Endothelial Cells (HBMECs). FIG. 44C iPS-derived HBMECs (right blue column) demonstrates higher and more physiologically-relevant TEER values as compared to a brain endothelial cell-line (left grey column). .Math.cm.sup.2 vs. Average TEER (ohm.Math.cm.sup.2).

[0117] FIG. 45 Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cells expressing typical markers of Astrocytes, Pericytes and HMBECs after 5 days on-chip. Middle fluorescent confocal micrograph shows a cross section of cells on a chip. Upper fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary pericytes (SMAred) and astrocytes (GFAPgreen) as part of a neural compartment (upper neuronal channel). Nuclei are stained with DAPI (blue). Lower fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary endothelial cells (iHBMEC (Glut1red) as part of a vascular compartment (lower vascular channel). Nuclei are stained with DAPI (blue).

[0118] FIG. 46A-B A Barrier Function test using Fluorescent Dextran (3 kDa) demonstrates exemplary low diffusion of Dextran indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 46A An exemplary chart showing Dextran Diffusion as % Dextran (fluorescence). The majority of dextrin is in the vascular channel. FIG. 46B An exemplary Barrier Function Schematic illustrating the movement of dextrin from the Vascular Channel to the Neuronal Channel.

[0119] In some embodiments, an IgG-to-albumin ratio (i.e. Albumin Ratio), see, data in FIG. 47A-F, is used as a diagnostic for clinical relevance. (www.mayomedicallaboratories.com/test-catalog/Overview/33038). As one on-chip example, IgG:Albumin in neuronal channel of one embodiment of a BBB On-Chip=0.005. According to Mayo Clinic resources, an IgG:Albumin in CSF in vivo=0.0-0.21. Thus, the IgG-to-albumin ratio on-chip is within a normal range. An increase in this index in vivo is a reflection of IgG accumulation in the CNS. For one clinical example, an IgG:Albumin ratio is used as a marker for CNS inflammatory diseases, such as Multiple Sclerosis. Thus, in additional embodiments, comparative IgG:Albumin ratios are contemplated for modeling CNS inflammatory diseases on-chip.

[0120] FIG. 47A-F A Barrier Function test demonstrates exemplary low diffusion of human IgG and albumin but not transferrin indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 47A-B Low diffusion of IgG indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47C-D Low diffusion of Albumin indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47E-F High levels of transferrin in the neuronal channel indicates presence of transferrin transporters functionally active on iPS-derived brain endothelial cells. FIGS. 47A, C and E Exemplary charts showing human IgG measured by ELISA, human albumin diffusion as albumin concentration ug/ml, and human transferrn (ELISA), respectively. The majority of human IgG and albumin is in the vascular channel. FIGS. 47B, D and F Illustrate exemplary barrier function schematics illustrating the movement of human IgG, albumin or transferrin, respectively, from the Vascular Channel to the Neuronal Channel.

[0121] In some embodiments, on-chip cells are contemplated for treating with inflammatory agents, such as IL-13, IFN-gamma, etc., for inducing inflammation on-chips. In some embodiments, on-chip cells are subjected to shear forces. In some embodiments, on-chip pericyte and/or astrocyte cells further include neuronal cells. In some embodiments, a panel of compounds are contemplated for use in testing permeability across the blood brain barrier on-chip. In further embodiments, cell cultures on-chip are contemplated to extend up to 14 days.

[0122] Although there is a strong need for a model of the human blood-brain barrier, it is also desirable to develop models of blood-brain barriers of other organisms (not limited to animals). Of particular interest are models of, for example, mouse, rat, dog, and monkey, as those are typically used in drug development. Accordingly, the BBB-on-chip can make advantage of not only human-derived cells but also cells from other organisms. Moreover, although it is preferable that all cell types used originate from the same species (for example, in order to ensure that cell-cell communication is effective), it may be desirable at time to mix species (for example, if a desired cell type is scarce or possess technical challenges).

[0123] In some embodiments, neuronal progenitor cells are MN progenitor cells, including but not limited to spinal MN cells. In fact, one of the discoveries during the development of the present inventions found that spinal cord progenitor motor neuron cells on-chip accelerates neuronal maturation in vitro.

[0124] Some human stem cell derived models of neurodegenerative diseases are limited by the lack of systems for providing mature functioning cells in vitro. For more accurate modeling with clinical applications, in part, large numbers of maturing functional neuronal cells are needed, in addition to having standard culture methods for accurate comparative analysis and drug testing.

[0125] In vivo, developing neurons interact with multiple non-neuronal cell types and extracellular substrates along with receiving extracellular signals as the developing neurons mature and migrate into their biological niche.

[0126] In vivo, during neuronal development, brain microvascular endothelial cells (BMECs) invade the neural tube from the perineural vascular plexus beginning at 4 weeks post-fertilization (PF), preceding astrocyte emergence, and form a primitive blood brain barrier that directly interacts with developing neural tube neurons and progenitors (Kurz, 2009; Engelhardt and Liebner, 2014) (FIG. 37a). During development, vascular angiogenesis and axon neurite outgrowth share common morphogenic mechanisms known as angioneurins (Zacchigna, Lambrechts, et al., 2008, Walchli, Wacker, et al., 2015). Brain Microvascular endothelial cells (BMECs) as described herein, share many common signaling pathways with neurons early in development. However, there is much speculation in relation to their contribution to human neuronal maturation. Thus, concordant developmental timing of the two cell types in vivo, as well as shared developmental signaling pathways supported testing a microfluidic chip containing BMECS derived from iPSCs combined with iPSC-derived MN cultures in vitro.

[0127] Thus, a microfluidic chip was developed for combining iPSC derived neuronal progenitor motor neuron cells with BMECs. In one example, induced pluripotent stem cells were used to derive spinal motor neuron progenitor cells. In another example, induced pluripotent stem cells were used to derive brain microvascular endothelial cells. During the development of the present inventions, neurons cultured alone in a chip microvolume, i.e. one channel, displayed increased calcium transient function and chip-specific gene expression compared to equivalent cell samples cultured in 96 well plates. In contrast, BMECs seeded into a distinct channel in the chip with neuronal progenitor cells seeded into the other channel of a 2-channel microfluidic chip, were seen to directly contact developing neural cells. This cellular interaction associated neural function was further enhanced over the single channel neuronal progenitor cultures as demonstrated by vascular-neural interactions and expression of niche genes. Moreover, transcriptomic comparisons to fetal and adult spinal cord tissue revealed enhanced in vivo-like signatures arising from the chip co-cultures.

[0128] Exemplary embodiments of cells on-chip, e.g. HBMECs and neuronal progenitor cells, are provided in the following figures.

[0129] FIG. 32A-G Exemplary Spinal Neural Progenitor (SNP) Neurons Survive And Mature In The Chip Microenvironment. FIG. 32a Schematic of Spinal neural progenitor (SNP) differentiation from induced pluripotent stem cell (iPSC) cultures. Cells were fated to neural ectoderm (NE) using WNT agonist CHIR99021 and SMAD inhibitors LDN193189 and SB431542 for 6 days then patterned to ventral spinal neurons using Retinoic Acid (RA) and Sonic hedgehog agonist (SAG) in 6-well plates. At day 12, SNPs were frozen, banked and thawed for experiments (Cryo-bank). SNPs were seeded into the top channel of the SC-chip (dotted line) and incubated for 6 days. FIG. 32b Schematic of dual-channel organ chip seeded with SNPs in top channel. FIG. 32c Phosphorylated neurofilament heavy chain (SMI32) is enriched in spinal motor neurons (spMNs) and expressed in cells populating entire top channel. Scale bar=200 microns. FIG. 32d Immunostaining of main channel of the chip of markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP), scale=200 microns (left); 40 microns (right). FIG. 32e Representative image of SC-chip neurons treated with Fluo-4 calcium activated dye and acquired live in FITC channel. FIG. 32f Florescence of individual active neurons normalized to baseline florescence and plotted over time (dF/F). FIG. 32g Raster plot showing detected events of corresponding traces.

[0130] FIG. 33A-H Exemplary Co-culture of iPSC derived MNs and BMECs in SC-Chip. FIG. 33a Schematic of dual-differentiation and seeding paradigm in EC/SC-Chip. Both SNPs and BMECs are generated from human iPSCs and seeded into top and bottom channels respectively. Transverse section of the EC/SC-Chip (right) shows two compartments separated by porous membrane. FIG. 33b Immunostaining of whole SC-Chip at 6-days incubation expressing SMI32 in top channel and ISL1 expressed by BMECs on bottom channel. FIG. 33c Maximum projection images cropped along Z-axis to show top and bottom compartments of seeding end of EC/SC-Chip immunostained with MN markers SMI32 and islet 1 (ISL1), and tight junction marker zona occludens 1 (ZO-1), Scale bar=400 microns (left) 40 microns (right). FIG. 33d Confocal optical reconstruction at along Z axis (top) of SC-Chip exhibits distinct separation of cultures at 6 days separated by porous PDMS membrane. Perspective view (bottom) exhibits confluent layer of BMECs surrounding entire bottom channel, Scale bar=100 microns. FIG. 33e Representative images of cells attached to porous membrane in the main channel (top and bottom planes). SC-Chip was seeded either with SNPs alone in the top channel (SC-Chip) or with the addition of non-GFP BMECs into the bottom compartment (EC/SC-Chip). FIG. 33f High magnification of top compartment show GFP negative, ISL1 positive clusters indicating BMEC infiltration. Quantification of FIG. 33g GFP and FIG. 33h ISL1 populations in top compartment. Error bars represent standard deviation. Plots were determined from 6 individual chips from three culture rounds (dots). No significant difference cultures determined by t-test (ns).

[0131] FIG. 34A-H Exemplary SC-Chip environment increases spontaneous neural activity. FIG. 34a Individual Representative calcium transient activity of SNP (top) and BMEC (bottom) cultures acquired at the seeding compartments plotted by change in florescence intensity over background (dF/F) with respect to time totaling 180 seconds. FIG. 34b Calcium transient activity plots of 30 representative neurons from each culture condition. Transient frequency of 194-297 neurons spanning three separate replicates per condition calculated in two individual lines FIG. 34c 83GFP and FIG. 34d 00iCTR. Neuron activity completely ablated with the administration of tetrodotoxin (TTX). FIG. 34e Immunocytochemistry staining of islet1 (ISL1) and smi32 (right) of site previously acquired for live calcium transients using fluro-4 dye (left). ISL1.sup.+ neurons (arrows) superimposed to determine MN firing specificity. FIG. 34f Mean transient frequency of representative active neurons compared to ISL1 overlaid MNs. Charts showing representative transient events from FIG. 34g ISL1 SMI32 double-positive neurons quantified and displayed as raster plot and plotted in FIG. 34h as frequency in hertz (Hz). Means represented by black bars and error bars represent standard error of the mean. Significance calculated by ANOVA, ** denotes P<0.01, ***=P<0.005, and ****=P<0.001.

