Ref. [36] |
16 independent function units, each unit consists of four uniform BBB regions, replicate the complex multicellular architecture, mechanical properties, 3D extracellular matrix |
Primary rat BMECs |
Primary rat astrocytes |
No membrane |
Rat tail type-I collagen |
No electrode in chip, normal resistance meter |
VE-Cadherin, ZO-1, Claunin-5, etc. (immunofluorescence staining) |
1298 Ω∙cm2
|
Sodium fluorescein (376 Da) |
Examination of brain metastasis and the therapeutic response of brain tumors |
Ref. [40] |
Three PDMS layers plus the PC membrane that divides the two chambers. |
Primary human BMECs |
Primary pericytes, Primary astrocytes and pluripotent stem cell -derived neurons |
PC membrane (0.2 μm pores) |
Laminin |
23 ga stainless steel, not in chip |
ZO-1 (immunofluorescence staining) |
Reported in Ω/cm2, need to be converted |
FITC-dextran (10 and 70 kDa) |
Ascorbate transport across the BBB as an indication of active transport |
Ref. [58] |
Multi-layered channel structure made from patterned PDMS substrate with embedded electrode layers. |
bEnd.3 cell line |
Astrocytes C6 cell line |
PC membrane (10 μm thick, 0.4 μm pores) |
Poly-lysine and fibronectin |
Two sets AgCl electrodes. |
ZO-1 (immunofluorescence staining) |
223–280 Ω∙cm2
|
Not used |
Permeability of seven neuroactive drugs and TEER were quantified in models. |
Ref. [59] |
3 μm pores to allow diffusion of media and tracer between the central and outer compartments |
HUVECs |
Astrocytes CTX-TNA2 cell line |
3 μm pores |
Martrigel and fibronectin |
No electrode |
Not detected |
Not measured |
Texas red dextran (370 kDa) and rhodamine 123 |
Comparing permeability of three passive permeability markers and one marker subject to efflux |
Ref. [60] |
Two separate microchannels supply their respective co-culture tissues independently of one another, and can serve as the microenvironment of the outside and the inside of the BBB respectively. |
HUVECs and human lung fibroblasts |
Primary rat astrocyte and neurons |
No membrane, using fibrin hydrogel |
Fibrin hydrogel |
No electrode |
ZO-1. (immunofluorescence staining) |
Not measured |
FITC-dextran (20 kDa, 70 kDa) |
A platform exhibits direct contact between neural and vascular tissues and a corresponding low permeability characteristic of in vivo BBB |
Ref. [61] |
Two-chamber system divided by PC membrane |
Primary human BMECs |
Human induced pluripotent stem cell-derived neurons and astrocytes |
PC membrane (0.2 μm pores) |
Laminin |
Custom-built multifrequency impedance analyzer |
ZO-1, claudin -5 (immunofluorescence staining) |
Reported in Ω, need to be converted |
FITC-dextran (10 kDa) |
Understand responses to inflammatory stimulation |
Ref. [62] |
Small model, two-layer microchannel and membrane with platinum electrodes. |
hCMEC/D3 cell line |
No co-cultured cells |
PC membrane (10 μm thick, 0.4 μm pores) |
Collagen I |
Platinum electrodes (diameter 200 μm) |
ZO-1 (immunofluorescence staining) |
36.9–120 Ω∙cm2
|
Not used |
Observe shear stress and TNF-α on BBB function |
Ref. [63] |
Vascular conduit overlaid on top of a neural chamber separated by a PC membrane |
RBE4 cell line |
Mixture of neurons (4%), astrocytes (95%), and microglia (1%). |
PC membrane (8 μm pores) |
Poly-lysine and fibronectin |
No electrode |
ZO-1 (western blot) |
Not measured |
Alexafluor-dextran (3 kDa) |
TNF-α simulation triggered neuroinflammation |
Ref. [64] |
The apical and basolateral side separated by 3 μm gaps formed by microfabricated pillars. |
RBE4 cell line |
No co-cultured cells |
No membrane, micro-gaps (50 μm long, 3 μm wide, 3 μm deep) in PDMS wall |
Fibronectin |
No electrode |
ZO-1, claudin (western blot) and P-glycoprotein protein efflux |
Not measured |
FITC- dextran (3–5 kDa) |
Astrocyte-conditioned medium on BBB function |
Ref. [65] |
Two-compartment microfluidic devices were a membrane between two channels. |
