An oxygenated nutrient delivery microfluidic brain organ chip and a preparation method thereof

Abstract

The application discloses an oxygen and nutrient delivery microfluidic brain organ chip and a preparation method thereof, and belongs to the technical field of biomedical engineering and microfluidic organ chip, which comprises the following steps: inoculating microglial cells treated by oxidative stress into an organoid culture cavity, and inoculating neurons treated by ischemia and hypoxia into a peripheral nerve culture cavity; a concentration distribution of inflammatory factors or damage-related factors is established through a nutrient mixing matrix and a distribution branch, so that a difference microenvironment of a lesion adjacent part and a distal end is simulated; a time sequence evaluation on an interaction process of microglial cell activation and neuron damage is performed in combination with morphological observation and molecular detection; the concentration distribution of the inflammatory factors or the damage-related factors is established between the two cavities through the nutrient mixing matrix, morphological and molecular readouts are acquired in a time sequence of 24 to 72 hours, so that the interaction process of the microglial cell activation and the neuron damage is reproduced.

Description

Technical Field

[0001] This invention relates to the field of biomedical engineering and microfluidic organ-on-a-chip technology, and more specifically, to a microfluidic brain organ-on-a-chip for oxygen and nutrient delivery and its preparation method. Background Technology

[0002] In recent years, brain organoids have been rapidly developed in conjunction with microfluidic technology: gas-permeable membranes are used for transmembrane oxygen supply, tree-shaped or network-type multi-inlet mixing is used for spatiotemporal delivery of nutrients and factors, and liquid-phase directional pumping is used for metabolic clearance.

[0003] To improve observability, on-chip optical detection windows, bypass sampling ports, and dissolved oxygen / metabolism sensing arrays have also been gradually introduced;

[0004] The overall trend is shifting from static batch processing to continuous programmable processing, and attempts are being made to handle oxygen stress, nutrient supply, and metabolic load simultaneously on the same platform.

[0005] However, most existing devices rely on whole-area or coarse-zone oxygen supply combined with single liquid-phase perfusion to achieve basic maintenance. Many solutions still implement oxygen supply, supply and removal, and online detection separately, making it difficult to form a repeatable, traceable, and mass-production quality control system solution within a single chip.

[0006] The existing technology mainly faces three bottlenecks:

[0007] Firstly, the gas phase side and the permeable membrane are not designed with fine patterning and one-to-one alignment, and lack flow-limiting elements and isolation ribs to suppress cross-regional crosstalk, resulting in blurred oxygen field boundaries and easily distorted gradients in the culture chamber, making it difficult to obtain a uniform and programmable oxygen tension distribution.

[0008] Secondly, the multi-inlet nutrient mixing and downstream scavenging often assume equal resistance or ideal shunt. The on-chip calibration port, bypass sampling channel and detection window do not form a closed-loop self-calibration. The output weight drifts with assembly and temperature drift, and the output linearity and error are difficult to quantify and control.

[0009] Third, it remains difficult for the culture chamber to achieve low shear and long-term stability while ensuring perfusion. There is a lack of replaceable cartridges and standardized reset verification procedures, and there is a lack of unified standards for endpoints such as dissolved oxygen uniformity, gradient steady-state time, flow deviation and leakage limits. Summary of the Invention

[0010] To achieve the above objectives, the present invention provides the following technical solution:

[0011] A microfluidic brain organoid chip for oxygen and nutrient delivery includes a chip substrate, an organoid culture chamber, and a gas phase oxygen supply microchannel layer and a liquid phase nutrient perfusion microchannel layer disposed vertically opposite to the culture chamber.

[0012] A permeable membrane is provided between the gas phase oxygen supply microchannel layer and the liquid phase nutrient perfusion microchannel layer. The permeable membrane is patterned to form at least two oxygen permeation zones. The ratio of the oxygen permeability coefficient between the zones is 2 to 8, and the alignment accuracy is no more than 20 micrometers.

[0013] In the gas phase oxygen supply microchannel layer, a zoned oxygen supply branch is set one-to-one above each membrane zone. Each zoned oxygen supply branch is connected to the main distribution pipe and is equipped with a flow limiting structure and an independently adjustable pressure interface. Adjacent zoned oxygen supply branches are isolated by isolation ribs to ensure that the cross-zone flow ratio is no more than 5% under a pressure difference of 1 kPa.

[0014] The organoid culture chamber is a replaceable cartridge structure. A guiding and sealing interface is provided between the cartridge and the chip substrate. The assembly tolerance is no more than 20 micrometers. The culture chamber is equipped with a replaceable three-dimensional porous bracket and a circumferential buffer groove. The bracket has a hole diameter of 20 to 80 micrometers and a porosity of 40 to 70%. The groove width is 100 to 300 micrometers. It is used to limit the average shear stress in the culture chamber to no more than 1 Pa within the rated perfusion flow range.

[0015] The liquid-phase nutrient perfusion microchannel layer includes a nutrient mixing matrix and a metabolic clearance matrix. The nutrient mixing matrix has no less than three inlets and is connected to the culture chamber via no less than four distribution branches. The metabolic clearance matrix is ​​connected to the culture chamber and has no less than one clearance outlet. An on-chip calibration port, a bypass sampling channel and an optical detection window are provided on the matrix for weight self-calibration.

[0016] The bottom or sidewall of the culture chamber integrates a dissolved oxygen sensor array and a metabolite sensor array, with an array pixel density of no less than 25 points per square millimeter and a reference channel, for online measurement and calibration of dissolved oxygen and metabolic indicators.

[0017] Furthermore, the permeable membrane is a gas-permeable elastomer or a nanoporous polymer, with an oxygen permeability of not less than 300 Barrer, a thickness of 20 to 100 micrometers, a boundary transition width of the membrane partitions of not more than 50 micrometers, and a number of partitions of 3 to 6.

[0018] The zoned oxygen supply branch includes flow-limiting micro-holes or equivalent liquid resistance components with an equivalent diameter of 10 to 50 micrometers and the equivalent liquid resistance deviation of each branch is not greater than 10%. The adjacent zones are provided with partition ribs with a height not less than the channel height and a thickness of 20 to 80 micrometers, so that the cross-zone flow ratio measured under a pressure difference of 1 kPa is not greater than 3%.

[0019] The projection overlap between the oxygen supply branch of the partition and the corresponding membrane partition is not less than 95%.

[0020] Furthermore, a selective unidirectional microchannel array is disposed between the organoid culture chamber and the peripheral nerve culture chamber. The microchannels, along the flow direction, sequentially include an inlet constriction section, a neck limiting section, a gradually expanding section, and an outlet guiding section, and satisfy the following geometric and surface energy constraints:

[0021] The aspect ratio of the entrance is greater than 1;

[0022] The equivalent width of the neck is 2 to 4 micrometers;

[0023] The relative rate of change of the width of the transition section along the flow direction is less than 0.1;

[0024] The export guidance angle is 10 to 20 degrees;

[0025] Furthermore, the inner surface of the channel has a surface free energy gradient that decreases along the channel direction, so as to enable unidirectional passage of axons and inhibit retrograde entry of cell bodies.

[0026] Furthermore, the nutrient mixing matrix and metabolic clearance matrix together with the on-chip calibration structure constitute a self-calibration unit. The self-calibration unit includes a bypass sampling microchannel with an error of no more than 10% that is equal in length to each outlet branch, an optical detection window with an equivalent optical path length of 0.2–1.0 mm, and a sampling port with a center distance of 2–10 mm from the tracer injection port.

[0027] A positioning pin and a thrust surface are provided between the organoid culture chamber cartridge and the chip body to achieve reset positioning, with a fitting gap of 5–20 μm.

[0028] The maximum value of the matrix weight estimation error obtained based on the least squares solution is no greater than 3% by calibrating the tracer sequence of the self-calibration unit.

[0029] A method for preparing a microfluidic brain organoid chip for oxygen and nutrient delivery includes the following steps:

[0030] S1: Prepare structural master molds for gas phase oxygen supply microchannel layer, liquid phase nutrient perfusion microchannel layer and selective unidirectional microchannel array, and obtain corresponding structural substrates by photolithography and replication molding.

[0031] S2: The permeable membrane is partitioned and patterned to form oxygen permeability zones, so that the ratio of oxygen permeability coefficients of different zones is 2 to 8, and the alignment with the microchannel layout is completed with a positioning deviation of no more than 20 micrometers.