[0132] FIG. 35A-I Exemplary SC-Chip induces neural differentiation and vascular interaction gene expression. FIG. 35a GFP-MN isolation schematic: nuclear GFP expressing SNPs seeded on top channel with isogenic non-GFP expressing BMECs on bottom. Fluorescence-assisted cell sorting (FACS) conducted on FITC to purify spMNs after BMEC co-culture. FIG. 35b Single population histogram of chips seeded only with non-GFP BMECs (black) or GFP-MNs (green) that defined negative and positive fractions respectively. Number of events displayed on y-axis and FITC intensity on x-axis FIG. 35c Mean RPKM data of purified SC-Chip and EC/SC-Chip cultures normalized to RPKM data from iPSC derived BMECs cultured alone. Canonical markers claudin 5 (CLDN5), occluding (OCLN), Tight junction protein 1 (TJP1), Glut-1 (SLC2A1), von willebrand factor (VWF), Tie2 receptor (TEK), endoglin (ENG), and melanoma cell adhesion molecule (MCAM). FIG. 35d Principle component analysis plots of first two principle components (PC) (top) and PCs 2, and 3 (bottom). Arrows indicate weighting along the axis of each respective PC. FIG. 35e Top 200 ranked genes from each PC displayed as Z-score calculated across all conditions for each row and indicated by colorimetric scale. Each PC was entered into DAVID ontology pathway analysis and top 7 categories listed for each PC. The number of genes (count) in each category is displayed with corrected significance values from DAVID analysis (Bonferonni). FIG. 35f FIG. 35g Differentially expressed genes contributing to the Response to Protein Stimulus and Neural Differentiation DAVID terms enriched in PC2. RPKM data averaged across sample replicates and normalized to 96 W control. (FIG. 35h, FIG. 35i) Differentially expressed genes contributing to Vascular Development and Extracellular Matrix DAVID terms enriched in PC3.

[0133] FIG. 36A-E Exemplary BMECs Induce Vascular Interaction Gene Expression In Chip. FIG. 36a PCA comparing expression data of differentially expressed genes from PC2 and PC3 and including in vivo adult laser captured MNs (green), in vivo fetal spinal cord (purple), and in vitro experimental data. FIG. 36b Heatmap of top ranking genes from Dev-PC2 that describe variance in fetal gene expression. Z-score calculated from Log.sub.2 RPKM data per gene row and displayed by colorimetric scale. FIG. 36c Mean Somatostatin (SST) expression plotted for each condition, error bars represent standard deviation. FIG. 36d Representative images of 96 well (96 W) and SC-Chips co-cultured with BMECs (EC/SC-Chip) immunostained with SST and MN markers SMI32 and ISL1, scale bars=40 microns. FIG. 36e Whole mount image of day 67 human fetal spinal cord (top) immunostained with SST, SMI32 and ISL1. Higher magnification of ventral horn ISL1 positive MN pool (box) co-expressing SST (bottom).

[0134] FIG. 37A-D Shows an illustrative comparison of fetal developmental stages (week 2 through birth) to differentiation of specialized cell types, including Motor Neurons, Endothelial Cells, Skeletal Muscle Cells and Astrocytes and examples of cell staining. FIG. 37a Schematic of relative emergence of cell types that directly interact with developing human MNs in vivo. FIG. 37b Days 53 and 67 human fetal spinal cords immunostained with neuronal marker neuofilament heavy chain (NFH), BMEC marker glucose transporter 1 (GLUT1), and transcription factor ISL1. High magnification of ventral horns (box) show ISL1 positive MN pools (yellow dotted line). FIG. 37c Images of iPSC-derived BMECs cultured in 96-well plate for 6 days and immunostained with GLUT1, ISL1, and tight junction markers zona occludens 1 (ZO-1) and Occludin (OCLDN). FIG. 37d Images of SNPs cultured in 96 W plate and immunostained with markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

[0135] FIG. 38A-C An Exemplary Comparison Of Gene Expression Showing Density Plots And Comparison Of Datasets. FIG. 38a Density plots of 10,001 genes for each sample for transcriptomic analysis. Quantile normalized data (right). FIG. 38b Pierson correlation coefficient analysis of normalized RPKM datasets displayed by colorimetric scale. FIG. 38c Notch pathway genes differentially expressed in the SC-Chip conditions.

[0136] FIG. 39A-B An Exemplary Comparison Of Gene Expression Between Density And Principle Component Analysis (PCA). FIG. 39a Density plots of 10,001 genes of developmental comparison containing adult laser captured MNs (green: peaks on the left), fetal Spinal cord (purple: peaks on the right) and sorted iPSC-derived neurons (black: middle peaks); upper graph. Quantile normalized data (bottom). FIG. 39b PCA analysis of variance across all samples and all 10,001 genes. Adult laser captured MNs (green: upper left dots), fetal Spinal cord (purple: cluster of dots on the left) and sorted iPSC-derived neurons (black: lower left dots). Quantile normalized data (bottom).

DESCRIPTION OF PREFERRED EMBODIMENTS

[0137] In one embodiment, the present invention contemplates a BBB-on-chip where at least one population of cells is derived from a patient diagnosed with a disorder of the nervous system. While it is not intended that the present invention be limited to a particular CNS disorder, in one embodiment, the disorder is ALS. Amyotrophic lateral sclerosis (ALS) is a severe neurodegenerative condition characterized by loss of motor neurons in the brain and spinal cord. In one embodiment, the present invention contemplates generating induced pluripotent stem cells (iPSCs) from patients with ALS and differentiating them into motor neurons progenitors for seeding on a microfluidic device. There are currently no effective treatments for ALS. In one embodiment, the present invention contemplates the BBB-on-chip as a model system for testing drugs so as to predict success in subsequent clinical trials.

[0138] In another embodiment, the CNS disorder is Parkinson's disease (PD). PD is a neurodegenerative disorder primarily characterized by a loss of dopamine neurons, but which also leads to many other pathological changes.

[0139] In yet another embodiment, the CNS disorder is Alzheimer's disease. Alzheimer's is a type of dementia that causes problems with memory, thinking and behavior. Symptoms usually develop slowly and get worse over time, becoming severe enough to interfere with daily tasks.

[0140] It is contemplated that iPSC technology can be used together with microfluidic chips to mimic patient-specific phenotypes in disease states. For example, in another embodiment, cells derived from patients diagnosed with MCT8-specific thyroid hormone cell-membrane transporter deficiency are contemplated for use in microfluidic devices as at least one of the cellular components of the BBB-on-chip. This disease is characterized by severe cognitive deficiency, infantile hypotonia, diminished muscle mass and generalized muscle weakness, progressive spastic quadriplegia, joint contractures, and dystonic and/or athetoid movement with characteristic paroxysms or kinesigenic dyskinesias. Seizures occur in about 25% of cases. Patients exhibit pathognomonic thyroid test results including high serum 3,3,5-triiodothyronine (T.sub.3) concentration and low serum 3,3,5-triiodothyronine (reverse T.sub.3 or rT.sub.3) concentration. Serum tetraiodothyronine (thyroxine or T.sub.4) concentration is often reduced, but may be within the low normal range; serum TSH concentrations are normal or slightly elevated. SLC16A2 (also known as MCT8) is the only gene in which mutations are known to cause this disorder.

TABLE-US-00001 TABLE 2 Exemplary Experimental Conditions Tested. Apical Apical Density Cells Basal Basal Chip Condition Coating Cell Type Apical Apical Evaluation Basal Coating Basal Density Evalutaion Number 1. Laminin DRY none Poor COL/Fibro 2x BMEC ?? Poor 150 (50 ug/mL) 2. Laminin EZ03 6.00E+06 Success (High Density) COL/Fibro 1x BMEC 2.0E+06 Poor 151 (50 ug/mL) 03iCTR 3. Laminin iMNP25i 2.50E+06 Success (Med Density) COL/Fibro 1x BMEC 2.0E+06 Poor 152 (50 ug/mL) 58iCTR 4. Laminin iNPC 1.50E+06 Poor COL/Fibro 1x BMEC 2.0E+06 Poor 087 (50 ug/mL) 03iCTR 5. Laminin iNPC 1.50E+06 Need eval . . . COL/Fibro 1x BMEC 2.0E+06 Poor 083 (50 ug/mL) 03iCTR 6. Laminin iMNp25i 1.25E+06 Success COL/Fibro 1x BMEC 2.0E+06 Poor 086 (50 ug/mL) (Sphere and endo) 03iCTR 7. Laminin iMNp25iSphere n/a Success (Lowest density) COL/Fibro 1x BMEC 1.0E+06 Success 166 (50 ug/mL) 03iCTR 8. Laminin iMNp25i 1.25E+06 Success (Med Density) COL/Fibro 1x BMEC 1.0E+06 Poor 164 (50 ug/mL) 83iCTR 9. Laminin EZ03 3.00E+06 Success (Med Density) COL/Fibro 1x BMEC 1.0E+06 Poor 165 (50 ug/mL) 58iMCT8 10. Laminin iNPC 3.00E+06 Success (Lowest density) COL/Fibro 1x BMEC 2.0E+06 Poor 116 (50 ug/mL) 58iMCT8 11. Laminin EZ03 1.50E+06 Success (High density) COL/Fibro 1x BMEC 2.0E+06 Poor 295 (50 ug/mL) 58iMCT8 12. Laminin iMNp25i 5.00E+06 Success (Lowest density) COL/Fibro 1x BMEC 2.0E+06 Poor 296 (50 ug/mL) 03iCTR 13. Laminin EZ03 1.20E+07 Success (High density) COL/Fibro 0.2x BMEC 1.2E+07 Fair 281 (50 ug/mL) 03iCTR 14. Laminin iMNp25i 6.00E+06 Success (High density) COL/Fibro 0.2x BMEC 6.0E+06 Poor 291 (50 ug/mL) 03iCTR

