WO2019068127A1 - Method for creating hypoxic conditions within a microfluidic device - Google Patents

Abstract

The present invention relates to a method for creating hypoxic and/or anoxic conditions within a microfluidic device comprising the step of introducing a fluid into a microfluidic device, wherein said microfluidic device comprises at least one inlet for introducing a fluid into said device, at least one outlet for removing a fluid from said device and at least one chamber fluidly connected to said at least one inlet and/or said at least one outlet, wherein at least one inner wall of the at least one chamber comprises an oxygen scavenging material, preferably on its surface.

Description

ME THOD FOR CREATING HYPOXIC CONDI TIONS WI THIN A MICROFLUIDIC

DEVICE

TECHNICAL FIELD

[0001] The present invention relates to methods and means to create hypoxic as well as anoxic conditions within a microfluidic device.

BACKGROUND ART

[0002] Oxygen is the by far most important chemical species for life on earth. It is involved in various cellular biochemical reactions while it modulates cellular

functions like metabolic pathways and plasma membrane integrity in animals, plants and microorganisms.

[0003] Many organisms depend on oxygen to maintain their aerobic metabolism through generation of ATP in

mitochondria, where it serves as the terminal electron acceptor. In contrast obligate anaerobes cannot survive nor grow in presence of oxygen. Oxygen is harmful for this type of organisms since oxygen reduction products (O2-, OH., O2*, H2O2) are generated inside the organisms and only limited or no detoxifying pathways are

available. For instance, the omnipresent enzyme in aerobes superoxide dismutase converts the superoxide anion into ground-state oxygen and hydrogen peroxide, and hence clearing the cytosol from destructive superoxide anions. In the cytosol of obligate anaerobes this enzyme is not or only limited available and consequently they do not possess tools to combat destructive reactions, therefore oxygen is leading to cell death. In contrast to oxygen-based metabolism, aerobes gain their energy by fermentation in which they reduce organic compounds to products such as organic acids or alcohols. These class of bacteria are found in different parts of the human body e.g. colon, nasal fold, urethra, vagina and tooth surfaces, and are potentially pathogenic when displaced from their indigenous habitat. Anaerobic bacterial infections can occur in all parts of the body. For instance, recurrent Clostridium difficile infections constitute a significant clinical issue as it occurs in 15-35% of patients after the first episode. It's a spore forming, gram positive pathogen which secretes toxins causing pseudomembranous colitis and antibiotic- associated diarrhea possibly leading to death. Hence, the early identification from blood or stool samples is essential to treat bacterial infections, but conventional procedures require extensive work and elaborate

techniques to prevent oxygen contact to specimen.

Furthermore, in pharmaceutical vaccine development, animal challenge models are used to study the effectivity and efficiency of drugs and therefore prior testing bacteria needs to be grown from frozen spores. This germination typically needs 2-7 days in total, prior bacteria can be used for the example. In this example a fast (<1 day) and rather simple technique to germinate anaerobic bacteria with the need of minimal bacteria volume is presented.

04] Apart from bacteria, oxygen plays a key role in mammalian cellular functions in normal human physiology as well as disease states. Its concentration varies tremendously in human body, from nearly anoxic in

cartilage to fully oxygenated in arterial blood 13.2%. Moreover, in the human body a sudden loss of oxygen supply can cause irreversible impairment like ischemic brain damage. The brain requires a high amount of energy which it derives from oxygen and glucose and is supplied by cerebral blood. Since the brain is not capable to reserve energy it is highly susceptible to interruption of blood flow. A blockade of the cerebral blood flow is called stroke and immediately causes brain damage. The blood-brain-barrier (BBB) separates the cerebral blood from the brain and constitutes the physical and metabolic barrier while in the early stage of ischemic stroke it loses its integrity leading to infiltration of local inflammatory cells and post-ischemic oedema and swelling. The cerebral endothelial cells are the most important constituent of BBB integrity and their tight junctions regulate barrier permeability. Although these cell types were intensively studied with the use of advanced in vitro models including co-cultures with astrocytes and pericytes, as well as fluid induced shear force mimicking their native microenvironment , the combination of

therefore mentioned with oxygen-glucose deprivation (OGD) still remains to be established.