[0032] S3: The gas phase channel layer, the partition membrane and the liquid phase channel layer are sequentially laminated and bonded at 60 to 90 degrees Celsius to form a three-layer stacked structure;

[0033] S4: Assemble organoid culture chamber cartridges, three-dimensional porous brackets and circumferential buffer grooves, wherein the bracket pore diameter is 20 to 80 micrometers, the porosity is 40 to 70%, and the groove width is 100 to 300 micrometers, so that the average shear stress in the culture chamber is not higher than 1 Pa under the rated perfusion flow rate.

[0034] S5: Integrate dissolved oxygen sensing array and metabolite sensing array at the bottom or side wall of the culture chamber to form an optical detection window and bypass sampling channel. The array pixel density is not less than 25 points per square millimeter and the sampling period is not greater than 10 seconds.

[0035] S6: Construct the nutrient mixing matrix and metabolic clearance matrix, and set the tracer injection port and calibration port. Inject the tracer solution according to the preset input sequence and collect the outlet concentration data. Use the least squares method combined with sparse regularization to solve the weight matrix so that the maximum weight estimation error is no more than 3%.

[0036] S7: Implement factory quality control endpoint tests, including oxygen uniformity index UO less than 0.1, gradient linearity determination coefficient not less than 0.99, gradient steady state time less than 2 minutes, and average shear stress in the culture chamber not higher than 1 Pa under the specified flow rate, and generate batch records.

[0037] Furthermore, in the membrane partitioning patterning process of step S2, plasma surface modification is performed first, followed by sol-gel local sealing to form oxygen permeation partitions;

[0038] The plasma surface modification process uses a mixture of oxygen and argon as the working gas, with an oxygen volume fraction of 50% to 100%, a total flow rate of 50 to 150 standard cubic centimeters per minute, a cavity pressure of 50 to 200 millitors, a radio frequency of 13.56 MHz, a radio frequency power of 50 to 150 watts, a processing time of 30 to 180 seconds, and annealing at 60 to 90 degrees Celsius for 10 to 30 minutes after processing.

[0039] Subsequently, a silica sol precursor solution was deposited on the target area by spin coating, with a solid content of 0.5% to 3.0% by mass, a spin coating speed of 500 to 2000 rpm, a spin coating time of 10 to 30 seconds, a curing temperature of 80 to 120 degrees Celsius, and a curing time of 5 to 20 minutes.

[0040] The alignment deviation between the aforementioned plasma surface modification and the subsequent sol-gel local sealing and membrane partitioning is no more than 20 micrometers, the resulting partition boundary transition width is no more than 50 micrometers, and the oxygen permeability coefficient ratio between the high-permeability zone and the low-permeability zone reaches 2 to 8.

[0041] Furthermore, in step S6, when performing weight self-calibration on the nutrient mixing matrix and the metabolic clearance matrix, the following steps are performed sequentially:

[0042] Under constant temperature of 23 degrees Celsius, set the tracer input sequence and the exit sampling position and establish isochronous sampling, with a sampling period of no more than 10 seconds;

[0043] Under a pressure difference of 1 kPa, tracer solutions with a mass fraction of 0.01% to 0.10% were sequentially injected into each inlet and maintained for a stabilization period of 60 seconds.

[0044] Concentration time series at each outlet were collected and averaged over the steady-state interval to form observation data;

[0045] The weight matrix is ​​solved by least squares combined with sparse regularization. The regularization coefficient is set to 0.001 to 0.01, and the column sum is 1 and non-negativity is imposed on the solution.

[0046] Calculate the weight estimation error. If the maximum value is no greater than 3%, write it to on-chip storage. If it exceeds 3%, return to step 2 to increase the number of measurements.

[0047] Perform a verification test and calculate the deviation of the target concentration at the outlet. If the maximum deviation is no more than 5%, the calibration is completed. If it exceeds 5%, the above steps are repeated after 8 to 24 hours.

[0048] Furthermore, after cartridge replacement and reset, a reset verification process is implemented, which includes the following steps:

[0049] A 10 kPa pressure difference was established using deionized water as the working fluid at 23 degrees Celsius and maintained for 300 seconds for a sealing leakage test. The pressure drop rate was no greater than 1% per minute and the equivalent leakage rate was no greater than 2 μL per minute.

[0050] The volumetric flow rate of each distribution branch was measured under a pressure difference of 1 kPa and compared with the factory archive value. The maximum relative deviation was no greater than 5%.

[0051] With the same set partial pressure applied to each oxygen supply branch in each zone, dissolved oxygen distribution was collected at 23 degrees Celsius, and the dissolved oxygen uniformity index U was calculated. O The relative deviation from the target field, where U O Less than or equal to 0.10 and with a relative deviation of no more than 10%;

[0052] If all the above steps are satisfied, the reset is considered successful and written into the batch record. If any one of them is not satisfied, the assembly is repeated and the process returns to the first step.

[0053] In summary, the present invention has the following beneficial effects:

[0054] By graphically partitioning the permeable membrane and aligning it one-to-one with the partitioned oxygen supply branches, and introducing a force-limiting structure and isolation ribs on the gas phase side, the cross-regional flow ratio is ≤5% under a pressure difference of 1 kPa and the oxygen field boundary is clearly programmable, thereby improving the uniformity and response speed of dissolved oxygen around organoids. This solves to some extent the gradient distortion and central hypoxia problems caused by whole-surface and coarse partitioned oxygen supply in the existing technology.

[0055] By integrating calibration ports, bypass sampling channels, and optical detection windows into the nutrient mixing and metabolic clearance network, and employing least squares weight self-calibration with column summation of 1 and non-negativity constraints, the linearity and accuracy of the outlet ratio can be corrected online (determination coefficient ≥ 0.99, estimation error ≤ 3%), thereby enhancing the predictability and traceability of long-term operation. This also solves, to some extent, the problem of uncontrollable weight drift caused by the medium resistance assumption and assembly deviation in the existing technology.

[0056] By adopting a replaceable cartridge-type culture chamber combined with a three-dimensional porous bracket and a circumferential buffer ring groove, and establishing a reset verification process for endpoints such as sealing leakage, flow deviation, dissolved oxygen uniformity, and steady-state time, the average shear stress of the culture chamber is ≤1 Pa and the batch-to-batch consistency can be quantified, thereby improving the long-term culture stability and reproducibility. To a certain extent, this solves the problems of high shear damage and lack of standardized quality control under perfusion conditions in existing technologies. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of the overall structure and connections of the present invention;

[0059] Figure 2 This is a schematic diagram of the stacked explosion of the present invention;

[0060] Figure 3 This is a schematic diagram showing the alignment relationship between the partitioned membrane and the partitioned oxygen supply branch of the present invention;

[0061] Figure 4 This is a geometric schematic diagram of the selective unidirectional microchannel unit of the present invention;

[0062] Figure 5 This is a flowchart of the self-calibration and verification process of the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] Example:

[0065] The following is in conjunction with the appendix Figure 1-5 The present invention will be described in further detail below.

[0066] Please see Figure 1-5 This invention provides a technical solution: a microfluidic brain organoid chip for oxygen and nutrient delivery, such as... Figure 1-5 As shown, it includes a chip substrate, an organoid culture chamber, and a gas phase oxygen supply microchannel layer and a liquid phase nutrient perfusion microchannel layer arranged opposite to the culture chamber.

[0067] A permeable membrane is set between the gas phase oxygen supply microchannel layer and the liquid phase nutrient irrigation microchannel layer. The permeable membrane is patterned to form at least two oxygen permeation zones. The ratio of the oxygen permeability coefficient between the zones is 2 to 8, and the alignment accuracy is no more than 20 micrometers.

[0068] In the gas phase oxygen supply microchannel layer, a zone oxygen supply branch is set one-to-one above each membrane zone. Each zone oxygen supply branch is connected to the distribution main pipe and is equipped with a flow limiting structure and an independently adjustable pressure interface. The gas path is isolated between adjacent zone oxygen supply branches through the isolation ribs, so that the cross-zone flow ratio is not greater than 5% under a pressure difference of 1 kPa.

[0069] The organoid culture chamber is a replaceable cartridge structure. There is a guide and sealing interface between the cartridge and the chip substrate. The assembly tolerance is no more than 20 micrometers. The culture chamber is equipped with a replaceable three-dimensional porous bracket and a circumferential buffer groove. The bracket has a pore diameter of 20 to 80 micrometers and a porosity of 40 to 70%. The groove width is 100 to 300 micrometers. It is used to limit the average shear stress in the culture chamber to no more than 1 Pa within the rated perfusion flow range.