TABLE-US-00002 TABLE 3 Exemplary Conditions Tested. Top Concen- Apical Basal Condi- Chip Top Density tration Bottom Cell Conc Function- Function- Flow tions ID Cells (mil/mL Top Coating (ug/mL) Cells Density Coating Basal (ug/mL) alization alization Rate Chips 1 573 03CTR 14E6 Col 320/80 N/A N/A Col 320/80 100 W, 100 W, None 1 iBMEC IV/Fibronectin IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (wet) (Fb) 1:4 (wet) sccm O2 sccm O2 2 574 03CTR 14E6 Col 160/40 N/A N/A Col 160/40 100 W, 100 W, None 1 iBMEC IV/Fibronectin IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (dry) (Fb) 1:4 (dry) sccm O2 sccm O2 3 663 iMNP 14E6 Laminin 1x 50 N/A N/A Laminin 1x 50 100 W, 100 W, None 1 30 s, 15 30 s, 15 sccm O2 sccm O2 4 664 iMNP TBD Laminin 1x 50 iMNP 14E6 Col 320/80 100 W, 100 W, None 1 IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (wet) sccm O2 sccm O2 5 665 iMNP TBD Laminin 1x 50 03CTR 14E6 Col 160/40 100 W, 100 W, None 1 iBMEC IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (dry) sccm O2 sccm O2 6 667 Col 320/80 03CTR 14E6 Col 320/80 100 W, Corona; None 1 IV/Fibronectin iBMEC IV/Fibronectin 30 s, 15 100 W, (Fb) 1:4 (wet) (Fb) 1:4 (wet) sccm O2 30 s, 15 sccm O2 7 668 Col 160/40 03CTR 14E6 Col 160/40 100 W, Corona; None 1 IV/Fibronectin iBMEC IV/Fibronectin 30 s, 15 100 W, (Fb) 1:4 (dry) (Fb) 1:4 (dry) sccm O2 30 s, 15 sccm O2 8 693 Neuron? TBD Laminin 50 03CTR 14E6 Col 320/80 100 W, Corona; None 1 iBMEC IV/Fibronectin 30 s, 15 100 W, (Fb) 1:4 (wet) sccm O2 30 s, 15 sccm O2

TABLE-US-00003 TABLE 4 Exemplary Conditions Tested. Top Concen- Apical Basal Condi- Chip Top Density tration Bottom Cell Conc Function- Function- Flow tions ID Cells (mil/mL Top Coating (ug/mL) Cells Density Coating Basal (ug/mL) alization alization Rate Chips 1 761 03CTR 14E6 Col 320/80 N/A N/A Col 320/80 100 W, 100 W, None 1 iBMEC IV/Fibronectin IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (wet) (Fb) 1:4 (wet) sccm O2 sccm O2 2 762 03CTR 14E6 Col 160/40 N/A N/A Col 160/40 100 W, 100 W, None 1 iBMEC IV/Fibronectin IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (dry) (Fb) 1:4 (dry) sccm O2 sccm O2 3 763 03CTR 14E6 Laminin 1x 50 N/A N/A Laminin 1x 50 100 W, 100 W, None 1 iBMEC 30 s, 15 30 s, 15 sccm O2 sccm O2 4 764 Neuron? TBD Laminin 1x 50 03CTR 14E6 Col 320/80 100 W, 100 W, None 1 iBMEC IV/Fibronectin 30 s, 15 30 s, 15 (Fb) 1:4 (wet) sccm O2 sccm O2 5 765 Col 320/80 03CTR 14E6 Col 320/80 100 W, Corona; None 1 IV/Fibronectin iBMEC IV/Fibronectin 30 s, 15 100 W, (Fb) 1:4 (wet) (Fb) 1:4 (wet) sccm O2 30 s, 15 sccm O2 6 766 Col 160/40 03CTR 14E6 Col 160/40 100 W, Corona; None 1 IV/Fibronectin iBMEC IV/Fibronectin 30 s, 15 100 W, (Fb) 1:4 (dry) (Fb) 1:4 (dry) sccm O2 30 s, 15 sccm O2

[0141] The following references are herein incorporated by reference in their entirety. [0142] Avior, Y., I. Sagi and N. Benvenisty (2016). Pluripotent stem cells in disease modelling and drug discovery. Nat Rev Mol Cell Biol 17(3): 170-182. [0143] Bhatia, S. N. and D. E. Ingber (2014). Microfluidic organs-on-chips. Nat Biotech 32(8): 760-772. [0144] Borghese, L., D. Dolezalova, T. Opitz, S. Haupt, A. Leinhaas, B. Steinfarz, P. Koch, F. Edenhofer, A. Hampl and O. Brstle (2010). Inhibition of Notch Signaling in Human Embryonic Stem Cell-Derived Neural Stem Cells Delays G1/S Phase Transition and Accelerates Neuronal Differentiation In Vitro and In Vivo. STEM CELLS 28(5): 955-964. [0145] Christopherson, K. S., E. M. Ullian, C. C. Stokes, C. E. Mullowney, J. W. Hell, A. Agah, J. Lawler, D. F. Mosher, P. Bornstein and B. A. Barres (2005). Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120(3): 421-433. [0146] Daneman, R., L. Zhou, A. A. Kebede and B. A. Barres (2010). Pericytes are required for blood-brain barrier integrity during embryogenesis. Nature 468(7323): 562-566. [0147] Demestre, M., M. Orth, K. J. Fohr, K. Achberger, A. C. Ludolph, S. Liebau and T. M. Boeckers (2015). Formation and characterisation of neuromuscular junctions between hiPSC derived motoneurons and myotubes. Stem Cell Res 15(2): 328-336. [0148] Domenech, M., H. Yu, J. Warrick, N. M. Badders, I. Meyvantsson, C. M. Alexander and D. J. Beebe (2009). Cellular observations enabled by microculture: paracrine signaling and population demographics. Integrative biology: quantitative biosciences from nano to macro 1(3): 267-274. [0149] Du, Z. W., H. Chen, H. Liu, J. Lu, K. Qian, C. L. Huang, X. Zhong, F. Fan and S. C. Zhang (2015). Generation and expansion of highly pure motor neuron progenitors from human pluripotent stem cells. Nat Commun 6: 6626. [0150] Engelhardt, B. and S. Liebner (2014). Novel insights into the development and maintenance of the blood-brain barrier. Cell and Tissue Research 355(3): 687-699. [0151] Ericson, J., P. Rashbass, A. Schedl, S. Brenner-Morton, A. Kawakami, V. van Heyningen, T. M. Jessell and J. Briscoe (1997). Pax6 controls progenitor cell identity and neuronal fate in response to graded Shh signaling. Cell 90(1): 169-180. [0152] Ferriero, D. M., R. A. Sheldon and R. O. Messing (1994). Somatostatin enhances nerve growth factor-induced neurite outgrowth in PC12 cells. Brain Res Dev Brain Res 80(1-2): 13-18. [0153] Gu, C. and E. Giraudo (2013). The role of semaphorins and their receptors in vascular development and cancer. Experimental Cell Research 319(9): 1306-1316. [0154] Guo, J.-H., W. Ma, J.-W. Yang, Y. Gao, Z. Liang, J. Liu, D.-Y. Wang, T. Luo, J.-R. Cheng and L.-Y. Li (2015). Expression pattern of NeuN and GFAP during human fetal spinal cord development. Child's Nervous System 31(6): 863-872. [0155] Hesselmans, L. F., F. G. Jennekens, C. J. Van den Oord, H. Veldman and A. Vincent (1993). Development of innervation of skeletal muscle fibers in man: relation to acetylcholine receptors. Anat Rec 236(3): 553-562. [0156] Ho, R., S. Sances, G. Gowing, M. W. Amoroso, J. G. O'Rourke, A. Sahabian, H. Wichterle, R. H. Baloh, D. Sareen and C. N. Svendsen (2016). ALS disrupts spinal motor neuron maturation and aging pathways within gene co-expression networks. Nat Neurosci 19(9): 1256-1267. [0157] Huang da, W., B. T. Sherman and R. A. Lempicki (2009). Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4(1): 44-57. [0158] Kaupmann, K., C. Bruns, D. Hoyer, K. Seuwen and H. Lubbert (1993). Distribution and second messenger coupling of four somatostatin receptor subtypes expressed in brain. FEBS Lett 331(1-2): 53-59. [0159] Kurz, H. (2009). Cell lineages and early patterns of embryonic CNS vascularization. Cell Adhesion & Migration 3(2): 205-210. [0160] Lacar, B., S. B. Linker, B. N. Jaeger, S. Krishnaswami, J. Barron, M. Kelder, S. Parylak, A. Paquola, P. Venepally, M. Novotny, C. O'Connor, C. Fitzpatrick, J. Erwin, J. Y. Hsu, D. Husband, M. J. McConnell, R. Lasken and F. H. Gage (2016). Nuclear RNA-seq of single neurons reveals molecular signatures of activation. Nat Commun 7: 11022. [0161] Lee, N. V., M. Sato, D. S. Annis, J. A. Loo, L. Wu, D. F. Mosher and M. L. Iruela-Arispe (2006). ADAMTS1 mediates the release of antiangiogenic polypeptides from TSP1 and 2. Embo j 25(22): 5270-5283. [0162] Lippmann, E. S., A. Al-Ahmad, S. M. Azarin, S. P. Palecek and E. V. Shusta (2014). A retinoic acid-enhanced, multicellular human blood-brain barrier model derived from stem cell sources. Sci Rep 4: 4160. [0163] Lukaszewicz, A. I. and D. J. Anderson (2011). Cyclin D1 promotes neurogenesis in the developing spinal cord in a cell cycle-independent manner. Proceedings of the National Academy of Sciences 108(28): 11632-11637. [0164] Maury, Y., J. Come, R. A. Piskorowski, N. Salah-Mohellibi, V. Chevaleyre, M. Peschanski, C. Martinat and S. Nedelec (2015). Combinatorial analysis of developmental cues efficiently converts human pluripotent stem cells into multiple neuronal subtypes. Nat Biotech 33(1): 89-96. [0165] Montoya, G. J., J. J. Sutachan, W. S. Chan, A. Sideris, T. J. Blanck and E. Recio-Pinto (2009). Muscle-conditioned media and cAMP promote survival and neurite outgrowth of adult spinal cord motor neurons. Exp Neurol 220(2): 303-315. [0166] Moody, W. J. and M. M. Bosma (2005). Ion channel development, spontaneous activity, and activity-dependent development in nerve and muscle cells. Physiol Rev 85(3): 883-941. [0167] O'Rourke, J. G., L. Bogdanik, A. Yanez, D. Lall, A. J. Wolf, A. K. Muhammad, R. Ho, S. Carmona, J. P. Vit, J. Zarrow, K. J. Kim, S. Bell, M. B. Harms, T. M. Miller, C. A. Dangler, D. M. Underhill, H. S. Goodridge, C. M. Lutz and R. H. Baloh (2016). C9orf72 is required for proper macrophage and microglial function in mice. Science 351(6279): 1324-1329. [0168] Palmer, T. D., A. R. Willhoite and F. H. Gage (2000). Vascular niche for adult hippocampal neurogenesis. J Comp Neurol 425(4): 479-494. [0169] Pasterkamp, R. J. (2012). Getting neural circuits into shape with semaphorins. Nat Rev Neurosci 13(9): 605-618. [0170] Patel, T. P., K. Man, B. L. Firestein and D. F. Meaney (2015). Automated quantification of neuronal networks and single-cell calcium dynamics using calcium imaging. J Neurosci Methods 243: 26-38. [0171] Persson, M., D. Stamataki, P. to Welscher, E. Andersson, J. Bse, U. Rther, J. Ericson and J. Briscoe (2002). Dorsal-ventral patterning of the spinal cord requires Gli3 transcriptional repressor activity. Genes & Development 16(22): 2865-2878. [0172] Pfaff, S. L., M. Mendelsohn, C. L. Stewart, T. Edlund and T. M. Jessell (1996). Requirement for LIM Homeobox Gene Is11 in Motor Neuron Generation Reveals a Motor Neuron-Dependent Step in Interneuron Differentiation. Cell 84(2): 309-320. [0173] Przybyla, L. and J. Voldman (2012). Probing embryonic stem cell autocrine and paracrine signaling using microfluidics. Annual review of analytical chemistry (Palo Alto, Calif.) 5: 293-315. [0174] Rezaie, P. and D. Male (1999). Colonisation of the developing human brain and spinal cord by microglia: a review. Microsc Res Tech 45(6): 359-382. [0175] Rushton, D. J., V. B. Mattis, C. N. Svendsen, N. D. Allen and P. J. Kemp (2013). Stimulation of GABA-induced Ca2+ influx enhances maturation of human induced pluripotent stem cell-derived neurons. PLoS One 8(11): e81031. [0176] Sasaki, M., S. Seo-Kiryu, R. Kato, S.-i. Kita and H. Kiyama (2001). A disintegrin and metalloprotease with thrombospondin type1 motifs (ADAMTS-1) and IL-1 receptor type 1 mRNAs are simultaneously induced in nerve injured motor neurons. Molecular Brain Research 89(1-2): 158-163. [0177] Schweitzer, J., T. Becker, J. Lefebvre, M. Granato, M. Schachner and C. G. Becker (2005). Tenascin-C is involved in motor axon outgrowth in the trunk of developing zebrafish. Dev Dyn 234(3): 550-566. [0178] Shen, Q., S. K. Goderie, L. Jin, N. Karanth, Y. Sun, N. Abramova, P. Vincent, K. Pumiglia and S. Temple (2004). Endothelial cells stimulate self-renewal and expand neurogenesis of neural stem cells. Science 304(5675): 1338-1340. [0179] Shen, Q., S. K. Goderie, L. Jin, N. Karanth, Y. Sun, N. Abramova, P. Vincent, K. Pumiglia and S. Temple (2004). Endothelial Cells Stimulate Self-Renewal and Expand Neurogenesis of Neural Stem Cells. Science 304(5675): 1338-1340. [0180] Tuncdemir, S. N., B. Wamsley, F. J. Stam, F. Osakada, M. Goulding, E. M. Callaway, B. Rudy and G. Fishell (2016). Early Somatostatin Interneuron Connectivity Mediates the Maturation of Deep Layer Cortical Circuits. Neuron 89(3): 521-535. [0181] Varnum-Finney, B., K. Venstrom, U. Muller, R. Kypta, C. Backus, M. Chiquet and L. F. Reichardt (1995). The Integrin Receptor (8)(1) Mediates Interactions of Embryonic Chick Motor and Sensory Neurons with Tenascin-C. Neuron 14(6): 1213-1222. [0182] Walchli, T., A. Wacker, K. Frei, L. Regli, M. E. Schwab, S. P. Hoerstrup, H. Gerhardt and B. Engelhardt (2015). Wiring the Vascular Network with Neural Cues: A CNS Perspective. Neuron 87(2): 271-296. [0183] Warp, E., G. Agarwal, C. Wyart, D. Friedmann, C. S. Oldfield, A. Conner, F. Del Bene, A. B. Arrenberg, H. Baier and E. Y. Isacoff (2012). Emergence of patterned activity in the developing zebrafish spinal cord. Curr Biol 22(2): 93-102. [0184] Yu, H., C. M. Alexander and D. J. Beebe (2007). Understanding microchannel culture: parameters involved in soluble factor signaling. Lab Chip 7(6): 726-730. [0185] Zacchigna, S., D. Lambrechts and P. Carmeliet (2008). Neurovascular signalling defects in neurodegeneration. Nat Rev Neurosci 9(3): 169-181.