[0005] A standard method to adjust the gas content of

culture media is to inject a gas mixture with the desired oxygen concentration, so called bubbling. The gas mixture is typically composed of O2, CO2 and an inert gas (most frequently nitrogen) . The conventional way to culture cells and identify potential targeted bacteria is done in hypoxic workstations (glove box isolator) or anaerobic jars containing oxygen scavenging material. However, these methods are either expansive, prone to leaks, or cannot recreate anoxic conditions and neither allows microscopic investigations.

[0006] In recent years microfluidic devices have been proven to be an ideal platform to mimic ^λίη vivo-like'

conditions for cell cultures. Such microdevices enable precise temporal and spatial control which allows the complex formation of biophysical and biochemical

microenvironments for in vitro studies. Several

approaches to control dissolved oxygen levels in

microfluidic devices have been reported previously. A straightforward method to control the oxygen level in microchambers was established with the use of so called oxygen scavenging components. Reducing chemicals such as sodium sulfite were directly supplemented to the culture medium and the oxygen gradient was controlled by media flow. Another approach is based on oxygen diffusion from a control channel across an oxygen permeable membrane to the cell culture chamber. The medium in the control channel was either pre-equilibrated with nitrogen or dissolved oxygen was depleted by chemical reaction using pyrogallol and sodium hydroxide. However, both methods require either external gas supply or handling with chemicals which are potentially harmful to cells. SUMMARY OF THE INVENTION

[0007] It was surprisingly found that the oxygen

concentration of a fluid within a microfluidic device can be controlled by providing a microfluidic device

comprising an oxygen scavenging material comprised within an inner wall of the microfluidic device.

[0008] Hence, the present invention relates to a method for creating hypoxic and/or anoxic conditions or controlling the oxygen concentration within a microfluidic device comprising the step of introducing a fluid into a

microfluidic device, wherein said microfluidic device comprises at least one inlet for introducing a fluid into said device, at least one outlet for removing a fluid from said device and at least one chamber fluidly

connected to said at least one inlet and/or said at least one outlet, wherein at least one inner wall of the at least one chamber an oxygen scavenging material,

preferably on its surface.

BRIEF DESCRIPTION OF THE FIGURES

[0009] Fig. 1 shows chemical structures of the oxygen

indicator dyes palladium ( I I ) and platinum(II) meso- tetra ( 4-fluorophenyl ) tetrabenzoporphyrin (PdTPTBPF and PtTPTBPF) used in the preparation of oxygen sensors.

[0010] Fig. 2 shows microdevices with integrated oxygen

indicator. A) Fabrication of the microdevices. First the liquid prepolymer is poured on the master (1), then it is exposed to UV-light wich initiates the thiol-ene reaction (2) and peeled off from the mould, and access holes drilled (3) . Next, the indicator beads are injected into the microchamber while their surface amino-groups

covalently bond to the epoxy groups of the microchamber walls (4) . The indicator solution is discharged and the polymer layer is removed from the COC-substrate (5) and subsequently bonded to a previously prepared OSTEMER bottom layer (6) . B) Schematic of a microchamber with indicated dimensions (left) and a SEM image of the indicator beads attached at the surface (right) .

Schematic not to scale. C) Image of a microchip (left) with oxygen indicator and the oxygen concentration detection principle (right) .

[0011] Fig. 3 shows oxygen sensor calibration and

characterization. Stern-Volmer calibration curves for (A) normal range indicator (PtTPTBPF) and (B) trace range indicator (PdTPTBFP) at 22°C and 37°C. The dots and error bars show experimental data from 3 independent

calibration measurements. C) Dynamic sensor response for detection of gaseous and dissolved oxygen. The response time (t90) was measured for gas and water was changed from O2 to N2 and deoxygenated water was switched to oxygenated water.

[0012] Fig. 4 shows DO scavenging in microchambers . A)

Variation of chamber height. DO scavenging in chambers with 45, 90 and 250 um height. Devices were

polymerization at 37°C for 35h?. B) Scavenging cycles during 18 days of perfusion. Long-time scavenging in 45 um high chamber polymerized at 37 °C. The microchamber was constantly perfused with diH20 while the perfusion was stopped and DO depletion was measured at indicated time points. 5 scavenging cycles at different time points are shown. Note that DO concentration does not reach 0 hPa since the ports were not closed and therefore DO

diffusion from the connecting tubes was allowed. C) DO scavenging at different polymerization temperatures of the thiol-ene-epoxy thermoset microchamber with a height of 90 μπι. Inlets were closed using PCR foil. D) Influence of the polymerization time duration. 90 um microchamber devices were backed at 150°C.