[0070] The liquid phase nutrient perfusion microchannel layer includes a nutrient mixing matrix and a metabolic clearance matrix. The nutrient mixing matrix has no less than three inlets and is connected to the culture chamber via no less than four distribution branches. The metabolic clearance matrix is ​​connected to the culture chamber and has no less than one clearance outlet. On-chip calibration ports, bypass sampling channels and optical detection windows are set on the matrix for weight self-calibration.

[0071] The bottom or sidewall of the culture chamber integrates a dissolved oxygen sensor array and a metabolite sensor array, with an array pixel density of no less than 25 points per square millimeter and a reference channel, for online measurement and calibration of dissolved oxygen and metabolic indicators.

[0072] In this embodiment: it is assumed that the chip has an external size of 30 mm in length, 20 mm in width, and 3.0 mm in thickness, and adopts a stacked structure: the upper layer is a gas phase oxygen supply microchannel layer, the middle layer is a patterned partitioned permeable membrane, and the lower layer is a liquid phase nutrient perfusion microchannel layer; a replaceable organoid culture chamber cartridge is set in the middle, which is assembled through a guide and sealing interface, and the assembly tolerance is no more than 20 micrometers.

[0073] The culture chamber cartridge has a cavity planar dimension of 6.0 mm × 4.0 mm and a cavity depth of 1.2 mm. A three-dimensional porous bracket is placed at the bottom of the culture chamber. The bracket material is medical-grade polyurethane foam with an average pore diameter of 40 μm, a porosity of 55%, and a thickness of 0.8 mm. A circumferential buffer ring groove with a width of 200 μm and a depth of 150 μm is formed around the cavity and communicates with the cavity to weaken boundary layer shear. Under rated perfusion conditions (total liquid volumetric flow rate of 6 μL / min, equally distributed by four side branches), the measured average shear stress in the culture chamber is 0.6 Pa, which meets the requirement of limiting the shear stress to no more than 1 Pa.

[0074] The zoned permeable membrane is made of gas-permeable elastomer, with a thickness of 50 micrometers, forming four oxygen permeation zones within the projected area of ​​the culture chamber, with a zone boundary transition width of 30 micrometers. Within the gas-phase oxygen supply layer, a zoned oxygen supply branch is configured, corresponding one-to-one with each membrane zone. Each branch is pre-connected with a flow-limiting micropore of equivalent diameter 20 micrometers. Separating ribs, 40 micrometers thick and 90 micrometers high (not less than the channel height), are installed between adjacent zones. Each branch can be independently pressure-regulated via an external interface. Tracer tests were conducted under a 1 kPa pressure difference, and the cross-zone flow rate was measured to be 2.6%, not exceeding 5%.

[0075] A nutrient mixing matrix and a metabolic scavenging matrix are set up on the liquid phase side. The nutrient mixing matrix has three inlets and is connected to the four sides of the culture chamber via four distribution branches; the metabolic scavenging matrix is ​​connected to the opposite side of the culture chamber and has one scavenging outlet, realizing the counter-coordination of supply and scavenging. A calibration port, a bypass sampling channel, and an optical detection window are set in the matrix outlet area. The equivalent optical path length of the detection window is 0.5 mm, which is used for on-chip weight self-calibration and online monitoring.

[0076] An oxygen sensing array and a metabolite sensing array are integrated at the bottom or sidewall of the culture chamber, with a pixel density of no less than 25 dots per square millimeter, and a reference channel is provided for optical baseline and drift correction. When the same partial pressure (40% oxygen in each of the four zones) is set, the oxygen uniformity index U of the steady-state dissolved oxygen field is measured. OThe ratio of standard deviation to mean was 0.066; when the central two subdivisions were adjusted from 21% to 60% and the sides were kept at 21%, the dissolved oxygen gradient reached the ±5% error band in 90 seconds.

[0077] Each gas-liquid interface is compatible with microfluidic ferrules or Luer connectors. The recommended pressure regulation range for the zoned oxygen supply branches is 10 kPa to 60 kPa, and the recommended liquid inlet pressure differential is 1 kPa. Without altering the stacking relationship of the upper gas-phase oxygen supply, the intermediate zoned permeable membrane, and the lower liquid-phase perfusion, and provided that the membrane zones and zoned oxygen supply have a one-to-one correspondence and that a dual-matrix network coexists with on-chip calibration, the membrane material can be replaced with a nanoporous polymer. The membrane thickness can be adjusted from 20 micrometers to 100 micrometers, the number of zones can be 3, 4, 5, or 6, and the flow-limiting structure can be replaced by a serpentine high-resistivity section, with its equivalent liquid resistance deviation calibrated and controlled within 10%.

[0078] like Figure 1-5 As shown, the permeable membrane is a gas-permeable elastomer or nanoporous polymer with an oxygen permeability of not less than 300 Barrer, a thickness of 20 to 100 micrometers, a boundary transition width of no more than 50 micrometers for membrane partitions, and a number of partitions of 3 to 6.

[0079] Each zoned oxygen supply branch contains flow-limiting micro-holes or equivalent liquid resistance components with an equivalent diameter of 10 to 50 micrometers and the equivalent liquid resistance deviation of each branch is no more than 10%. A partition rib with a height no less than the channel height and a thickness of 20 to 80 micrometers is set between adjacent zones so that the cross-zone flow ratio measured under a pressure difference of 1 kPa is no more than 3%.

[0080] The overlap between the projection of the zoned oxygen supply branch and the corresponding membrane zone is not less than 95%.

[0081] In this embodiment: the partitioned permeable membrane is made of a gas-permeable elastomer with a prepolymer to curing agent mass ratio of 10:1. The membrane thickness is 45 micrometers after spin coating. The oxygen permeability is 650 Barrer when measured by constant volume pressure variation at 23 degrees Celsius.

[0082] Four oxygen permeation zones were formed within the projection area of ​​the culture chamber through mask exposure and surface modification. These zones were defined as strips of equal width, with a measured boundary transition width of 25 micrometers between the strips. The membrane pattern and the oxygen supply branches of the upper gas phase oxygen supply layer were aligned one-to-one using microscopic alignment. After alignment, the projection overlap was calculated by image overlay, yielding a result of 97%, which meets the requirement of no less than 95%.

[0083] Each oxygen supply branch in the zone has a flow-limiting micro-orifice at its inlet, serving as an equivalent liquid resistance element. The micro-orifice has a diameter of 20 micrometers and an effective thickness length of 30 micrometers. The volumetric flow rate of the four branches is calibrated with air under a pressure difference of 1 kPa. After converting the equivalent liquid resistance, the relative deviation is calculated, with a maximum deviation of 6% and a limit of no more than 10%. Separating ribs are arranged within the gas phase channel between adjacent zones. These ribs are 40 micrometers thick and 90 micrometers high, exceeding the height of the channel layer to achieve physical isolation of the gas path between zones.

[0084] The cross-flow ratio test was conducted at 23 degrees Celsius: maintaining a pressure difference of 1 kPa between two adjacent zones, 90% oxygen was continuously supplied to the target zone, and nitrogen was supplied to the opposite zone. Oxygen-sensitive signals were acquired at the corresponding positions below the membrane through an optical detection window, and the cross-flow ratio was calculated by the ratio of the steady-state signal intensities. The results showed that the steady-state signal of the adjacent zone was 2.3% of that of the target zone, not exceeding the limit of 3%. This result was replicated on three sets of samples, with a mean of 2.4% and a standard deviation of 0.3%.

[0085] To verify the stability of thickness and boundary transition, three representative locations on the film were measured using a step gauge and line scan. The film thicknesses were 44 μm, 46 μm, and 45 μm, respectively, and the boundary transition widths were 24 μm, 27 μm, and 25 μm, respectively. To verify the alignment accuracy and overlap, the mask marking and channel layout superposition method was used for calculation. The maximum local misalignment of a single sample was 8 μm, and the overall projection overlap was no less than 96%.

[0086] like Figure 1-5 As shown, a selective unidirectional microchannel array is set between the organoid culture chamber and the peripheral nerve culture chamber. The microchannels, along the flow direction, sequentially include an inlet constriction section, a neck limiting section, a gradually expanding section, and an outlet guiding section, and satisfy the following geometric and surface energy constraints:

[0087] The aspect ratio of the entrance is greater than 1;

[0088] The equivalent width of the neck is 2 to 4 micrometers;

[0089] The relative rate of change of the width of the transition section along the flow direction is less than 0.1;

[0090] The export guidance angle is 10 to 20 degrees;

[0091] Furthermore, the inner surface of the channel has a surface free energy gradient that decreases along the channel direction, so as to enable unidirectional passage of axons and inhibit retrograde entry of cell bodies;

[0092] In this embodiment, the array is disposed between the organoid culture chamber and the peripheral nerve culture chamber, and the unit microchannel sequentially includes an inlet constriction section, a neck limiting section, a gradually expanding section and an outlet guiding section along the flow direction.