EXPERIMENTAL

Immunohistochemistry

[0186] Fetal tissue was received from the Birth Defects Research Laboratory at the University of Washington under their approved Institutional review board (IRB), consent and privacy guidelines. Protocols were performed in accordance with the Institutional Review Board's guidelines at the Cedars-Sinai Medical Center under the auspice IRB-SCRO Protocol Pro00021505.

[0187] Spinal cord samples arrived with estimated age and as partially intact spinal columns which were partitioned into approximated cervical, thoracic and lumbar sections. Fetal tissue was subsequently fixed in 4% paraformaldehyde for 48 hr, and placed in 30% sucrose for an additional 24 hour. Finally, spinal cords were embedded in Tissue-Tek OCT (VWR) and sectioned at 25 m using a cryostat (Leica) at 20 C. and directly mounted on glass slides (Fisher Scientific). Tissue sections were permeabilized in cold MeOH for 20 minutes, and blocked in PBS containing 5% normal donkey serum (Sigma, D9663) and 0.25% Triton-X for 1.5 hour, then transferred to primary antibody solution containing mouse anti-GLUT1 (R&D Systems, 1:100) SMI-32P-100 (Covance, 1:1,000), and goat anti-Islet-1 (R&D, AF1387, 1:500), and rabbit anti-NFH (Sigma, N4142, 1:1000), rabbit anti-SIRT (Sigma, SAB4502861, 1:100) and incubated overnight at 4 C. Samples were then incubated for 1 h at room temperature in donkey anti-mouse Alexa Fluor 488 and donkey anti-goat 594 secondary antibodies (Life Technologies, A21202 and A21289, 1:1,000 each). Fetal samples were mounted in Fluoromount-G (Southern Biotech, 0100-01) and acquired at 20 using automated stitching on a Leica DM 6000 microscope for whole mount image.

[0188] iPSC-derived cultures were fixed in 4% paraformaldehyde, and rinsed with PBS. Cultures were permeabilized in 10% Triton X at RT for 10 min and blocked in 5% donkey serum and 0.1% Triton-X at RT for 1 h. Samples were incubated overnight at 4 C. in primary antibody solution containing the following antibodies: mouse anti-SMI-32 (Covance, SMI-32P-100, 1:1,000), rabbit anti-ZO-1 ( ) mouse anti-GLUT1 (GLUT1), mouse anti-Occludin ( ) mouse anti-Claudin5 ( ) rinsed in PBS, and incubated for 1 h at room temperature in donkey anti-mouse Alexa Fluor 488, donkey anti-goat 594, and donkey anti-goat 647 secondary antibodies (Life Technologies, A21202 and A21289, 1:1,000 each).

Imaging and Population

[0189] Confocal images were acquired using an A1 confocal microscope (Nikon) using a Plan Apo 10 objective at 1-micron increments. FIG. 33D was generated from auto-stitching three confocal stacks at 10 (Nikon Elements). Images were rendered with maximum projection and quantified using Imaris (Bitplane). For population and ISL1 nuclei size analysis, images were cropped to only top channel and max projected using IMARIS software. Sites were prioritized as containing maximal GFP positive nuclei in the frame. GFP positive nuclei were quantified using spots algorithm and ISL1 positive nuclei were quantified by filtering by GFP co-localization.

Transcriptomic Analysis

[0190] BMECs were seeded in endothelial cell media into either a T75 flask, or the bottom channel of the SC-chip. SNPs were thawed and seeded into either the 96 well plates or the top channel of the Chip and incubated overnight. The following day, media was replaced with Stage 3 media in all conditions. BMEC flask was washed 2 with minimal neural media (IMDM:F12 0.5% N2, 1% B27, 0.5 NEAA, 1 PSA). 24 hours later BMEC media was collected, centrifuged and filtered. BMEC conditioned media was then supplemented with remaining ingredients for Stage 3 media. Media for all conditions was replaced every two days. At 6 days post-seeding, samples were dissociated with Accutase (Sigma-Aldrich). 3 Chip or 3 wells of a 96 well plate were pooled for each experimental replicate for a total of three replicates. Pooled samples were washed in PBS, resuspended in cold MACS buffer (Miltenyl Biotech), and filtered using a 40-micron screen. Cells were sorted using an Influx FACS sorter (BD). GFP positive gating was established using SC-chip seeded exclusively with 83iGFP MNs for a positive control, and 83iCTR BMECs for a negative control. Samples were sorted using this established gating. Positive fraction cell pellets were frozen at 80 C. until processing. For population counts at day 30, both channels were dissociated as before, and one Chip per replicate was quantified by number of GFP positive events.

Live Calcium Transient Imaging and Analysis

[0191] Fluo-4 calcium dye (Invitrogen), was prepared at 10 mM in 50% Pluronic solution and DMSO, and diluted to a final concentration of 20 M in ECS. Tissue cultures were incubated at RT for 30 minutes, then washed in fresh ECS and incubated an additional 30 minutes before acquiring. After a 2-minute burn in phase, 16 bit videos were acquired for 3 minutes at 20 Hz on an Eclipse Ti microscope (Nikon) using a Plan Flor 20 objective (Nikon) equipped with an Orca-Flash4.0LT digital camera (Hamamatsu). As no difference in event detection was determined between 16- and 8-bit data, all datasets were down-sampled to 8-bit (ImageJ). Calcium activity was counted by tracing ROI in blinded fashion, i.e. the treatment conditions were not associated with the results, and creating masks for use in extracting intensity data (ImageJ). 20-50 neurons were counted per site, and at least 3 Chip or wells were included for each condition. A total of 148-400 neurons per condition were analyzed using MATLAB. dF/F traces were extracted through FluoroSNNAP (Patel, Man et al. 2015). Automated calcium event detection was accomplished through template libraries described previously at a threshold of 0.05. Events were then curated manually by a blinded counter, in other words a person evaluated the events without knowing the associated treatment conditions.