[0013] Fig. 5 shows FTIR analysis of thiol-enes (OSTEMER) . A) FTIR spectrum of OSTEMER sheets polymerized at 37, 85 and 130°C. B) Free SH-group stretching intensity at different polymerization temperatures. C) Band areas of OH, CH and SH stretching at different thermoset

polymerization temperatures. Integration areas: Thiol: 2607-2536, CH: 3000-2800, OH: 3650-3300 cm-1. n=3

samples .

[0014] Fig. 7 shows Clostridium difficile germination in OSTEMER microchambers. A) Germination in 15 ml centrifuge tubes with oxygenated (+DO) and deoxygenated ( -DO) medium as controls. White arrow indicates bacteria. 24 h after inoculation of C. difficile spores in OSTEMER

microchambers polymerized at B) room temperature, C) room temperature and deoxygenated medium (control) and D) at 150°C. Scale bars 20 μπι. E) Oxygen consumption of dilutions of E. coli suspensions and antibiotic

susceptibility testing

[0015] Fig. 7 shows BBB ischemia model. A) Representative fluorescence images of cerebral endothelial cells visualizing nuclei (DAPI), cytoskeleton (f-actin) and tight junction associated molecules (ZO-1) . White arrows indicate inter-endothelial gap formation after 4h OGD- treatment. B) mRNA expression profiles of HIF-1 target genes VEGF and glutl after 4h of exposure to aglycemia, GOD and normoxia as control. n=3; Error bars indicate S.D.; * p < 0.05, Student's t-test.

DESCRIPTION OF EMBODIMENTS

[0016] The present invention relates to a method for

creating hypoxic and/or anoxic conditions or controlling the oxygen concentration within a microfluidic device comprising the step of introducing a fluid into a

microfluidic device, wherein said microfluidic device comprises at least one inlet for introducing a fluid into said device, at least one outlet for removing a fluid from said device and at least one chamber fluidly

connected to said at least one inlet and/or said at least one outlet, wherein at least one inner wall of the at least one chamber comprises an oxygen scavenging

material, preferably on its surface

[0017] It was surprisingly found that the concentration of oxygen within a fluid being introduced into a

microfluidic device can be controlled by using a

microfluidic device which comprises an oxygen scavenging material which upon contact with a fluid is able to bind reversibly or irreversibly oxygen. [0018] The term "oxygen scavenging material", as used herein, refers to any material capable of gathering or removing oxygen from a fluid.

[0019] The architecture of the microfluidic device used in the method of the present invention can be identical to the devices commonly known in the art. These devices comprise usually at least one inlet for introducing a fluid into said device and at least one outlet for removing a fluid from said device. The microfluidic device of the present invention comprises further at least one chamber which is fluidly connected to said at least one inlet and said at least one outlet. The at least one chamber has typically a larger diameter than the channels present in the microfluidic device of the present invention and may act as a reaction or incubation chamber. In said at least one chamber chemical or

biochemical reactions may be observed or measured by appropriate detection devices (e.g. photometer) .

[0020] The microfluidic device of the present invention

allows to create hypoxic and/or anoxic conditions within the entire microfluidic device or at least parts thereof (e.g. within the at least one chamber) .

[0021] The term "hypoxic", as used herein, refers to a fluid which free oxygen content is lower than typically

occurring in said fluid without having contact with an oxygen scavenging material in a microfluidic device of the present invention. A typically occurring oxygen concentration within a fluid can be determined by

measuring the oxygen concentration within the fluid under normal conditions in relation to temperature (20°C) and pressure (101.325 kPa) before the fluid is introduced into the microfluidic device of the present invention. "Anaerobic conditions", as used herein, refer to "anoxic conditions" where the oxygen content within the fluid is not detectable.

[0022] According to a preferred embodiment of the present invention the oxygen content in a fluid is under "hypoxic conditions" preferably at least 1% at, more least 10%, preferably at least 25%, more preferably at least 50%, more preferably at least 75%, more preferably at least 80%, more preferably at least 90%, more preferably at least 95%, more preferably at least 99.9%, lower compared to the oxygen content within the same fluid under normal conditions before said fluid is introduced into the microfluidic device of the present invention.