bEnd.3 cell line |
C8D1A astrocytes cell line |
PTFE or PE membrane (0.4 μm pores) |
Fibronectin or collagen I |
No electrode |
Claudin-5 (immunofluorescence staining) |
Not measured |
FITC-dextran (70 kDa) |
Study the optically transparent membrane used in models |
Ref. [66] |
Composed of an upper and a lower part that are combined with an adhesive film, three microchannel systems are integrated |
hCMEC/D3 cell line |
Mouse embryonic stem cells derived cortical spheroids |
Polyethylene terephthalate membrane and PC membrane |
Collagen A |
No electrode |
VE cadherin, ᵦ-catenin, ZO-1 (immunofluorescence staining) |
Not measured |
FITC-dextran (3 kDa) |
Detect effects of neuroinflammation upon disruption of the endothelial layer in response to inflammatory signals. |
Ref. [67] |
The porous membrane is situated between the upper and the lower channels made of PDMS. This core is sandwiched by two glass slides with gold electrodes. |
hCMEC/D3 cell line, primary rat brain endothelial cells |
primary astrocytes and brain pericytes. |
PET membrane, (23 μm thick, 0.45 μm pores, pore density 2 × 106/cm2) |
Rat tail collagen |
A pair of 25-nm thick, transparent, gold electrodes was formed on each glass slide |
ᵦ-catenin, ZO-1 (immunofluorescence staining) |
monolayer: 28.5 ± 7.2 Ω∙cm2 Co-culture: 114.2 ± 35.7 Ω∙cm2
|
Sodium fluorescein (376 Da), FITC- dextran (4.4 kDa), Evans blue-labeled albumin (67 kDa) |
Design a new device, which can co-culture of 3 types of cells, observe the cells by microscopy, monitor the TEER, and measure the monolayer permeability |
Ref. [68] |
Creating a cylindrical collagen gel containing a central hollow lumen inside a microchannel |
Primary human BMECs |
Primary human brain pericytes, primary human brain astrocytes |
No membrane |
Rat tail collagen I |
No electrode |
VEcadherin, ZO-1 (immunofluorescence staining) |
Not measured |
Alexa488- dextran (3 kDa) |
Study the secretion profiles of G-CSF, IL-6 and IL-8 when the BBB stimulated with TNF-α |
Ref. [69,70] |
Two PDMS components are separated by PC membrane and form two-chamber system |
hCMEC/D3 cell line |
No co-cultured cells |
PC membrane (0.4 μm pores) |
Fibronectin |
Four platinum wire electrodes inserted into two channels |
ZO-1 (immunofluorescence staining) |
22 ± 1.3 Ω∙cm2
|
Not used |
Developed a stable and easily method to determine TEER in organ-on-chip applications. |
Ref. [71] |
Transparent polyester porous membrane sandwiched between a top and a bottom overlying channel made of PMMA. |
bEnd.3 cell line |
No co-cultured cells |
Polyester membrane (3 μm pores) |
Not used |
Platinum electrodes |
Claudin-5 (immunofluorescence staining) |
About 1000 Ω∙cm2
|
FITC- bovine serum albumin |
Test the ability of a peptide to transport nanoparticles across BBB under flow conditions. |
Ref. [72] |
Two central hydrogel regions for co-culturing astrocytes and neurons, two side channels for hosting endothelial cells and media. |
HUVECs and hCMEC/D3 cell line |
Primary rat neurons and astrocytes |
No membrane, separated by 9 trapezoidal structures |
Poly-lysine and collagen I |
No electrode |
ZO-1 (immunofluorescence staining) |
Not measured |
Oregon green 488- dextran (10 kDa), Texas red dextran (70 kDa) |
Compounds and factors on neural growth and maturation |
Ref. [73] |
Four rectangular channels with different heights to allow simultaneous measurements at different shear stresses. |
Human BMECs from the BC1 human induced pluripotent stem cell line |
No co-cultured cells |
No membrane |
Fibronectin and collagen IV |
No electrode |
claudin-5, occludin, and ZO-1 (immunofluorescence staining) |
Not measured |
Not used |
Study the role of shear stress in modulating the character of human brain microvascular endothelial cells derived from induced pluripotent stem cells. |
Ref. [74] |
Consists of a cell insert and three 3D printed plastic layers with two electrodes |