Example 1

[0192] Human induced pluripotent stem cells were derived and maintained. Cells are prepared either directly from cultured iPSCs or from frozen lots of pre-differentiated cells.

[0193] In the case of MNs, for example, cells are seeded at day 12 of differentiation either from freshly differentiated cultures or directly from a thawed vial. iPSC-derived forebrain neural progenitor cultures (dubbed EZs) were cultured in chip either dissociated or as neural spheres that attached and extended in 3 dimensions (See FIG. 3A-B apical). The various factors used in the protocol (see above chart and tabs) for the generation of motor neurons are provided (using iPSCs as the starting material).

[0194] A. Spinal Motor Neuron Progenitors (SNP).

[0195] More specifically, spinal motor neuron derivation was accomplished using a modification of a previously published protocol (i.e. Du, Chen et al. 2015). Briefly, iPSCs were mechanically passaged and maintained at low density. IPSCs were harvested then seeded into chips. Differentiation was induced for 6 days in neural induction media (days 0-6) consisting of IMDM/F12 (Gibco), B27, N2, 1% NEAA, 0.2 uM LDN193189, 10 uM SB431542, 3 uM CHIR99021 (Cayman Chemical); and Penicllin-Streptomycin-Amphotericin B (PSA).

[0196] Cells were then passaged using Accutase (Sigma-Aldrich) and reseeded onto matrigel and patterned in Stage 2 media (days 6-12) consisting of IMDM/F12 (Gibco), B27, N2, 1% NEAA, 200 ng/ml ascorbic acid, 0.1 uM retinoic acid, 1 uM SAG and Penicllin-Streptomycin-Amphotericin B (PSA) for an additional 6 days. Cells were then dissociated and cryogenically frozen for later use. At day 12 of culture cells are labeled ventral spinal neuron progenitors (SNP) when cultured using the method described in this and the above paragraph.

[0197] At day 12, SNPs may be frozen, banked and then thawed when used for seeding chips or used to directly seed chips, see Example 6, below.

[0198] B. BMEC Differentiation.

[0199] Cells are thawed (or dissociated fresh) and seeded into the chip at day 8-9 (in the case of BMECs differentiation) and at various points in neural differentiation. BMECs were differentiated as described previously (Lippmann, Al-Ahmad et al. 2014).

[0200] FIG. 48 illustrates exemplary materials and methods including factors used in the protocol for the generation of motor neurons (using iPSCs as the starting material).

Example 2

[0201] In this example, another protocol for the generation of motor neurons is provided using iPSCs as the starting material.

[0202] FIG. 49 illustrates exemplary materials and methods for one embodiment including various factors used in the protocol for the generation of motor neurons are provided (using iPSCs as the starting material).

Example 3

[0203] This example explores various conditions tested for seeding neural (EZ spheres and iMNPs) and endothelial cells (iBMECs) from frozen stocks of cells on surfaces treated with different extracellular matrices (ECMs). The best results for iBMECs were achieved with a mixture of collagen and fibronectin (4:1 ratio) using a seeding concentration of 510.sup.6 cells/ml (Table 1). Given these results, seeding was attempted on microfluidic devices, i.e. chips. Table 2 shows various conditions tested for seeding neural (EZ spheres, iNPCs and iMNPs) and endothelial cells (iBMECs) on the apical and basal sides of a microfluidic chip using frozen stocks of cells.

[0204] While a variety of protocols were explored, one embodiment for preparing and seeding a microfluidic chip comprising six steps. FIG. 1 shows the workflow. First, the chip (or regions thereof) are treated to promote wetting or protein adhesion (e.g. by plasma treatment). One or more channels are then plugged (see the top schematic of FIG. 2, where an X indicates a channel is blocked in a microfluidic device or chip with top and bottom channels). FIG. 2 (bottom schematic) shows how the ports of a microfluidic device can be utilized to introduce fluid (e.g. with ECMs) or cells using pipette tips. Using the protocol, the ECM mixture for the bottom channel is introduced first, with the excess removed, and the remainder dried. Thereafter (step 3), the ECM for the top channel is introduced. The BMECs can be seeded on the bottom channel. The top channel can be washed. Finally, the neural cells can be introduced and incubated for attachment. Cultures were seeded into chips following the seeding of BMECs described above either on the same day or the following day after BMECs had been seeded onto the chip. The chips were cultured for 14 days and fixed and stained for relevant markers. Confocal imaging shows the transmigration in z-stack.

[0205] FIG. 3A-B provides a microscopic analysis of Chip 166 from Table 2, showing neural cells in the top channel of the microfluidic device (left) and BMECs on the bottom channel of the microfluidic device (right). The neural cells and BMECs have attached.

[0206] The attached cells were then tested for markers to confirm their identity. FIG. 4A-C is a vertical 2D projection of a 3D confocal stack of images slices, which allows for visualization of the neurons and endothelial cells together, even though they are not in the same plane on the microfluidic device. The BMECs display the Glut 1 marker, while the neurons are positive for NFH. DAPI was used to stain the nuclei.

[0207] FIG. 5 provides an image from a microfluidic chip wherein at least a portion of an apical astrocyte (i.e. the endfoot) has transmigrated the membrane and contacted the BMECs on the other side. The astrocytes are shown in white against the red stained BMECs.

Example 4

[0208] The present invention contemplates, in one embodiment, utilizing nanopatterned surfaces for seeding cells. FIG. 6A-B shows a first image (top) where iMNPs were seeded on a plain (un-patterned) surface, as well as a second image (bottom) where the same cells were seeded on a nanopatterned surface. Clearly, the nanopatterned surface results in directed neurite growth (e.g. in a line pattern)

Example 5

[0209] While frozen stocks of cells can be used (particular for the neural cells), it was found that better results can be obtained (particularly for BMECs) when fresh cells are used for seeding. FIG. 7A-D show microscopic examination of the morphology of fresh (not frozen) BMECs seeded on a mixture of collagen and fibronectin that has either been dried (FIG. 7A, top left) or remained wet (FIG. 7B, top right), as well as an example where the same fresh cells were seeded on laminin (FIGS. 7C and D). Interestingly, when laminin was used, the BMECs were not free of neurons (see the arrow in FIG. 7D indicating contamination of the cells with neurons).

[0210] Tables 3 and 4 show various conditions tested for seeding fresh neural (iMNPs) and fresh endothelial cells (iBMECs), where the particular conditions are associated by microfluidic chip number, allowing for a correlation of good tight junctions with the seeding conditions. Staining results (not shown) for microfluidic chip 574 (see Table 3 for conditions) indicated the cells are positive for Glut 1 (red stain), which is a marker of BMEC tight junctions (the nuclei were also stained blue from DAPI). The seeding conditions for chip 574 resulted in good tight junctions. Staining results (not shown) for microfluidic chip 665 (see Table 3 for conditions) indicated that the cells are positive for Glut 1. Thus, the seeding conditions for chip 665 also resulted in good tight junctions. Staining results (not shown) for microfluidic chip 667 (see Table 3 for conditions) indicated the cells are positive for Glut 1. Thus, the seeding conditions for chip 667 resulted in good tight junctions. Staining results for microfluidic chip 693 (see Table 3 for conditions) indicated the cells are positive for Glut 1. Thus, the seeding conditions for chip 693 resulted in good tight junctions.

[0211] Staining results (not shown) for microfluidic chip 733 (see Table 4 for conditions) indicated the cells are positive for Glut 1. The results (not shown) also revealed that coating with laminin alone (before seeding) results in poor BMEC tight junction formation.

Example 6

[0212] Seeding Chips Using Day 12 Spinal Neural Progenitor (SNP) Cells.

[0213] SNPs (fresh or thawed) were seeded into the top channel of the SC-chip and incubated for 6 days. SNPS or MNs were cultured in Stage 3 media (after day 12) consisting of IMDM/F12 (Gibco), B27, N2, 1% NEAA, 200 ng/ml ascorbic acid, 0.5 uM retinoic acid, 0.1 uM cAMP, 0.1 uM SAG, 10 ng/ml GDNF, 10 ng/ml BDNF, 1% PSA and fed every two days.

[0214] After 6 days, cultures underwent immunostaining of cells in the main channel of the chip, including but not limited to markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

[0215] FIG. 32A-G Exemplary Spinal Neural Progenitor (SNP) Neurons Survive And Mature In The Chip Microenvironment. FIG. 32a Schematic of Spinal neural progenitor (SNP) differentiation from induced pluripotent stem cell (iPSC) cultures. Cells were fated to neural ectoderm (NE) using WNT agonist CHIR99021 and SMAD inhibitors LDN193189 and SB431542 for 6 days then patterned to ventral spinal neurons using Retinoic Acid (RA) and Sonic hedgehog agonist (SAG) in 6-well plates. At day 12, SNPs were frozen, banked and thawed for experiments (Cryo-bank). SNPs were seeded into the top channel of the SC-chip (dotted line) and incubated for 6 days. FIG. 32b Schematic of dual-channel organ chip seeded with SNPs in top channel. FIG. 32c Phosphorylated neurofilament heavy chain (SMI32) is enriched in spinal motor neurons (spMNs) and expressed in cells populating entire top channel. Scale bar=200 microns. FIG. 32d Immunostaining of main channel of the chip of markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP), scale=200 microns (left); 40 microns (right). FIG. 32e Representative image of SC-chip neurons treated with Fluo-4 calcium activated dye and acquired live in FITC channel. FIG. 32f Florescence of individual active neurons normalized to baseline florescence and plotted over time (dF/F). FIG. 32g Raster plot showing detected events of corresponding traces.

Example 7

[0216] Seeding Chips Using Day 12 Spinal Neural Progenitor (SNP) Cells and BMECs in SC-Chip.