[0023] The term "anoxic conditions", as used herein, means that the fluid does not comprise any free oxygen.

[0024] The fluid introduced into the microfluidic device via the at least one inlet is a liquid or a gas. However, according to a preferred embodiment of the present invention the fluid is liquid, more preferably an aqueous liquid .

[0025] The oxygen scavenging material may be coated on the surface of the inner wall of the at least one chamber and/or of the at least one channel. It is also possible to use a material which comprises or consists of an oxygen scavenging material.

[0026] The oxygen scavenging material may be porous allowing the oxygen present in the fluid to be absorbed also in the pores of said material. This is particular

advantageous because the presence of pores increases the surface and thus the oxygen scavenging properties of the oxygen scavenging material.

[0027] According to a preferred embodiment of the present invention the oxygen scavenging material comprises at least one free -SH (thiol) group and/or at least one R-C- S-C-R' (thioether) group.

[0028] It has been found that free thiol groups as well as thioether groups show surprisingly good oxygen scavenging properties .

[0029] According to another preferred embodiment of the

present invention the at least one inner wall comprises a polymer comprising free -SH groups and/or R-C-S-C-R' groups .

[0030] The free -SH and R-C-S-C-R' groups may be part of a polymer, preferably a polymer which can be used to manufacture the microfluidic devices of the present invention or at least parts thereof. [0031] The oxygen transmission rate of the polymers used in the present invention is preferably between 0.5 and 5 cm³/ (m² day bar), more preferably between 1 and 4.5 cm³/ (m² day bar), more preferably between 1.5 to 4 cm³/ (m² day bar), more preferably between 2 and 4 cm³/ (m² day bar), more preferably between 2.5 and 4 cm³/ (m² day bar), more preferably between 3 and 4 cm³/ (m² day bar), in particular 3 or 3,5 cm³/ (m² day bar) .

[0032] "Oxygen transmission rate", as used herein, refers to the measurement of the amount of oxygen gas that passes through a polymer of the present invention over a given period of time. Methods for determining the oxygen transmission rate are well known in the art, whereby it is particularly preferred to use standard method ASTM F2622.

[0033] According to a further preferred embodiment of the present invention the at least inner wall comprises a thiol-ene, thiol-vinyl, thiol-allyl, thiol-alkene, thiol- yne, thiol-maleimide, thiol-isocyanate and/or thiol-epoxy based polymer.

[0034] According to another preferred embodiment of the

present invention the thiol-ene based polymer is OSTEmer 322, OSTEmer 220, OSTEmer 325, OSTEmer 324 (Mercene Labs, Stockholm, Sweden) , NOA 81 (Norrland products, Cranbury, NY ) .

[0035] According to a further preferred embodiment of the present invention the at least inner wall comprises a polymer which is obtainable by at least partially

polymerizing a multifunctional thiol with a

multifunctional allyl (vinyl) and/or multifunctional an isocyanate and optionally a multifunctional epoxy

monomer .

[0036] According to another preferred embodiment of the

present invention the multifunctional allyl is selected from the group consisting of 1, 3, 5-triallyl-l , 3, 5- triazine-2 , 4 , 6 ( 1H, 3H, 5H) -trione, Glyoxal bis (diallyl acetal), 1 , 1 , 2 , 2-Tetraallyloxyethane, Vinyl Ether, Allyl Ether, Propenyl Ether, Allyl Triazine, Allyl

Isocyanurate, Alkene, Acrylate, Unsaturated Ester, Maleimide, Methacrylate, Acrylonitrile, Styrene, Diene and N-Vinyl Amide.

[0037] According to a preferred embodiment of the present invention the multifunctional thiol is selected from the group consisting of pentaerythritol tetrakis(3- mercaptopropionate ) , Alkyl 3-mercaptopropinate,

alkylthioglycolate and Alkyl thiol.

[0038] According to another preferred embodiment of the

present invention the epoxy monomer is selected from the group consisting of D.E.N. 431 Epoxy, Bisphenol A

diglycidyl ether and Tris ( 2 , 3-epoxypropyl ) Isocyanurate .

[0039] According to a further preferred embodiment of the present invention the at least one inlet, the at least one outlet and the at least one chamber are fluidly connected via at least one channel.