[0093] The aspect ratio of the inlet contraction section is set to 1.5 (height 6 micrometers, equivalent width 4 micrometers), and the length is 30 micrometers;

[0094] The neck limiting section has an equivalent width of 3 micrometers, a height of 5 micrometers, and a length of 15 micrometers; the gradually expanding section is 300 micrometers long, and the equivalent channel width linearly expands from 3 micrometers to 8 micrometers. The calculated maximum relative change rate along the flow direction is less than 0.02, meeting the requirement of less than 0.1; the outlet guide section adopts a wedge-shaped outward expansion to connect with the downstream cavity, with a guide angle of 15 degrees. The array has 100 parallel channels with a channel center distance of 20 micrometers. A 5-micrometer-wide isolation rib is set between adjacent channels to avoid axonal lateral crosstalk.

[0095] The unidirectional channel was fabricated using photolithography and soft lithography replication processes. A master microstructure of the contraction and gradient sections was constructed on a silicon wafer through two exposures using negative photoresist. The first exposure formed the neck confinement and entrance contraction geometry, while the second exposure used a grayscale mask to achieve linear gradient expansion. After development, an elastomer structure layer was obtained by casting and curing with polydimethylsiloxane. This structure layer was then activated by oxygen plasma and bonded to a cover plate, with an interlayer alignment deviation of no more than 2 micrometers. The nominal height fluctuation of the channel after bonding was less than 0.5 micrometers, verified using a profilometer and a profilometer.

[0096] A gradient of decreasing surface free energy is constructed along the channel direction to enhance the directionality of axon growth and inhibit retrograde cell entry. After bonding, the entire assembly is first treated with oxygen plasma to improve hydrophilicity. Then, a surface modification solution is infused between the two end cavities at a constant rate. A mixed solution of poly-L-lysine and laminin is loaded near the inlet side, while a polyethylene glycol silane shielding solution is simultaneously loaded near the outlet side. A coordination adsorption gradient along the channel direction is formed by controlling the counter-diffusion and the propulsion speed.

[0097] Finally, after standing at 23 degrees Celsius for 30 minutes, it was rinsed with deionized water and dried. The contact angle line scan results showed that the static water contact angle at the inlet was 38 degrees, in the middle of the channel it was 60 degrees, and at the outlet it was 82 degrees. The corresponding surface free energy decreased from about 62 millinewtons per meter to about 33 millinewtons per meter, with a monotonic gradient and no step change.

[0098] Organoids from the same batch were placed in the upstream culture chamber, and peripheral nerve cell suspensions were inoculated in the downstream culture chamber. The same culture medium was used in the upstream and downstream chambers, and the liquid phase pressure difference was maintained at less than 50 Pa to avoid significant body flow. Fluorescence images were collected at the optical detection window at 24 hours and 48 hours after inoculation, and the number of axon cross-windows and cell body crossing events were counted.

[0099] The results showed that at 48 hours, the axonal passage rate along the set direction was 86% (86 channels showed continuous axonal penetration), while the axonal passage rate in the reverse direction was 12%; the cell body retrograde entry rate was 1.8% (2 events out of 100 channels). In the control channel with the equivalent neck width increased to 5 micrometers, the cell body retrograde entry rate increased to 7.4%, demonstrating that a neck confinement range of 2 to 4 micrometers is necessary to inhibit cell body entry.

[0100] In the control group where the surface free energy gradient was removed and only the geometric inequality was retained, the positive axon throughput decreased to 61%, indicating that the synergy between geometry and surface energy gradient can significantly improve unidirectionality.

[0101] To verify the impact of the relative rate of change of the transition section on axon continuity, the length of the transition section was shortened to 100 micrometers (other dimensions remained unchanged) as a control. The maximum value of the relative rate of change along the flow direction was calculated to be approximately 0.05, still less than 0.1, with a forward axon throughput of 79%. When the length was further shortened to 50 micrometers (the boundary of the maximum relative rate of change of approximately 0.1), the forward throughput dropped to 68%, and end folding and adhesion hotspots appeared, supporting the necessity of a relative rate of change of less than 0.1. In the control where the outlet guide angle was adjusted from 15 degrees to 5 degrees, the forward throughput decreased to 65%; when adjusted to 25 degrees, end detachment and backflow occurred, suggesting that 10 to 20 degrees is the optimal range.

[0102] like Figure 1-5 As shown, the nutrient mixing matrix and metabolic clearance matrix together with the on-chip calibration structure constitute a self-calibration unit. The self-calibration unit includes a bypass sampling microchannel with an error of no more than 10% that is equal in length to each outlet branch, an optical detection window with an equivalent optical path length of 0.2–1.0 mm, and a sampling port with a center distance of 2–10 mm from the tracer injection port.

[0103] A positioning pin and a thrust surface are installed between the organoid culture chamber cartridge and the chip body to achieve reset positioning, with a fitting gap of 5–20 μm.

[0104] By calibrating the tracer sequence of the self-calibration unit, the maximum error of the matrix weight estimation obtained based on the least squares solution is no greater than 3%.

[0105] In this embodiment, the partitioned permeable membrane is made of a gas-permeable elastomer with a prepolymer to curing agent mass ratio of 10:1. A 45-micrometer thick membrane is obtained through spin coating, and the oxygen permeability is measured to be 650 Barrer using the constant volumetric pressure variation method at 23 degrees Celsius. Four oxygen permeability zones are formed within the projection area of ​​the culture chamber through mask exposure and surface modification. These zones are equal-width strips, with a measured boundary transition width of 25 micrometers between the strips. The membrane pattern and the partitioned oxygen supply branches of the upper gas-phase oxygen supply layer are aligned one-to-one using microscopic alignment. After alignment, the projection overlap is calculated by image superposition, resulting in 97%, which meets the requirement of not less than 95%.

[0106] Each oxygen supply branch in the zone has a flow-limiting micro-orifice at its inlet, serving as an equivalent liquid resistance element. The micro-orifice has a diameter of 20 micrometers and an effective thickness length of 30 micrometers. The volumetric flow rate of the four branches is calibrated with air under a pressure difference of 1 kPa. After converting the equivalent liquid resistance, the relative deviation is calculated, with a maximum deviation of 6% and a limit of no more than 10%. Separating ribs are arranged within the gas phase channel between adjacent zones. These ribs are 40 micrometers thick and 90 micrometers high, exceeding the height of the channel layer to achieve physical isolation of the gas path between zones.

[0107] The cross-flow ratio test was conducted at 23 degrees Celsius: maintaining a pressure difference of 1 kPa between two adjacent zones, 90% oxygen was continuously supplied to the target zone, and nitrogen was supplied to the opposite zone. Oxygen-sensitive signals were acquired at the corresponding positions below the membrane through an optical detection window, and the cross-flow ratio was calculated by the ratio of the steady-state signal intensities. The results showed that the steady-state signal of the adjacent zone was 2.3% of that of the target zone, not exceeding the limit of 3%. This result was replicated on three sets of samples, with an average value of 2.4% and a standard deviation of 0.3%.

[0108] To verify the stability of thickness and boundary transition, three representative locations on the film were measured using a step gauge and line scan. The film thicknesses were 44 μm, 46 μm, and 45 μm, respectively, and the boundary transition widths were 24 μm, 27 μm, and 25 μm, respectively. To verify the alignment accuracy and overlap, the mask marking and channel layout superposition method was used for calculation. The maximum local misalignment of a single sample was 8 μm, and the overall projection overlap was no less than 96%.

[0109] like Figure 1-5 As shown, a method for preparing a microfluidic brain organoid chip for oxygen and nutrient delivery includes the following steps:

[0110] S1: Prepare structural master molds for gas phase oxygen supply microchannel layer, liquid phase nutrient perfusion microchannel layer and selective unidirectional microchannel array, and obtain corresponding structural substrates by photolithography and replication molding.

[0111] S2: The permeable membrane is partitioned and patterned to form oxygen permeability zones, so that the ratio of oxygen permeability coefficients of different zones is 2 to 8, and the alignment with the microchannel layout is completed with a positioning deviation of no more than 20 micrometers.