BMECs from human induced pluripotent stem cells |
Primary rat astrocytes |
PC membrane (0.4 μm pores) |
Collagen IV and fibronectin |
Two 0.8 mm diameter Ag/AgCl pellet electrode |
ZO-1, Claunin-5. (immunofluorescence staining) |
Peaked above 4000 Ω·cm2, sustained above 2000 Ω·cm2
|
FITC-dextran (70, 20 and 4 kDa), Caffeine, cimetidine, and doxorubicin |
Model research |
Ref. [75] |
A double layer microfluidic device with an embedded membrane, the top layer contains a single channel, the bottom channel contains an array of 6 channels |
hCMEC |
No co-cultured cells |
Polyester membrane (0.4 μm pores) |
Fibronectin |
No electrode |
Not detected |
Not measured |
fluorescent sodium salt (376 Da), FITC- dextran (70 kDa) |
Pulsed electric fields may enhance drug delivery to the brain by disrupting the integrity of the BBB and allowing otherwise impermeable drugs to reach target areas. |
Ref. [76] |
Two isolated compartments with the hydrogel reservoir |
hCMEC/D3 cell line |
p5–p7 normal human astrocytes |
No membrane |
Collagen I, matrigel, hyaluronan |
No electrode in chip, normal impedance spectroscopy |
ZO-1. (immunofluorescence staining) |
Static condition: about 200 Ω·cm2 flow condition: about 1000 Ω·cm2
|
FITC-dextran (4 kDa) |
Indicated that the mechanical stress exerted by blood flow is an important regulator of transport both across and along the walls of cerebral microvasculature. |
Ref. [77] |
A 4 × 4 intersecting microchannel array forms 16 BBB sites on a chip, with a multielectrode array integrated to measure the TEER from all 16 different sites. |
Primary mouse BMECs |
Primary mouse astrocytes |
PC membrane (10 μm thick, 0.4 μm pores) |
Fibronectin or matrigel |
Multielectrode arrays, a thin titanium adhesion layer and a gold layer |
ZO-1. (immunofluorescence staining) |
Reported in Ω, need to be converted |
Texas Red dextran (3 kDa), Alexa 546 dextran (10 kDa), FITC dextran (70 kDa) |
Developed multisite BBB chip is expected to be used for screening drug by more accurately predicting their permeability through BBB as well as their toxicity. |
Ref. [78] |
Microchannel with temporary chitosan-based membrane |
hCMEC/D3 cell line |
P6-P10 human astrocytes from the cerebral cortex |
temporary chitosan-based membrane |
Hydrogel matrigel |
No electrode |
Not detected |
Not measured |
Not used |
To obtain a co-culture without a nonphysiological membrane making use of a temporary chitosan membrane in a microfluidic channel. |
Ref. [79] |
Bio-printing, 10 μm average diameter tubes encasing a liquid flow having around 1 mm·s−1 average speed. On the surface of each tube regular pores allowing for mass transport. |
bEnd.3 cell line |
U87 glioblastoma cells |
No membrane, porous tubular structures on tube surface (pore diameter: 1 μm) |
Not used |
No electrode in chip, commercially Voltohmmeter with two electrodes |
ZO-1. (immunofluorescence staining) |
75 ± 2 Ω∙cm2
|
Dextran |
Presented a dynamic 3D biohybrid model of the BBB able to reproduce at 1:1 scale the capillaries of the neurovascular system. |
Ref. [80] |
High-throught, the model harbors 96 or 40 chips in a 384-well plate. In each chip, a perfused vessel of BMECs was grown against an extracellular matrix gel, astrocytes and pericytes were added on the other side of the gel to complete the BBB model. |
Human TY10 cell line (isolated from normal brain tissue from a patient with meningioma) |
Human hBPCT cell line pericytes from brain tissue of a patient t. Human hAst cell line astrocytes from human primary astrocytes distributed by Lonza. |
No artificial membranes, using extracellular matrix gel |
Collagen-I |
No electrode |
claudin-5, VE-cadherin, PECAM-1 (immunofluorescence staining) |
Not measured |
FITC-dextran (20 kDa) |
Developed a high-throughput plate-based model, and used to assess passage of large biopharmaceuticals across the BBB. |