[0040] The at least one inlet, the at least one outlet and the at least one chamber may be fluidly connected to each other via at least one channel. In a particularly

preferred embodiment of the present invention all or some of these parts may also comprise an oxygen scavenging material. This is advantageous if the microfluidic device of the present invention shall have a constant level of oxygen or substantially no oxygen.

[0041] According to a preferred embodiment of the present invention the fluid is an aqueous fluid.

[0042] According to a further preferred embodiment of the present invention the fluid comprises a cell culture medium.

[0043] The method of the present invention can be used for various purposes. If biological cells are cultivated or if artificial hypoxic conditions shall be established within the microfluidic device of the present invention the fluid introduced in the device may be a cell culture medium. The type of cell culture medium depends on the cells to be cultivated.

[0044] A major advantage of the present invention is the fact that there is no need to degas the medium before applying it in the microfluidic device of the present invention. The oxygen scavenging material present on or in the least one inner wall of the device is sufficient to reduce the amount of oxygen within the fluid to create anoxic and hypoxic conditions.

[0045] According to another preferred embodiment of the

present invention the fluid comprises cells, preferably animal and/or microbial cells.

[0046] According to a preferred embodiment of the present invention the animal cells are mammalian cells, more preferably human or mouse cells.

[0047] According to another preferred embodiment of the

present invention the animal cells are selected from the group consisting of brain endothelial cells, enterocytes.

[0048] According to a further preferred embodiment of the present invention the microbial cells are bacterial cells, preferably anaerobic or facultative anaerobic bacterial cells.

[0049] According to another preferred embodiment of the

present invention the microbial cells are selected from the group consisting of bacterial cells of the class Clostridia, preferably of the genus Clostridium, more preferably Clostridium difficile.

[0050] According to a further preferred embodiment of the present invention the microfluidic device is used for cultivating cells under hypoxic or anoxic conditions comprising the step of incubating the cells in a

microfluidic device comprising at least one chamber.

[0051] According to a preferred embodiment of the present invention the microfluidic device is used to monitor the influence of hypoxic or anoxic conditions on the growth or biochemical behaviour of cells.

[0052] According to another preferred embodiment of the

present invention the microfluidic device is used to monitor the influence of cells of different sources to each other or of substances on the growth of cells under hypoxic or anoxic conditions.

[0053] The present invention is further illustrated by the following examples, however, without being restricted thereto . EXAMPLES

[0054] Materials and Methods

[0055] Master mould fabrication

[0056] The 45 and 90 μπι high chambers were fabricated as previously described (Sticker D, et al . Lab Chip.

2015;15(24) :4542-54) . Briefly, silicon wafers were sonicated in 2% Hellmanex III solution (Hellma

Analytics), dd¾0 and isopropanol for 10 min at 30 °C. The TMMF S2045 dryfilm resist (Tokyo Ohka Kogyo Co., Ltd) was thermally laminated onto the wafer using a HeatSeal H425 A3 office laminator (GBC) , followed by UV exposure (mask aligner EVG 620) and postexposure bake at 90 °C for 5 min. Micropatterned wafers were developed using EBR solvent (PGMEA/ l-methoxy-2-propyl-acetate,

MicroChemicals ) under magnetic stirring followed by isopropanol and ddH20 rinsing, and a hardbake at 200 °C for 1 h. To prevent OSTEMER-adhesion, moulds were spin- coated with 0.5% Teflon AF (60151-100-6, Dupont) in

Fluorinert FC-40 solution (F9755, Sigma Aldrich) at 3000 rpm for 60 s, followed by two bake steps for 60 s at 125°C and 2 min at 175 °C to evaporate the solvent. For 250 μπι high chambers, master moulds were assembled by placing PDMS sheets (HT-6240, Silex, UK) onto clean silicon wafers.

[0057] Preparation of the oxygen sensor

[0058] Amine-functionalized polystyrene beads (500 μΐ of a 50 mg/mL stock solution, micromer® product code 01-01- 303, micromod, Germany) were diluted with water (2mL) and THF (200 μL) . This dispersion was stirred for 30 min at room temperature. Afterwards, an oxygen indicator

solution (188 μL of a THF solution containing either 2 mg/mL PtTPTBPF or PdTPTBFP, Fig. 1) was added dropwise, the dispersion was sonicated for 30 min, centrifuged at a relative centrifugation force of 6200 and the supernatant was decanted. The particles were purified using multiple washing, centrifugation and decantation steps. The purification process was repeated with water (1-2 times) at the beginning, followed by ethanol (3-5 times) and finished with water (1-2 times) . After the last decantation step the particles were diluted with water to a particle concentration of approx. 50 mg/mL.