[0112] S3: The gas phase channel layer, the partition membrane and the liquid phase channel layer are sequentially laminated and bonded at 60 to 90 degrees Celsius to form a three-layer stacked structure;

[0113] S4: Assemble organoid culture chamber cartridges, three-dimensional porous brackets and circumferential buffer grooves, wherein the bracket pore diameter is 20 to 80 micrometers, the porosity is 40 to 70%, and the groove width is 100 to 300 micrometers, so that the average shear stress in the culture chamber is not higher than 1 Pa under the rated perfusion flow rate.

[0114] S5: Integrate dissolved oxygen sensing array and metabolite sensing array at the bottom or side wall of the culture chamber to form an optical detection window and bypass sampling channel. The array pixel density is not less than 25 points per square millimeter and the sampling period is not greater than 10 seconds.

[0115] S6: Construct the nutrient mixing matrix and metabolic clearance matrix, and set the tracer injection port and calibration port. Inject the tracer solution according to the preset input sequence and collect the outlet concentration data. Use the least squares method combined with sparse regularization to solve the weight matrix so that the maximum weight estimation error is no more than 3%.

[0116] S7: Implement factory quality control endpoint tests, including oxygen uniformity index UO less than 0.1, gradient linearity determination coefficient not less than 0.99, gradient steady state time less than 2 minutes, and average shear stress in the culture chamber not higher than 1 Pa under the specified flow rate, and generate batch records.

[0117] In this embodiment, structural master molds for a gas phase channel, a liquid phase channel, and a selective unidirectional microchannel are fabricated using a silicon wafer as a substrate. The nominal height of the gas phase channel is set to 80 micrometers, and the main channel width is set to 300 micrometers. The nominal height of the liquid phase channel is set to 80 micrometers, and the branch width is set to 150 micrometers. The unidirectional microchannel unit includes inlet contraction, neck limiting, gradient expansion, and outlet guiding geometry. The equivalent neck width is designed to be 3 micrometers, the relative change rate of the gradient section is less than 0.1, and the outlet guiding angle is 15 degrees. The master molds are obtained by spin coating with negative photoresist and a two-exposure process. After development, they are baked and cured, and the surface is sprayed with a fluorosilane release agent for later use.

[0118] The gas phase channel layer and liquid phase channel layer were obtained through replication molding. The elastomer prepolymer and curing agent were mixed at a mass ratio of 10:1, degassed, and then poured into each master mold. After standing at room temperature for 10 minutes and curing at 80 degrees Celsius for 45 minutes, the structural layer was peeled off, and external interface holes were opened. The permeable membrane was formed on a flat substrate using spin coating, with a target thickness of 50 micrometers. After curing, four oxygen permeability zones were patterned, with the transition width of the zone boundaries controlled at 30 micrometers. Low-permeability zones were obtained by surface densification treatment of some zones, while the rest remained high-permeability zones. The measured ratio of the oxygen permeability coefficient between the high-permeability and low-permeability zones on the membrane was 4.5 (within the range of 2 to 8).

[0119] The membrane plate and the oxygen supply branches of the gas phase layer are aligned under a microscope, with an alignment error of no more than 10 micrometers.

[0120] The gas phase channel layer, partitioned membrane and liquid phase channel layer are sequentially laminated and bonded. Before bonding, the contact surfaces of each layer are activated by oxygen plasma. The process conditions are 80 watts power and 60 seconds time.

[0121] After coarse positioning by aligning the positioning holes and using the positioning pins, fine adjustment is performed on a microscopic alignment platform, with the final positioning deviation not exceeding 15 micrometers; a three-layer stacked structure is formed by hot pressing and bonding under 80 degrees Celsius and 0.2 MPa flat pressure conditions for 20 minutes.

[0122] After bonding is completed, a replaceable organoid culture chamber cartridge is assembled and an annular seal is installed. The assembly tolerance of the cartridge guide and sealing interface is not greater than 20 micrometers. A three-dimensional porous bracket with an average pore diameter of 40 micrometers, a porosity of 55%, and a thickness of 0.8 millimeters is placed in the culture chamber. A circumferential buffer ring groove with a width of 200 micrometers and a depth of 150 micrometers is machined around the chamber.

[0123] The nutrient mixing matrix and metabolic clearance matrix were constructed on the liquid phase side. The nutrient mixing matrix had 3 inlets and was connected to the culture chamber via 4 distribution branches, while the metabolic clearance matrix was connected to the opposite side of the culture chamber and had 1 clearance outlet.

[0124] A calibration port and bypass sampling channel were established in the matrix exit area, and an optical detection window with an equivalent optical path length of 0.5 mm was fabricated. Subsequently, dissolved oxygen and metabolite sensing arrays were integrated at the bottom of the culture chamber, and an oxygen-sensitive fluorescent microarray was deposited using micro-contact printing with a pixel density of 25 dots per square millimeter. A reference channel was set on the sidewall for optical baseline correction. The sampling period of the sensing readout system was set to 5 seconds to meet the requirement of not exceeding 10 seconds.

[0125] On-chip weight self-calibration is performed to determine the weights of the nutrient mixing matrix and the metabolic clearance matrix.

[0126] The tracer solution was injected into the three inlets sequentially according to the preset input sequence. Each sequence was kept in steady state for 60 seconds at 23 degrees Celsius and 1 kPa inlet pressure difference, and the concentration data of each outlet was collected in the detection window. After summing the data to form an observation matrix, the least squares method was used to solve the problem, combined with sparsity regularization. The regularization coefficient was set to 0.005, and the column sum of 1 and non-negativity constraints were applied to the solution.

[0127] The calculated matrix weight estimation error was 2.1%, and it was written to on-chip memory as the running weights. A verification test was conducted using a different set of input sequences to validate the target outlet concentration; the maximum deviation was 3.4%. If further convergence is required, the calibration process is repeated.

[0128] Implement factory quality control endpoint testing;

[0129] First, a 10 kPa pressure difference was established using deionized water as the working fluid and maintained for 300 seconds. The pressure drop rate was measured to be 0.3% per minute and the equivalent leakage rate was 0.6 μL per minute. Second, the volumetric flow rate of each distribution branch was measured under a 1 kPa pressure difference and compared with the archived values. The maximum relative deviation was 3.8%. Third, dissolved oxygen distribution was collected under the condition of equal pressure oxygen supply in four zones, and the dissolved oxygen uniformity index was calculated to be 0.066.

[0130] In the switching test where the pressure in the two central compartments was adjusted from 21% to 60%, the dissolved oxygen gradient reached steady state in 90 seconds. At the rated flow rate of 6 μL / min, the average shear stress within the culture chamber was 0.6 Pa based on flow field measurements and calculations. All the above endpoints met the requirements of oxygen uniformity index less than 0.1, gradient linearity determination coefficient not less than 0.99 (calibrated to 0.995), steady-state time less than 2 minutes, and average shear stress not exceeding 1 Pa, and batch records were generated and archived.

[0131] like Figure 1-5 As shown, in the membrane partitioning patterning process in step S2, plasma surface modification is performed first, followed by sol-gel local sealing to form oxygen permeation partitions.

[0132] The plasma surface modification process uses a mixture of oxygen and argon as the working gas, with an oxygen volume fraction of 50% to 100%, a total flow rate of 50 to 150 standard cubic centimeters per minute, a cavity pressure of 50 to 200 millitors, a radio frequency of 13.56 MHz, a radio frequency power of 50 to 150 watts, a processing time of 30 to 180 seconds, and annealing at 60 to 90 degrees Celsius for 10 to 30 minutes after processing.

[0133] Subsequently, a silica sol precursor solution was deposited on the target area by spin coating, with a solid content of 0.5% to 3.0% by mass, a spin coating speed of 500 to 2000 rpm, a spin coating time of 10 to 30 seconds, a curing temperature of 80 to 120 degrees Celsius, and a curing time of 5 to 20 minutes.

[0134] The alignment deviation between the aforementioned plasma surface modification and the subsequent sol-gel local sealing and membrane partitioning is no more than 20 micrometers, the resulting partition boundary transition width is no more than 50 micrometers, and the oxygen permeability coefficient ratio between the high-permeability zone and the low-permeability zone reaches 2 to 8.