[0217] Tall channel microphysiological systems (Emulate Inc.) were treated with plasma in 100% oxygen for 2:00 minutes and immediately coated with matrigel for the neural channel and a mixture of collagen IV (Invitrogen), Fibronectin (Invitrogen), and diluted in water in a ratio of 1:4:5 respectively, and incubated overnight at 37 C. and 5% CO.sub.2. Chips were seeded with MNs and BMECs on the same day sequentially by flipping the Chip and allowing cells to attach to the membrane by gravity. Chips were fed with approximately 25 uL of Stage 3, with an extra 25 uL in reservoir tips and media was replaced every other day.

[0218] FIG. 33A-H Exemplary Co-culture of iPSC derived MNs and BMECs in SC-Chip. FIG. 33a Schematic of dual-differentiation and seeding paradigm in EC/SC-Chip. Both SNPs and BMECs are generated from human iPSCs and seeded into top and bottom channels respectively. Transverse section of the EC/SC-Chip (right) shows two compartments separated by porous membrane. FIG. 33b Immunostaining of whole SC-Chip at 6-days incubation expressing SMI32 in top channel and ISL1 expressed by BMECs on bottom channel. FIG. 33c Maximum projection images cropped along Z-axis to show top and bottom compartments of seeding end of EC/SC-Chip immunostained with MN markers SMI32 and islet 1 (ISL1), and tight junction marker zona occludens 1 (ZO-1), Scale bar=400 microns (left) 40 microns (right). FIG. 33d Confocal optical reconstruction at along Z axis (top) of SC-Chip exhibits distinct separation of cultures at 6 days separated by porous PDMS membrane. Perspective view (bottom) exhibits confluent layer of BMECs surrounding entire bottom channel, Scale bar=100 microns. FIG. 33e Representative images of cells attached to porous membrane in the main channel (top and bottom planes). SC-Chip was seeded either with SNPs alone in the top channel (SC-Chip) or with the addition of non-GFP BMECs into the bottom compartment (EC/SC-Chip). FIG. 33f High magnification of top compartment show GFP negative, ISL1 positive clusters indicating BMEC infiltration. Quantification of FIG. 33g GFP and FIG. 33h ISL1 populations in top compartment. Error bars represent standard deviation. Plots were determined from 6 individual chips from three culture rounds (dots). No significant difference cultures determined by t-test (ns). FIG. 34A-H: SC-Chip environment increases spontaneous neural activity.

[0219] FIG. 34A-H Exemplary SC-Chip environment increases spontaneous neural activity. FIG. 34a Individual Representative calcium transient activity of SNP (top) and BMEC (bottom) cultures acquired at the seeding compartments plotted by change in florescence intensity over background (dF/F) with respect to time totaling 180 seconds. FIG. 34b Calcium transient activity plots of 30 representative neurons from each culture condition. Transient frequency of 194-297 neurons spanning three separate replicates per condition calculated in two individual lines FIG. 34c 83GFP and FIG. 34d 00iCTR. Neuron activity completely ablated with the administration of tetrodotoxin (TTX). FIG. 34e Immunocytochemistry staining of islet1 (ISL1) and smi32 (right) of site previously acquired for live calcium transients using fluro-4 dye (left). ISL1.sup.+ neurons (arrows) superimposed to determine MN firing specificity. FIG. 34f Mean transient frequency of representative active neurons compared to ISL1 overlaid MNs. Charts showing representative transient events from FIG. 34g ISL1 SMI32 double-positive neurons quantified and displayed as raster plot and plotted in FIG. 34h as frequency in hertz (Hz). Means represented by black bars and error bars represent standard error of the mean. Significance calculated by ANOVA, ** denotes P<0.01, ***=P<0.005, and ****=P<0.001.

[0220] FIG. 35A-I Exemplary SC-Chip induces neural differentiation and vascular interaction gene expression. FIG. 35a GFP-MN isolation schematic: nuclear GFP expressing SNPs seeded on top channel with isogenic non-GFP expressing BMECs on bottom. Fluorescence-assisted cell sorting (FACS) conducted on FITC to purify spMNs after BMEC co-culture. FIG. 35b Single population histogram of chips seeded only with non-GFP BMECs (black) or GFP-MNs (green) that defined negative and positive fractions respectively. Number of events displayed on y-axis and FITC intensity on x-axis FIG. 35c Mean RPKM data of purified SC-Chip and EC/SC-Chip cultures normalized to RPKM data from iPSC derived BMECs cultured alone. Canonical markers claudin 5 (CLDN5), occluding (OCLN), Tight junction protein 1 (TJP1), Glut-1 (SLC2A1), von willebrand factor (VWF), Tie2 receptor (TEK), endoglin (ENG), and melanoma cell adhesion molecule (MCAM). FIG. 35d Principle component analysis plots of first two principle components (PC) (top) and PCs 2, and 3 (bottom). Arrows indicate weighting along the axis of each respective PC. FIG. 35e Top 200 ranked genes from each PC displayed as Z-score calculated across all conditions for each row and indicated by colorimetric scale. Each PC was entered into DAVID ontology pathway analysis and top 7 categories listed for each PC. The number of genes (count) in each category is displayed with corrected significance values from DAVID analysis (Bonferonni). FIG. 35f FIG. 35g Differentially expressed genes contributing to the Response to Protein Stimulus and Neural Differentiation DAVID terms enriched in PC2. RPKM data averaged across sample replicates and normalized to 96 W control. (FIG. 35h, FIG. 35i) Differentially expressed genes contributing to Vascular Development and Extracellular Matrix DAVID terms enriched in PC3.

[0221] FIG. 36A-E Exemplary BMECs Induce Vascular Interaction Gene Expression In Chip. FIG. 36a PCA comparing expression data of differentially expressed genes from PC2 and PC3 and including in vivo adult laser captured MNs (green), in vivo fetal spinal cord (purple), and in vitro experimental data. FIG. 36b Heatmap of top ranking genes from Dev-PC2 that describe variance in fetal gene expression. Z-score calculated from Log.sub.2 RPKM data per gene row and displayed by colorimetric scale. FIG. 36c Mean Somatostatin (SST) expression plotted for each condition, error bars represent standard deviation. FIG. 36d Representative images of 96 well (96 W) and SC-Chips co-cultured with BMECs (EC/SC-Chip) immunostained with SST and MN markers SMI32 and ISL1, scale bars=40 microns. FIG. 36e Whole mount image of day 67 human fetal spinal cord (top) immunostained with SST, SMI32 and ISL1. Higher magnification of ventral horn ISL1 positive MN pool (box) co-expressing SST (bottom).

[0222] FIG. 37A-D Shows an illustrative comparison of fetal developmental stages (week 2 through birth) to differentiation of specialized cell types, including Motor Neurons, Endothelial Cells, Skeletal Muscle Cells and Astrocytes and examples of cell staining. FIG. 37a Schematic of relative emergence of cell types that directly interact with developing human MNs in vivo. FIG. 37b Days 53 and 67 human fetal spinal cords immunostained with neuronal marker neuofilament heavy chain (NFH), BMEC marker glucose transporter 1 (GLUT1), and transcription factor ISL1. High magnification of ventral horns (box) show ISL1 positive MN pools (yellow dotted line). FIG. 37c Images of iPSC-derived BMECs cultured in 96-well plate for 6 days and immunostained with GLUT1, ISL1, and tight junction markers zona occludens 1 (ZO-1) and Occludin (OCLDN). FIG. 37d Images of SNPs cultured in 96 W plate and immunostained with markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

[0223] FIG. 37A-D Shows an illustrative comparison of fetal developmental stages (week 2 through birth) to differentiation of specialized cell types, including Motor Neurons, Endothelial Cells, Skeletal Muscle Cells and Astrocytes and examples of cell staining. FIG. 37a Schematic of relative emergence of cell types that directly interact with developing human MNs in vivo. FIG. 37b Days 53 and 67 human fetal spinal cords immunostained with neuronal marker neuofilament heavy chain (NFH), BMEC marker glucose transporter 1 (GLUT1), and transcription factor ISL1. High magnification of ventral horns (box) show ISL1 positive MN pools (yellow dotted line). FIG. 37c Images of iPSC-derived BMECs cultured in 96-well plate for 6 days and immunostained with GLUT1, ISL1, and tight junction markers zona occludens 1 (ZO-1) and Occludin (OCLDN). FIG. 37d Images of SNPs cultured in 96 W plate and immunostained with markers of spMNs SMI32, nuclear marker islet1 (ISL1), Beta 3 tubulin (TUBB3), NKX6.1, neurofilament marker microtubule-associated protein 2 (MAP2), and synaptic marker synaptophysin (SYNP).

[0224] FIG. 38A-C An Exemplary Comparison Of Gene Expression Showing Density Plots And Comparison Of Datasets. FIG. 38a Density plots of 10,001 genes for each sample for transcriptomic analysis. Quantile normalized data (right). FIG. 38b Pierson correlation coefficient analysis of normalized RPKM datasets displayed by colorimetric scale. FIG. 38c Notch pathway genes differentially expressed in the SC-Chip conditions.

[0225] FIG. 39A-B An Exemplary Comparison Of Gene Expression Between Density And Principle Component Analysis (PCA). FIG. 39a Density plots of 10,001 genes of developmental comparison containing adult laser captured MNs (green: peaks on the left), fetal Spinal cord (purple: peaks on the right) and sorted iPSC-derived neurons (black: middle peaks); upper graph. Quantile normalized data (bottom). FIG. 39b PCA analysis of variance across all samples and all 10,001 genes. Adult laser captured MNs (green: upper left dots), fetal Spinal cord (purple: cluster of dots on the left) and sorted iPSC-derived neurons (black: lower left dots). Quantile normalized data (bottom).

Example 8

[0226] Unlike conventional static cultures, the present invention contemplates microfluidic devices where the cells are exposed to a constant flow of media providing nutrients and removing waste. FIG. 8 is a photograph showing one embodiment of a system for introducing flow into the microfluidic chips. A plurality of fluid reservoirs are in fluidic communication with a corresponding plurality of microfluidic chips via inlet ports, with tubing coming from the exit ports and attached to a plurality of syringes used to draw fluid through the chip at a flow rate. FIG. 9A-D comprises photographs of microscopic examination of cell morphology on the bottom (left-hand side) and top (right-hand side) of the membrane in a microfluidic device where the cells have been exposed to flow (using the system of FIG. 8) for a number of days. FIG. 10 shows fluorescent staining of cells in a microfluidic device where the cells have been exposed to flow (using the system of FIG. 8) for a number of days. The image is a 3D image of the BMEC in the bottom channel showing a complete contiguous BMEC lumen being formed in the chip. FIG. 11 is a photograph of fluorescent staining of cells showing the presence of neural stem cells (red) in addition to neurites (green), with the nuclei stained with DAPI.