[0059] Chip fabrication and oxygen indicator dye integration

[0060] Devices were fabricated using OSTEMER 322-40 (Mercene Labs AB, Sweden) as previously described (Sticker D, et al. Lab Chip. 2015; 15 (24 ) : 4542-54 ) . Total polymer layer height was adjusted to 1.1 mm using glass slides

(Menzel) . UV polymerization was performed with 365 nm Hg- tubes in a crosslinker equipped with integrated energy irradiation sensor (Bio-link BLX Crosslinker, Vilber Lourmat) at a dose of 700 mJ/cm². Similarly, OSTE+ was prepared from monomers as a reference material. Therefore 1, 3, 5-triallyl-l, 3, 5-triazine-2 , 4, 6 (1H, 3H, 5H) -trione (114235, Sigma Aldrich, Austria), pentaerythritol

tetrakis ( 3-mercaptopropionate ) (381462, Sigma Aldrich, Austria) and D.E.N. 431 Epoxy Novolac (38573, Enorica, Germany) were mixed at a stoichiometric ratio of

1/1.4/0.4, and 1% of the allyl-thiol photoinitiator

Irgacure TPO-L (BASF, Germany) , and 1% of the thiol-epoxy initiator DMP30 (T58203, Sigma Aldrich, Austria) added. All further process steps were conducted as described above, however, UV-irradiation was increased to

2.5 J/cm². After first cure the polymer was gently delaminated from the master mould, access holes drilled, developed in a stream of ethanol, dried and cut in shape. The elastic thermoset was then sealed using a COC slide and indicator-bead solution was injected. After 10 min incubation at RT the solution was removed, the chamber rinsed three times with dd¾0 and dry boosted. Next, the thermoset was gently peeled off from the COC substrate and positioned on a 1.1 mm thick OSTEMER bottom slide and fixed in place before final polymerization was initiated (RT or 50 to 150°C) . For devices which were polymerized at temperatures above 80°C no indicator beads were attached to the inner surfaces since at elevated

temperatures the functionality was lost. Therefore oxygen sensing in these fully-polymerized devices was performed by insertion of a 1:5 bead solution in dd¾0 prior

measurement . [0061 ] Calibration of the integrated oxygen sensors

[0062] The integrated oxygen sensors were gas-phase

calibrated using two mass flow controller instruments (Read Y smart series) by Vogtlin instruments

(www.voegtlin.com) to obtain gas mixtures of defined oxygen partial pressures (p02) . Compressed air, 2% (v/v) oxygen in nitrogen and nitrogen were used as calibration gases (Linde, www.linde-gas.at) . Oxygen calibrations at different temperatures were performed by putting the microfluidic chips into a temperature controlled water bath .

[0063] Oxygen transmission rate measurements

[0064] The OTR of OSTEMER 322-40 was determined using an instrument constructed for ultra-barrier membrane samples and is described in Tscherner M, et al . (Sensors, 2009 IEEE; 2009 25-28 Oct. 2009) . Briefly, the 175 μπι thick sample sheets were UV-exposed at a dose of 0.35 J/cm² and either baked at 150°C for 4h or polymerized at RT for 24h. The sheets were clamped in-between two gas chambers; one chamber was initially purged with pure oxygen and the other with nitrogen. Oxygen transmission across the sheets to the oxygen-free chamber was monitored using opto-chemical sensors. The OTR limit of detection is in the 10^~5 [cm³ nr² day^-1 bar^-1] regime.

[0065] Luminescence measurements

[0066] The dissolved and gas phase oxygen concentration was determined by phase-shift measurements using an oxygen meter (Piccolo2, Pyro Science GmbH, Germany) . The USB- controlled read-out system uses an excitation wavelength of 620 nm and reads the emission at 760 nm, with a maximum sampling rate of 20 samples/sec.