[0135] In this embodiment, the membrane partitioning patterning in step S2 employs a sequential process of "first plasma surface modification, then sol-gel local sealing." The substrate is the gas-permeable elastomer film prepared above, with a thickness of 50 micrometers, attached to a glass slide to ensure flatness. A rigid mask made by laser windowing of a stainless steel sheet is used as the masking template. The template and the membrane surface are bonded together by vacuum adsorption, and micrometer-level registration is achieved by aligning with the reference edges and positioning holes. The measured alignment deviation is no greater than 12 micrometers. The mask window corresponds to the future high-permeability partition, and the remaining masked areas are the subsequent low-permeability partitions.

[0136] The plasma surface modification step was completed within an RF plasma reaction chamber. The working gas was a mixture of oxygen and argon, with an oxygen volume fraction of 80%, a total flow rate of 100 standard cubic centimeters per minute, a chamber pressure of 100 mTorr, an RF frequency of 13.56 MHz, an RF power of 100 W, and a processing time of 90 seconds. Immediately after treatment, the surface functional groups were stabilized and the pattern durability was improved by annealing at 80°C for 20 minutes. This step activated the surface of the windowed areas, forming an oxygen-loving interface, while the masked areas remained largely unchanged. The first mask was removed and replaced with a second mask corresponding to the low-permeability partition pattern, completing the localized sealing deposition of the sol-gel.

[0137] Sol-gel local sealing utilizes a silica precursor sol with a solids content of 1.0% by mass. It is deposited in low-permeability target zones via spin coating at a speed of 1200 rpm for 20 seconds, followed by curing at 100°C for 10 minutes. This coating fills the micropores of the membrane at the molecular scale and reduces local oxygen permeability. After mask removal, alternating high-permeability and low-permeability strips aligned with the oxygen supply pathways are obtained. The zone boundaries are measured using confocal linear scanning and oxygen-sensitive fluorescence imaging, with a boundary transition width of 25 μm. Using constant volumetric pressure variation at 23°C, the oxygen permeability in the high-permeability and low-permeability zones is measured to be 650 Barrer and 145 Barrer, respectively, with a ratio of 4.5, falling within the defined range of 2 to 8. To verify uniformity, repeated measurements were taken at three equidistant locations on the membrane surface, yielding boundary transition widths of 24 μm, 26 μm, and 25 μm, and high-to-low permeability ratios of 4.6, 4.4, and 4.5, respectively. The batch stability meets the requirements for mass production.

[0138] After completing the partitioned membrane, it was aligned and bonded to the gas-phase oxygen supply channel layer. Alignment was achieved using a cross-shaped reference mark under a microscope, and the projected overlap was calculated to be 97% using image overlay, meeting the requirement of no less than 95%. The assembled chip was then subjected to cross-regional flow verification under a 1 kPa pressure difference: the target partition supplied 90% oxygen, and the adjacent partition supplied nitrogen. Steady-state oxygen-sensitive signals were collected at the corresponding positions under the membrane through an optical detection window, and the calculated flow ratio was 2.3%, meeting the requirement of no more than 3%. Additional adhesion and durability tests showed that the local sealing coating showed no visible peeling after three tape peel tests; after aging for 72 hours at 37°C and 95% humidity, the high-low permeability ratio remained above 4.3, and the boundary transition width change did not exceed 2 micrometers.

[0139] The aforementioned sequential process ensures that a stable oxygen-loving interface and patterned boundary are first constructed through plasma, followed by localized densification of the low-permeability zones using sol-gel, thereby simultaneously meeting the three thresholds of "alignment deviation not exceeding 20 micrometers, zone boundary transition width not exceeding 50 micrometers, and high-to-low oxygen permeability coefficient ratio of 2 to 8". Compared to the control process using only a single surface modification or a single coating, this embodiment achieves a steeper and repeatable permeation gradient at the same film thickness, and achieves a lower cross-zone crossflow ratio and higher alignment overlap after assembly.

[0140] like Figure 1-5 As shown, in step S6, when performing weight self-calibration on the nutrient mixing matrix and the metabolic clearance matrix, the following steps are performed sequentially:

[0141] Under constant temperature of 23 degrees Celsius, set the tracer input sequence and the exit sampling position and establish isochronous sampling, with a sampling period of no more than 10 seconds;

[0142] Under a pressure difference of 1 kPa, tracer solutions with a mass fraction of 0.01% to 0.10% were sequentially injected into each inlet and maintained for a stabilization period of 60 seconds.

[0143] Concentration time series at each outlet were collected and averaged over the steady-state interval to form observation data;

[0144] The weight matrix is ​​solved by least squares combined with sparse regularization. The regularization coefficient is set to 0.001 to 0.01, and the column sum of 1 and non-negativity constraints are imposed on the solution.

[0145] Calculate the weight estimation error. If the maximum value is no greater than 3%, write it to on-chip storage. If it exceeds 3%, return to step 2 to increase the number of measurements.

[0146] Perform a verification test and calculate the deviation of the target concentration at the outlet. If the maximum deviation is no more than 5%, the calibration is completed. If it exceeds 5%, the above steps are repeated after 8 to 24 hours.

[0147] In this embodiment, a reproducible implementation of the weight self-calibration process for the nutrient mixing matrix and metabolic clearance matrix is ​​provided. After the device is assembled according to the aforementioned embodiment, it is calibrated at 23 degrees Celsius. The equivalent optical path length of the detection window is 0.5 mm, and the sampling period of the optical readout system is set to 5 seconds (not exceeding 10 seconds). The sampling time reference of each outlet is achieved through isochronous sampling using the same trigger signal. The tracer solution uses an inert dye compatible with the culture medium, prepared with a mass fraction of 0.05%, and is sterilized and degassed by a 0.22-micron filter membrane before use to reduce bubble interference.

[0148] The calibration sequence was executed using a combination of individual inlet excitation and a small amount of mixed excitation. Under an inlet pressure difference of 1 kPa, tracer solutions were sequentially injected into inlets A, B, and C, each maintaining a stabilization period of 60 seconds. Subsequently, two sets of mixed input sequences [A,B,C]=[0.5,0.5,0],[0.5,0,0.5] were executed to improve the matrix condition number. The first 30 seconds of each sequence were discarded as a transition period, and the mean of the steady-state concentration time series at each outlet for the following 30 seconds was used as the observation data for that sequence, forming the observation matrix. To eliminate the influence of optical path and illumination fluctuations, the reference channel signal was recorded once at the beginning and once at the end, and baseline normalization was performed on all observations.

[0149] The weights are calculated using least squares combined with sparse regularization, with the regularization coefficient set to 0.005. Two constraints, "column sum of 1" and "non-negativity", are applied to satisfy mass conservation and physical feasibility.

[0150] Taking a three-inlet, four-outlet configuration as an example, the obtained 4×3 weight matrix is:

[0151] ;

[0152] The sum of each column is equal to 1 and all elements are non-negative.

[0153] The training sequence was back-substituted using this matrix, and the maximum relative residual was calculated to be 2.4%. The maximum weight estimation error was no more than 3%, which met the judgment threshold. The coefficient set was then written into on-chip storage as running weights.

[0154] To perform independent verification, a verification sequence [A,B,C]=[0.2,0.5,0.3] was constructed without participating in the solution process. The steady-state concentration at each outlet was collected under a 1 kPa differential pressure. The target outlet concentration was predicted using the obtained weight matrix. The maximum deviation compared to the measured value was 3.1%, and the root mean square deviation was 1.7%, not exceeding the 5% verification threshold. To assess repeatability, the same chip was calibrated three times. The median intra-batch coefficient of variation for the weight matrix elements was 1.9%, with a maximum not exceeding 3.8%. The entire process was repeated once for different chip samples, resulting in a maximum cross-chip elemental difference of 5.2%, falling within the allowable fluctuation range for production.

[0155] During the above process, if the residual of any training sequence exceeds 3%, the system will automatically return to the separate excitation step of the corresponding entry, increase the number of measurements by 1, and extend the steady-state holding time to 90 seconds. If the maximum deviation exceeds 5%, the entire calibration will be repeated after 8 to 24 hours (to eliminate the influence of temperature drift and slow evolution of surface state).

[0156] After calibration, the weight matrix version number and timestamp are written into the batch record and archived in conjunction with the flow distribution and dissolved oxygen uniformity data from the factory quality control for subsequent traceability and reproduction.

[0157] The above embodiments can make equivalent adjustments to the number of sequences, sampling duration and regularization coefficient without changing the limitations such as "isochronous sampling, pressure difference of 1 kPa, tracer mass fraction of 0.01% to 0.10%, least squares combined with sparse regularization and applying column sum and non-negativity constraints, maximum error of no more than 3%, and verification deviation of no more than 5%".