Example 9

[0227] Good cell viability and function on the BB-on-chip allow for measurements of barrier integrity and physiology, whether by TEER, patch clamp or other testing measures.

[0228] TEER: FIG. 12A shows the results/readout from transepithelial electrical resistance (TEER) measurements on the microfluidic chip under flow, static, and control conditions. Cells were plated on tall channel PDMS chips equipped with incorporated gold electrodes on each channel (see FIG. 13A). Post seeding of BMECs, transendothelial electrical resistance was measured by connecting the electrodes to an EVOM2 voltohmmeter (see FIG. 13B). FIG. 12A displays preliminary data indicating the beneficial effect of flow in the BBB-on-chip, i.e. higher TEER in response to flow. In particular, at around the 40 hour time point, the TEER value observed for a BBB-on-chip under flow was significantly higher than a similar chip under static conditions, i.e. that the iPS brain microvascular endothelial cells (iBMECs) formed tighter cell-cell junctions or barrier function under flow conditions on a prototype TEER-Chip as compared to a chip maintained in static culture. The damaged chip was a failure due to the TEER-Chips being a prototype. FIG. 12B shows TEER results where the cells were cultured in transwells.

[0229] FIG. 14A was a follow-up experiment on another round of prototype TEER chips that showed iBMEC barrier function increasing in the presence of flow on a chip followed by a weakening of barrier function with the exposure of the chips to TNFa, a proinflammatory cytokine. Higher TEER values generally indicate a tighter barrier, which is typically desirable for a blood-brain barrier.

[0230] PATCH CLAMP: FIG. 18A-B shows electrophysiology recordings collected by patch-clamp from neurons in the microfluidic device (BBB-on-Chip). These measurements on-chip can be used to provide an indication of neuronal maturation, i.e. more precisely describe the maturity of a neuronal cell. Cells were cultured as described above in a specially designed openable chip (where the chips can be partially disassembled to expose directly cells on the semi-porous membrane) with a stiff PET membrane to aid in patch-clamp recording. PDMS was attempted, but was unsuccessful. PET membrane chips were opened at endpoint at 6 and 24 days in chip. Individual neurons seeded into the chip were directly accessed with a glass micropipette, and cell electrophysiology was recorded including capacitance, membrane resting voltage, spontaneous activity and induced activity. FIG. 18A-B is a whole cell patch recording of an induced action potential from a neuron cultured on the chip. An arrow (FIG. 18A) indicates single action potential. Current recordings (FIG. 18B, right) show negative sodium channel currents (Na.sup.+) and positive potassium channel (K.sup.+) are necessary for normal neuron function and become more pronounced as a neuron matures.

[0231] CALCIUM FLUX: FIG. 19A-D show the results of calcium flux imaging in the neural channel. Using a florescent calcium influx-activated dye (Fluo-4), neurons seeded in chip were imaged using high resolution high frame-rate camera. Florescence intensity changes of up to hundreds of neurons were analyzed simultaneously by recording average pixel intensity over time (dF/F). These values were plotted with respect to time and are analyzed for waveform properties, which correlate spontaneous neural activity and neural network formation. This is accomplished through multi-step video post-processing and signal analysis (including video compression, signal cleanup, automatic or manual ROI detection, etc. which can be implemented from open-source MATLAB software packages). The photograph (FIG. 19A, top left) is a single fluorescent image from a movie of such images. The colored circles indicate the positions that correspond to the time traces in the 3 graphs. The traces show that it is possible to observe neuronal function in the microfluidic chips in real-time. In this case, it is shown that Ca2+ fluxes can be measured on chips to give a direct readout of neuronal activity. The addition of tetrodotoxin (TTX), which is a potent blocker of voltage-gated calcium channels, ablates this activity (FIG. 19D, bottom right). This type of experiment will be important when the neuronal activity is modulated by pharmacological stimulation.

[0232] ICC overlay data: By overlaying images taken after staining the cells, specific cell identification can be combined with original activity traces to determine specific activities of individual cell types in the chip. The overlay data (not shown) indicates that motor neurons are indeed more active in the chip. This can also be accomplished with cell type specific reporter lines.

Example 10

[0233] Brain microvascular endothelial cells (BMECs) constitute the blood-brain barrier (BBB) which forms a dynamic interface between the blood and the central nervous system (CNS) in vivo. This highly specialized interface restricts paracellular diffusion of fluids and solutes including chemicals, toxins and drugs from entering the brain. In this example, fluorescein sodium is used in a paracellular permeability assay of the BMECs seeded on a microfluidic device.

[0234] Albumin or Dextran conjugated to a fluorescent probe (e.g., FITC or TRITC) are frequently used to monitor changes in leakage, and thus barrier function. In this case, Dextran-FITC, a green fluorescent molecule of 4 KDa, or sodium fluorescein (a 0.3 KDa molecule), was added to the bottom (blood side) channel. Paracellular permeability was calculated by measuring the permeability of the fluorescent molecule on the Top (brain side) channel. Low permeability is an indication for proper barrier functions. FIG. 14B involves TNF-alpha exposure, but the readout is membrane permeability as measured by Dextran-FITC. FIG. 14B confirms that TNFa exposure results in a decrease in barrier function and TEER by an increase in permeability through the semi-porous membrane by dextran-FITC, a fluorescently labeled small molecule.

[0235] FIG. 15 shows the results for (and structure of) fluorescein sodium from a paracellular permeability assay. Chips were seeded with iPSC-derived BMECs taken from healthy controls (CTR) or MCT8-deficient patients, and the paracellular permeability was determined by monitoring Blood to brain permeability of the sodium fluorescein tracer as described above. Flow is clearly important.

[0236] In the present experiment, the agent used was fluorescein. In some aspects of the present invention, it is contemplated that similar testing will be done to ascertain permeability for various additional agents (e.g. drugs, chemicals, hormones, blood components, biomarkers). Such methods can allow qualitative or quantitative estimation of the permeability of the in vivo blood-brain barrier to the one or more agents. Furthermore, according to some aspects of the present invention, the permeability of one agent is measured in response to a second agent, treatment or experimental condition (for example, measuring the effect of a medication on the blood-brain barrier permeability of another medication). It is important to note that although we refer to permeability, we do not mean to exclude active transport, pumping or any other means for an agent to pass from one side of the barrier to the other (regardless of direction). The penetration of an agent through the barrier can be measured, for example, using mass spectroscopy, antibody-based methods (e.g. ELISAs, Western blots, bead-based assays), or optical methods (e.g. fluorescence signature, Raman spectroscopy, absorbance).

Example 11

[0237] Some embodiments include blood or blood components, optionally perfused through one or more fluidic channels within the device. The use of blood of blood components is desired as the blood or blood components can improve BBB-on-chip function, for example, by providing biochemical cues, or conversely hurt the BBB-on-chip, for example, because the blood may contain a harmful agent that may be under investigation. In some aspects, permeability assays include blood or blood components in order to provide a potentially more in vivo like result. In other aspects, individual-specific blood or blood components are used in order to potentially provide individualized BBB-related measures. This can include, for example, the measurement of the permeability of one or more agents or components from the blood or components, the effect of the blood or components on the permeability of one or more agents that may be added to the blood or another fluid included in the device, the effect of the blood or components on the health of the BBB-on-chip or any of its components (whether positive or negative), etc. This may include diagnostic uses, for example, to identify a disease, biomarker or infectious agent carried by the blood or blood components.

[0238] In this example, hormone transport across the BMECs was measured in the BBB-on-chip in healthy and diseased tissue by mass spectrometry. Thyroid hormone was added to the bottom channel and measured on the top channel. Thyroid hormones (T3 and T4) were detected using Liquid chromatography tandem-mass spectrometry (LC-MS/MS).

[0239] BMECs from a MCT8 background were used. FIG. 16A shows the user interface and the conditions during the run of human blood across the blood brain barrier. FIG. 16B shows the setup for measuring the transport of solutes from human blood across the blood brain barrier, a barrier created in vitro in the microfluidic devices describes herein using a layer of BMECs. FIG. 16B shows how human blood was perfused into the bottom channel of the tall chip. In this experiment thyroid hormones were measured by LC-MS/MS as described above. This setup will also be used to test the filtration of proteins across the BBB.

[0240] FIG. 17A-C shows the measurement of thyroid hormone transport by mass spectrometry (FIG. 17A) using the setup shown in FIG. 16B, along with the graphed results (FIG. 17B). After flowing patient blood through the microfluidic chips into the channel under the BMECs, it was possible to measure the transport of compounds from the blood into the neural compartment, i.e. through the BMEC barrier. In this case, the experiment included a control set of BBB-on-chips comprising iPS-derived cells originating from a non-diseased individual, and a second set of BBB-on-chips comprising iPS-derived cells originating from a patient diagnosed with Allan-Herndon-Dudley syndrome (AHDS). Briefly, iBMECs were generated from a patient with an inactivating genetic mutation in the MCT8 thyroid hormone transporter. This mutation leads to a defect in T3/T4 transport across the BBB and defects in neural development in patients. The mass-spectrometry data in FIG. 17A is an initial experiment to confirm that the MCT8 transporter defect can be recapitulated on an Organ-Chip.

Example 12

[0241] In this example, the disease model was further evaluated. Samples were prepared by taking 100 ul of each sample of T3 and mixing it with the equivalent sample of T4. This was done for each sample and also for the calibration curve. Proteins and salts were precipitated from the solution; the samples were dried and resuspended in the same volume. The calibration curve permitted the calculation of the concentrations (in mM) for both T3 and T4.

[0242] For the T3/T4 experiments, the following 4 conditions were tested in the microfluidic chip:

[0243] 1. 1 nM T3 in normal media in the bottom channel and media without T3 on top channel. Both sides were running at a 30 ul/hr flow rate.

[0244] 2. 100 nM T3 and T4 in normal media in the bottom channel and media without T3 on top channel. Again, both sides were running at a 30 ul/hr flow rate.

[0245] 3. Human plasma on bottom channel at 90 ul/hr and media without T3 on top channel kept static for 1 hour.

[0246] 4. Human plasma on bottom channel at 90 ul/hr and media without T3 on top channel kept static for 1 hour.

[0247] For each experiment, Dextran-FITC was used in the bottom channel to correct for paracellular diffusion.