[0067] Fourier transform infrared spectroscopy to study

thiol conversion

[0068] The Fourier transform-infrared (FTIR) spectrometer (Tensor 37, Bruker, Germany) was used to determine the conversion of thiol-groups for several polymerization temperatures. OSTEMER sheets were mounted on the Platinum ATR (Bruker, Germany) with a single reflection diamond ATR element and FTIR spectra were recorded at 4 cnr¹ resolution .

[0069] Bacterial culture

[0070] Clostridium difficile spores (VPI 10463, ATCC) were cultures in a biosafety level 2 laboratory. Frozen samples (1.33xl0⁹ CFU) were diluted 1:100 in brain heart infusion (BHI) broth (BD Biosciences, Boston, USA) with the addition of 1% Tauchlorin, to induce sporulation. Samples of 4.7 μΐ were injected into microchambers , while controls were cultured in 15 ml centrifuge tubes with fully oxygenated medium and medium deoxygenated overnight in anaerobic jar (10729081, Fisher Scientific, Austria) supplemented with 1% L-cysteine.

[0071] Cell culture

[0072] As a blood-brain barrier model, mouse brain

endothelial cell line (cerebEND) were cultivated in DMEM medium supplemented with 10% FCS (PAA, A15-101) and 1% penicillin/streptomycin on gelatine coated 25 cm² cell culture flasks. The coating was performed using 1% gelatine solution (9000-70-8, Sigma Aldrich) in DPBS and was incubated for 30 min at 37°C. Cells were split 1:3 once a week and cultivated in humidified atmosphere at 37 °C and 5% CO2 atmosphere. For the OGD treatment cells were seeded at confluence 33% surface coverage on 12mm diameter, gelatin coated glass coversides (Menzl) . After 7 days, confluent slides were sealed with UV-cured

OSTEMER fluidic top layers, complete culture medium was aspirated and OGD medium (-FCS, -glucose) was injected into the microchambers. After closing the chamber ports with PCR adhesive foil, microdevices were placed in a CO2 incubator until further analysis.

[0013] Immunohistochemistry

[0074] For imaging of cytoskeletal organization of cells grown on OSTEMER and cell culture substrates, the F-actin and tight junction staining procedures was used. Using F- actin/ ZO-1 double staining, cells were initially fixed in 4% glutaraldehyde (340855, Sigma Aldrich) and

permeabilized in 0.2% Triton X-100 solution (X100, Sigma Aldrich) for 20 min and 5 min, respectively. To reduce unspecific antibody binding a blocking solution of 5% w/w albumin from human serum (HSA) (A1653, Sigma Aldrich) in IX DPBS was added for 30 min at room temperature. A primary 1:100 rabbit anti-ZO-1 antibody dilution (FIRMA WINNI) in 1% BSA/PBS (Fraction V, human BSA) was

incubated for 60 min at 37°C. Next, a 1:200 dilution of a secondary Alexa488-con ugated anti-rabbit antibody (FIRMA WINNI) in 1% BSA/PBS was incubated for additional 60min at 37°C.Next, TRITC-con ugated phalloidin (90228,

Millipore) solution in 1% HSA/ lx DPBS at a concentration of 6.6 μΜ was incubated for 1 h at room temperature and 0.4 μg/ml DAPI (90229, Millipore) in DPBS was added for 5min. Samples were embedded in Vectashield® mounting medium (Thermo Scientific) prior analysis. Between all steps samples were washed three times in lx DPBS.

[0075] mRNA analysis (Winni)

[0076] Fluorescence and confocal microscopy

[0077] Fluorescence images were taken using a TE2000-S

inverted fluorescence microscope (Nikon) equipped with a DS-QilMc digital camera. All fluorescence images were processed using the manufacturer' s NIS-elements software (Nikon) . CLSM imaging was performed using a Leica TCS SP5 II system (Leica) . Images were recorded with a 63* oil immersion objective using the manufacturer's LAS AF imaging software.

[0078] Scanning electron microscopy (SEM)

[0079] Indicator-bead covered OSTEMER chambers were mounted on aluminium stubs (Christine Gropl

Elektronenmikroskopie ) and coated with a 5 nm layer of gold using an EM SCD005 sputter coater (Leica) . SEM imaging was performed using an Inspect S50 scanning electron microscope (FEI) at 15 kV, spotsize of 3 and high vacuum mode.