[0158] like Figure 1-5 As shown, the reset verification process after cartridge replacement and reset includes the following steps:

[0159] A 10 kPa pressure difference was established using deionized water as the working fluid at 23 degrees Celsius and maintained for 300 seconds for a sealing leakage test. The pressure drop rate was no greater than 1% per minute and the equivalent leakage rate was no greater than 2 μL per minute.

[0160] The volumetric flow rate of each distribution branch was measured under a pressure difference of 1 kPa and compared with the factory archive value. The maximum relative deviation was no greater than 5%.

[0161] With the same set partial pressure applied to each oxygen supply branch in each zone, dissolved oxygen distribution was collected at 23 degrees Celsius, and the dissolved oxygen uniformity index U was calculated. O The relative deviation from the target field, where U O Less than or equal to 0.10 and with a relative deviation of no more than 10%;

[0162] If all the above steps are satisfied, the reset is recorded as qualified and written into the batch record. If any one of them is not satisfied, the assembly is repeated and the first step is returned to repeat the above steps.

[0163] In this embodiment, the reset verification process involves replacing the culture chamber cartridge and completing the reset assembly. The device is placed in a 23°C environment, with deionized water as the working fluid on the liquid side, and a seal leak test is first performed.

[0164] A pressure difference of 10 kPa was established at the inlet and maintained for 300 seconds. The inlet pressure was continuously recorded by a pressure sensor and the pressure drop rate was calculated. At the same time, the equivalent leakage rate was converted by the mass method. The pressure drop rate obtained in one test was 0.3% per minute and the equivalent leakage rate was 0.6 μL per minute. Both of these values ​​did not exceed the limits of 1% per minute and 2 μL per minute, respectively, and the test was deemed successful.

[0165] Subsequently, a 1 kPa pressure difference was established at the liquid phase inlet. The volumetric flow rate of each branch was measured sequentially according to the branch numbers recorded in the factory archive. The maximum relative deviation from the corresponding archived value was taken as the criterion. In this embodiment, the relative deviations of the four branches were 1.8%, 2.5%, 3.2%, and 2.1%, respectively, with the maximum relative deviation being 3.2%, which is no greater than 5%, and the test was passed.

[0166] After completing the flow consistency verification, apply the same set partial pressure to each zone oxygen supply branch according to the factory calibration table, and maintain a stable pressure supply on the gas phase side that is independent of the liquid phase side.

[0167] Under steady-state conditions, dissolved oxygen distribution within the culture chamber plane was acquired using an on-chip dissolved oxygen sensor array. The dissolved oxygen uniformity index U was calculated using 400 pixels from a regular grid. O (The ratio of standard deviation to mean), and the relative deviation between fields is calculated with the target field at the factory as a reference.

[0168] In this embodiment, the average dissolved oxygen was measured to be 6.7 mg / L with a standard deviation of 0.50 mg / L, yielding U O =0.075, while the relative deviation from the target field is 6%;

[0169] Both satisfy U O The requirements of less than or equal to 0.10 and relative deviation not exceeding 10% are considered met. Once all three criteria are met, "Reset Pass" is recorded in the batch record, including the test date, ambient temperature, pressure setting, sampling period, flow rate and relative deviation of each branch, and U. O The numerical value deviates from the target field, the data file path, and the operator's signature.

[0170] If any of these conditions are not met, immediately remove the cartridge and seals, check the guide and thrust surfaces, seal compression and assembly tolerances. If necessary, replace the seals and clean the interface again before reassembling. Repeat all steps starting from the seal leakage test with the same diameter until the judgment threshold is met or the part is judged as unqualified and transferred to the rework process. This embodiment does not change the "seal leakage test 10kPa for 300 seconds, flow deviation not greater than 5%, U O Under the premise of "less than or equal to 0.10 and relative deviation not greater than 10%, all passing is considered as successful reset, and if any one is not met, reinstallation and retesting are required", the sampling frequency and number of pixels can be adjusted equivalently according to the equipment range.

[0171] A method for neurotoxicological screening based on microfluidic brain organoid microarrays for oxygen and nutrient delivery, referring to... Figure 1-5 ,include:

[0172] S1 seeded microglia in the organoid culture chamber and neurons in the peripheral nerve culture chamber to form a two-cell co-culture.

[0173] S2 uses the nutrient mixing matrix and distribution branches in the liquid phase nutrient perfusion microchannel layer, adopts equal volume flow rate and sets the volume fraction of drug solution and culture medium according to the weight matrix obtained by on-chip calibration, so as to form a continuous concentration distribution in the culture chamber.

[0174] S2 uses the weight matrix obtained from the factory calibration, and the calibration satisfies the following conditions: the coefficient of determination of the linearity of the export ratio is not less than 0.99 and the maximum value of the weight estimation error is not greater than 3%.

[0175] S3 morphological and molecular readouts were collected during culture to determine the toxicological effects of low-concentration drugs activating microglia and indirectly damaging neurons via inflammatory factors.

[0176] Among them, selective unidirectional microchannel arrays are used to limit the unidirectional growth of axons toward the peripheral nerve culture cavity, and axonal penetration rate, number of branches and cell body retrograde entry rate are used as one of the morphological interpretation indicators.

[0177] Molecular readouts include at least one of the following: Iba1, IL1β, TNFα, Caspase3, or TUNEL;

[0178] The time series evaluation covers 24 to 72 hours and includes quantitative readouts at at least three time points.

[0179] In this embodiment, microglia are seeded in the organoid culture chamber and neurons are seeded in the peripheral nerve culture chamber. The volume fractions of the drug solution and culture medium are set in the nutrient mixing matrix according to the weight matrix specified on the chip and the flow rate is maintained at the same volume, so as to form a continuous concentration distribution in the culture chamber. The morphological indicators are axonal penetration rate, branch number and retrograde cell entry rate, and the molecular indicators are Iba1, IL1β, TNFα, Caspase3 or TUNEL, to determine the microglia-mediated indirect neurotoxicity induced by low concentration of drugs.

[0180] A method for constructing and evaluating stroke-related pathological microenvironments based on microfluidic brain organoid chips for oxygen and nutrient delivery, referring to... Figure 1-5 ,include:

[0181] Microglia were seeded into organoid culture chambers after being treated with oxidative stress, and neurons treated with ischemia and hypoxia were seeded into peripheral nerve culture chambers.

[0182] The concentration distribution of inflammatory factors or damage-related factors is established by using a nutrient mixing matrix and distribution branches to simulate the differential microenvironment between the vicinity and distal parts of the lesion.

[0183] We combined morphological observations with molecular detection to evaluate the time-series interaction between microglia activation and neuronal damage.

[0184] In this embodiment, microglia were activated by oxidative stress and then seeded into the organoid culture chamber, and neurons treated with ischemia and hypoxia were seeded into the peripheral nerve culture chamber.

[0185] By establishing the concentration distribution of inflammatory factors or damage-related factors between the two chambers through a nutrient mixing matrix, and obtaining morphological and molecular readouts in a 24 to 72-hour time series, the interaction process between microglial activation and neuronal damage can be reproduced.