[0248] From the above-mentioned 4 conditions, only 100 nM was significantly above detection and these worked well as shown in FIG. 21. Chips 2280, 2289, and 2284 are populated with cells from a single control line. Chips 2285 and 2286 are populated with cells from the isogenic mutated MCT8 line. Chips 2287 and 2288 are populated with cells from a mutated MCT8 patient. FIG. 21 is a bar graph showing the corrected T3 concentration in the top channel of each chip. Clearly, there is reduced T3 transport in mutated MCT8 lines as compared to normal, demonstrating one aspect of disease modeling using the blood-brain barrier, organ-on-chip device.

Example 13

[0249] In one embodiment, the present invention contemplates contact of neurons and brain related vascular cells, and more preferably, direct contact of iMNs and iBMECs on the microfluidic chip to enhance neuronal physiology as measured by electrophysiology and transcriptomics. It has been found that the chip accelerates diMN electrophysiological maturation.

[0250] In this experiment, diMNs seeded into the chip were recorded after 12 days after seeding. FIG. 24A-B provides a whole cell patch clamp recording of a non-invoked spontaneously active neuron showing highly complex and repetitive bursts of neuronal activity indicative of neuronal networks being established in the chip.

[0251] When induced to fire by injecting current into the neuron at day 6 in chip, more resolved action potentials are observed (FIG. 25B) compared to traditional culture (FIG. 25A).

[0252] Neurons that are co-cultured with BMECs in chip (MN/BMEC) show more pronounced currents (FIG. 26B) than MNs cultured alone (FIG. 26A) on chip (MN Only) as depicted by current traces recorded as the neuron is induced to fire an action potential. These observed electrophyisiological properties are well established in the field as indicating neurons are more mature at this time point.

Example 14

[0253] In a controlled study, calcium influx live cell imaging was performed on diMNs that had been cultured in the chip (MN Chip) and in co-culture with BMECs (MN/BMEC). Neuron calcium influx was recorded as described previously, and plotted with respect to time (FIG. 27A-D, right panels). Calcium influx events or peaks correspond to neural activity and were counted by both automated software and blinded human technician. Each event was assigned a time-stamped value and depicted for each tracked neuron with respect to time.

[0254] FIG. 28 is a bar graph showing that the frequency of recorded neurons on the chip is significantly increased in both chip conditions compared to traditional 96 well culture control (CTRL 96). This increase was not observed in 96 well cultures that had been treated with media preconditioned with BMECs (ECCM 96) indicating the increase in the neurons ability to flux was achieved exclusively in the chip. This effect was further increased with the addition of BMECs to the chip in co-culture. Increased frequency is known to occur in vivo as MNs mature and indicate neurons mature faster in the chip.

Example 15

[0255] In this experiment, diMNs were stably transfected with a nuclear-tagged GFP reporter transgene and seeded on the top channel. NON-GFP BMECs were seeded into the bottom channel. Chips were allowed to mature either in this configuration, or non-BMEC controls (both diMN only on chip and diMN in a standard 96 well plate). The cells were FACS sorted to purify the diMN cultures away from the NON-GFP BMECs after 6 days on the chip. These purified cells were mRNA sequenced in all conditions, and a non-biased principle component analysis (PCA) was conducted on all samples. The first principle components separated the conditions by different genes expressed. PC1 separates all cultures from a progenitor pool (black) PC2 genes separated 96-well culture from diMNs in chip, and PC3 separated genes that were exclusively expressed in co-culture with BMECs (FIG. 29A-B).

[0256] The top 200 highly expressed genes and bottom 100 low expressed genes from each PC were entered into the non-biased gene ontology platform DAVID. The resulting pathways included increased neural differentiation in the chip-specific PC2 gene set (FIG. 30, middle list). Vascular interaction gene pathways were found in the co-culture chips indicating that known in vivo gene pathways between the vascular system and neurons were recapitulated in the chip device. The colored bars on the right in FIG. 30 represent the expression of each gene (row) in each of the 5 conditions (columns). Column order is MN Only, BMEC/MN, 96-well control, 96 well ECCM, MN progenitor. Red=high and blue=low. These vascular gene pathways have not been shown to be induced in any other culture system and may be inducing the observed increase in maturity and activity.

Example 16

[0257] Brain Blood Vessel On-Chip Further Comprising Pericytes and Astrocytes.

[0258] FIG. 40A-B Exemplary schematic illustrations of a brain blood vessel in vivo and a microfluidic chip in vitro. FIG. 40A shows a schematic illustration of the histology of a cross section of a brain blood vessel in vivo depicting a blood vessel lumen formed by and surrounded by endothelial cells (EC), a vascular basement membrane separating ECs from pericytes (PC), which in turn are overlain with astrocytes. Kim, et. al., Journal of Cerebral Blood Flow and Metabolism. 2006. FIG. 40B shows one embodiment of a chip in vitro: 1. Upper channel, where the arrow depicts directional fluid flow. 2. Lower channel, where arrow depicts directional fluid flow. 3. Cell layer in upper channel. 4. Enodothelial cell layer in lower channel. 5. Porous Membrane. 6. Optional vacuum chambers for providing membrane stretch.

[0259] FIG. 41A-D Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating exemplary micrographs of immunofluorescently labeled cells in neuronal and vascular channels. FIG. 41A shows an exemplary overview (looking up through the lower channel (dark-purple staining of cells) into the upper channel (light-green staining of cells)) demonstrating cell coverage of both channels along with cell morphology. FIG. 41B shows an exemplary micrograph of immunofluorescently labeled cell markers in an enlarged view from FIG. 41A where the two channels begin overlapping. The cells on the upper right in the upper channel are labeled light-green for a GFAP marker. The cells on the upper left in the lower channel are stained for a Phalloidin marker (dark-purple) for cells restricted to the vascular channel. FIG. 41C shows an exemplary micrograph of an immunofluorescently labeled cell marker showing ZO-1 tight junctions (outlines) of endothelial cells in the vascular channel. Nuclei are stained with DAPI (blue). FIG. 41D shows an exemplary micrograph of an immunofluorescently labeled cell marker showing light-green for a GFAP marker in the neuronal channel, as an enlarged view from FIG. 41A. Nuclei are stained with DAPI (blue).

[0260] FIG. 42A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary endothelial cells in the Vascular Compartment (lower channel). FIG. 42A schematic illustration of one embodiment of an On-Chip System showing neuronal cells in a neural compartment (upper channel) and cells lining a vasculature forming a simulated blood vessel in the lower channel as one embodiment for a chip as shown in FIG. 42B. FIG. 42B shows a micrograph demonstrating a cross section of one embodiment of a blood brain barrier on-chip as a 3D-Reconstruction of a Confocal Micrograph Image of fluorescently labeled cells, i.e. neuronal progenitors in the upper channel and HBMECs in the lower channel. As shown in the insert, endothelial cell markers (left to right) show immunofluorescent staining of Glut-1, PECAM-1, Occludin-5, and ZO-1.

[0261] FIG. 43A-B Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cell morphology of exemplary pericytes and astrocytes as part of a neural compartment (upper channel). FIG. 43A shows exemplary micrographs of fluorescently labeled cells using markers for demonstrating exemplary pericytes (upper micrograph) labeled for alpha-SMA (green) and astrocytes (lower micrograph) labeled for GFAP (red) on-chips. FIG. 43B shows an exemplary micrograph of a lower power confocal micrograph image of fluorescently labeled cells as in FIG. 43A. Nuclei are stained with DAPI (blue). As shown in the insert, these labeled pericytes and astrocytes are an example of cells illustrated schematically in the neuronal compartment (upper channel) above a vascular channel in one embodiment of a blood brain barrier (BBB) on-chip.

[0262] FIG. 44A-C Exemplary micrographs of differentiating iPSC-derived endothelial cells on-chip and an exemplary chart showing that on-chip iPSC-derived endothelial cells form a tighter cell layer simulating a blood brain endothelial barrier. FIG. 44A An exemplary timeline for providing iPSC-derived endothelial cells on-chip. Day 0: exemplary morphology of an iPSC colony stained with GFP showing nuclei (blue). After expansion of iPSc colonies, Day 8-12 Start HBMECs Differentiation. Day 18-22 HBMECs selection for providing HBMECs. FIG. 44B shows a phase contrast micrograph of Human Brain Microvascular Endothelial Cells (HBMECs). FIG. 44C iPS-derived HBMECs (right blue column) demonstrates higher and more physiologically-relevant TEER values as compared to a brain endothelial cell-line (left grey column). .Math.cm.sup.2 vs. Average TEER (ohm.Math.cm.sup.2).

[0263] FIG. 45 Shows one embodiment of a chip in vitro as a blood brain barrier (BBB) on-chip demonstrating cells expressing typical markers of Astrocytes, Pericytes and HMBECs after 5 days on-chip. Middle fluorescent confocal micrograph shows a cross section of cells on a chip. Upper fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary pericytes (SMAred) and astrocytes (GFAPgreen) as part of a neural compartment (upper neuronal channel). Nuclei are stained with DAPI (blue). Lower fluorescent confocal micrograph at a higher magnification shows cell morphology of exemplary endothelial cells (iHBMEC (Glut1red) as part of a vascular compartment (lower vascular channel). Nuclei are stained with DAPI (blue).

[0264] FIG. 46A-B A Barrier Function test using Fluorescent Dextran (3 kDa) demonstrates exemplary low diffusion of Dextran indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 46A An exemplary chart showing Dextran Diffusion as % Dextran (fluorescence). The majority of dextrin is in the vascular channel. FIG. 46B An exemplary Barrier Function Schematic illustrating the movement of dextrin from the Vascular Channel to the Neuronal Channel.

[0265] FIG. 47A-F A Barrier Function test demonstrates exemplary low diffusion of human IgG and albumin but not transferrin indicating intact barrier function of the vasculature in one embodiment of a blood brain barrier (BBB) on-chip. FIG. 47A-B Low diffusion of IgG indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47C-D Low diffusion of Albumin indicates robust and physiological barrier integrity at the vascular-neuronal interface. FIG. 47E-F High levels of transferrin in the neuronal channel indicates presence of transferrin transporters functionally active on iPS-derived brain endothelial cells. FIGS. 47A, C and E Exemplary charts showing human IgG measured by ELISA, human albumin diffusion as albumin concentration ug/ml, and human transferrn (ELISA), respectively. The majority of human IgG and albumin is in the vascular channel. FIGS. 47B, D and F Illustrate exemplary barrier function schematics illustrating the movement of human IgG, albumin or transferrin, respectively, from the Vascular Channel to the Neuronal Channel.