[0186] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0187] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A microfluidic brain organoid chip for oxygen and nutrient delivery, comprising a chip substrate, an organoid culture chamber, and a gas-phase oxygen supply microchannel layer and a liquid-phase nutrient perfusion microchannel layer disposed vertically opposite to the culture chamber, characterized in that: A permeable membrane is disposed between the gas phase oxygen supply microchannel layer and the liquid phase nutrient perfusion microchannel layer. The permeable membrane is patterned to form at least two oxygen permeation zones, with the oxygen permeation coefficient ratio between the zones being 2 to 8 and the alignment accuracy not exceeding 20 micrometers. In the gas phase oxygen supply microchannel layer, a zone oxygen supply branch is set one-to-one above each membrane zone. Each zone oxygen supply branch is connected to the distribution main pipe and is equipped with a flow limiting structure and an independently adjustable pressure interface. The gas path is isolated between adjacent zone oxygen supply branches through a partition rib. The organoid culture chamber is a replaceable cartridge structure. A guiding and sealing interface is provided between the cartridge and the chip substrate. The assembly tolerance is no more than 20 micrometers. The culture chamber is equipped with a replaceable three-dimensional porous bracket and a circumferential buffer groove. The bracket has a pore diameter of 20 to 80 micrometers and a porosity of 40 to 70%. The groove has a width of 100 to 300 micrometers and is used to limit the average shear stress in the culture chamber to no more than 1 Pa within the rated perfusion flow range. The liquid phase nutrient perfusion microchannel layer includes a nutrient mixing matrix and a metabolic clearance matrix. The nutrient mixing matrix has no less than three inlets and is connected to the culture chamber via no less than four distribution branches. The metabolic clearance matrix is ​​connected to the culture chamber and has no less than one clearance outlet. On the matrix, an on-chip calibration port, a bypass sampling channel and an optical detection window are provided for weight self-calibration. The bottom or sidewall of the culture chamber integrates a dissolved oxygen sensor array and a metabolite sensor array, with an array pixel density of no less than 25 points per square millimeter and a reference channel, for online measurement and calibration of dissolved oxygen and metabolic indicators. The permeable membrane is a gas-permeable elastomer or a nanoporous polymer with an oxygen permeability of not less than 300 Barrer, a thickness of 20 to 100 micrometers, a boundary transition width of no more than 50 micrometers for membrane partitions, and a number of partitions of 3 to 6. The partitioned oxygen supply branches include flow-limiting micropores or equivalent liquid resistance components with an equivalent diameter of 10 to 50 micrometers, and the equivalent liquid resistance deviation of each branch is no greater than 10%. The height between adjacent partitions is no less than the channel height. The partition ribs, with a thickness of 20 to 80 micrometers, ensure that the cross-regional flow ratio measured under a pressure difference of 1 kPa is no greater than 3%. The overlap between the projection of the oxygen supply branch and the corresponding membrane zone is not less than 95%; A selective unidirectional microchannel array is disposed between the organoid culture chamber and the peripheral nerve culture chamber. The microchannels, along the flow direction, sequentially include an inlet constriction section, a neck limiting section, a gradually expanding section, and an outlet guiding section, and satisfy the following geometric and surface energy constraints: The aspect ratio of the entrance is greater than 1; The equivalent width of the neck is 2 to 4 micrometers; The relative rate of change of the width of the transition section along the flow direction is less than 0.1; The export orientation angle is 10 to 20 degrees; Furthermore, the inner surface of the channel has a surface free energy gradient that decreases along the channel direction, so as to enable unidirectional passage of axons and inhibit retrograde entry of cell bodies; The nutrient mixing matrix and metabolic clearance matrix together with the on-chip calibration structure constitute a self-calibration unit. The self-calibration unit includes a bypass sampling microchannel with an error of no more than 10% that is equal in length to each outlet branch, an optical detection window with an equivalent optical path length of 0.2–1.0 mm, and a sampling port with a center distance of 2–10 mm from the tracer injection port. A positioning pin and a thrust surface are provided between the organoid culture chamber cartridge and the chip body to achieve reset positioning, with a fitting gap of 5–20 μm. The maximum value of the matrix weight estimation error obtained based on the least squares solution is no greater than 3% by calibrating the tracer sequence of the self-calibration unit.

2. A method for preparing a microfluidic brain organoid chip for oxygen and nutrient delivery, wherein the chip is the microfluidic brain organoid chip for oxygen and nutrient delivery as described in claim 1, characterized in that, Includes the following steps: S1: Prepare structural master molds for gas-phase oxygen supply microchannel layer, liquid-phase nutrient perfusion microchannel layer and selective unidirectional microchannel array, and obtain corresponding structural substrates through photolithography and replication molding; S2: The permeable membrane is partitioned and patterned to form oxygen permeability zones, so that the ratio of oxygen permeability coefficients of different zones is 2 to 8, and it is aligned with the microchannel layout with a positioning deviation of no more than 20 micrometers. S3: The gas phase channel layer, the partition membrane, and the liquid phase channel layer are sequentially laminated and bonded at 60 to 90 degrees Celsius to form a three-layer stacked structure; S4: Assemble organoid culture chamber cartridges, three-dimensional porous brackets and circumferential buffer grooves, wherein the bracket pore diameter is 20 to 80 micrometers, the porosity is 40 to 70%, and the groove width is 100 to 300 micrometers, so that the average shear stress in the culture chamber is not higher than 1 Pa under the rated perfusion flow rate; S5: Integrate dissolved oxygen sensing array and metabolite sensing array at the bottom or side wall of the culture chamber to form an optical detection window and bypass sampling channel. The array pixel density is not less than 25 points per square millimeter and the sampling period is not greater than 10 seconds. S6: Construct the nutrient mixing matrix and metabolic clearance matrix, and set the tracer injection port and calibration port. Inject the tracer solution according to the preset input sequence and collect the outlet concentration data. Use the least squares method combined with sparse regularization to solve the weight matrix so that the maximum weight estimation error is no more than 3%. S7: Implement factory quality control endpoint tests, including oxygen uniformity index UO less than 0.1, gradient linearity determination coefficient not less than 0.99, gradient steady state time less than 2 minutes, and average shear stress in the culture chamber not higher than 1 Pa under the specified flow rate, and generate batch records.

3. The method for preparing a microfluidic brain organoid chip for oxygen and nutrient delivery according to claim 2, characterized in that, In the membrane partitioning patterning process in step S2, plasma surface modification is performed first, followed by sol-gel local sealing to form oxygen permeation partitions. In this process, plasma surface modification uses a mixture of oxygen and argon as the working gas, with an oxygen volume fraction of 50% to 100%, a total flow rate of 50 to 150 standard cubic centimeters per minute, a cavity pressure of 50 to 200 millitors, a radio frequency of 13.56 MHz, a radio frequency power of 50 to 150 watts, a processing time of 30 to 180 seconds, and annealing at 60 to 90 degrees Celsius for 10 to 30 minutes after processing. Subsequently, a silica sol precursor solution was deposited on the target area by spin coating, with a solid content of 0.5% to 3.0% by mass, a spin coating speed of 500 to 2000 rpm, a spin coating time of 10 to 30 seconds, a curing temperature of 80 to 120 degrees Celsius, and a curing time of 5 to 20 minutes. The alignment deviation between the aforementioned plasma surface modification and the subsequent sol-gel local sealing and membrane partitioning is no more than 20 micrometers, the resulting partition boundary transition width is no more than 50 micrometers, and the oxygen permeability coefficient ratio between the high-permeability zone and the low-permeability zone reaches 2 to 8.

4. A microfluidic brain organoid for oxygen and nutrient delivery according to any one of claims 2 or 3. A method for fabricating a chip, characterized in that, In step S6, when performing weight self-calibration on the nutrient mixture matrix and metabolic clearance matrix, the following steps are performed sequentially: Under a constant temperature of 23 degrees Celsius, the tracer input sequence and the exit sampling position are set, and isochronous sampling is established, with a sampling period of no more than 10 seconds; Under a pressure difference of 1 kPa, tracer solutions with a mass fraction of 0.01% to 0.10% were sequentially injected into each inlet and maintained for a stabilization period of 60 seconds. Concentration time series at each outlet were collected and averaged over the steady-state interval to form observation data; The weight matrix is ​​solved by least squares combined with sparse regularization. The regularization coefficient is set to 0.001 to 0.01, and the column sum of 1 and non-negativity constraints are applied to the solution. Calculate the weight estimation error. If the maximum value is no greater than 3%, write it to on-chip storage. If it exceeds 3%, return to step 2 and increase the number of measurements. Perform a verification test and calculate the deviation of the target concentration at the outlet. If the maximum deviation is no more than 5%, the calibration is terminated. If it exceeds 5%, the above steps are repeated after 8 to 24 hours.

5. The method for preparing a microfluidic brain organoid chip for oxygen and nutrient delivery according to claim 4, characterized in that, After cartridge replacement and reset, a reset verification process is implemented, which includes the following steps: At 23 degrees Celsius, a pressure difference of 10 kPa was established using deionized water as the working fluid and maintained for 300 seconds to conduct a sealing leakage test. The pressure drop rate was no greater than 1% per minute and the equivalent leakage rate was no greater than 2 μL per minute. The volumetric flow rate of each distribution branch was measured under a pressure difference of 1 kPa and compared with the factory archived value. The maximum relative deviation was no greater than 5%. With the same set partial pressure applied to each oxygen supply branch in each zone, dissolved oxygen distribution was collected at 23 degrees Celsius. The relative deviation between the dissolved oxygen uniformity index UO and the target field was calculated, where UO is less than or equal to 0.10 and the relative deviation is not greater than 10%. If all the above steps are satisfied, the reset is considered successful and written into the batch record. If any one of them is not satisfied, the assembly is repeated and the process returns to the first step.

Images