[0080] Example 1: Microdevice with integrated oxygen

indicator

[0081] A crucial aspect of any in-vitro cell culture is the control and monitoring of DO (dissolved oxygen) tension. Especially, for anaerobe bacteria where oxygen is

potentially harmful to cells but also for aerobe cells when cultured in microchambers made of low oxygen

permeable material. Therefore, the microchambers using standard casting method were fabricated (Fig. 2A, 1-3), as previously described (Sticker D, et al . Lab Chip.

2015; 15 (24) : 4542-54) and made use of the advantageous two-step curing process of OSTEMER. First the oxygen indicator dye (PtTPTBPF or PdTPTBFP) was incorporated into amine-functionalized polystyrene beads and injected into the OSTEMER microchannel . Since reactive epoxy groups are present at the OSTEMER surface after the first curing step, the indicator-dye beads were covalently linked via the NH2-group to the epoxy group by

nucleophilic attack of the amine nitrogen on the terminal carbon of the epoxy (Fig. 2A-4) . After 15 min of resting time the microchamber was released from the support substrate (Fig. 2A-5) und subsequently bonded to a partially polymerized OSTEMER cover to seal off the microchamber (Fig. 2A-6) . The devices (Fig. 2B) with square-shaped microchambers (45, 90 and 250 μπι high x 2.5 mm wide x 9.5 mm long) were polymerized at different temperatures. To investigate the particle-to-polymer bonding strength, an open microchamber was immersed into an ultrasonic bath for 10 min and microscopic

investigation revealed no visible loss of beads.

Furthermore, electron microscopy showed that the beads form small aggregates while they spread homogeneously across the polymer surface (Fig. 2B, right) . For DO measurements the optical fibre (0 2 mm) of the handheld oxygen meter was placed orthogonally onto the device (Fig. 2C, left) .

82] Next the optode was calibrated in gas for two indicator dyes, the normal-range indicator PtTPTBPF and the PdTPTBFP for trace level detection. Therefore a defined mixture of nitrogen and oxygen from compressed reservoirs was injected into the 90 μπι high flow-through chambers, and oxygen concentration was determined with the oxygen meter. The Stern-Volmer calibration curves were measured in triplicates for each sensor, while the error bars in Fig. 3A-B represent the standard deviation. The two-side model was used to fit the data (equ. 1) (Lafleur JP, et al . Analyst. 2013; 138 (3) : 845-9) .

τ / l+K_vp0₂ 1+Kl_vp0₂) ⁽¹⁾

[0083] Here τ and To are the lifetimes of the luminophore in presence and absence of oxygen, I and Io are the

luminescence intensities in the presence and absence of oxygen, f represents the fraction of the total emission for the first site, and Ksv¹ and Ksv² are the Stern-Volmer quenching constants for the two sites. The fitting results and particularly the high R-squares, show that the indicator beads immobilized at the OSTEMER surfaces behave according to literature and hence the thiol-groups from the thermoset surface does not impede the sensor functionality (Koren K, et al . Sensors and Actuators B, Chemical. 2012 ; 169 (5) : 173-81 ) .

[0084] To determine the sensor response dynamics in solution deoxygenated water was pumped through the chamber from a nitrogen-bubbled reservoir while fully oxygenated water was rapidly flushed through at t=0 (Fig. 3C) using a syringe. The sensor response for gaseous oxygen was determined by a rapid change from flushing nitrogen gas to atmospheric air. The sudden rise in oxygen

concentration was monitored at a resolution of 250 ms and a response time of t9o=0.75 sec and t9o=7.75 sec was achieved for DO and gas phase oxygen, respectively. The microfluidic sensor dynamics for DO was very fast

(t9o=0.75 sec), which allows for high-speed flow-through applications .

[0085] To the best of our knowledge, this is the simplest and fastest technique to permanently incorporate oxygen sensors into microfluidic channels without the need for specialized equipment.

[0086] Since oxygen distribution is limited by diffusion in water and hence oxygen levels could spatially vary in the chamber, it is evaluated on which time scales this affects the oxygen measurements performed on the top surface of the chambers without flow (no convection) .

Given the diffusion coefficient of oxygen in water at 25°C D=2.1*10^~5 cm²/s, the time for diffusive mixing of oxygen in a channel with height h (maximal diffusion length towards the indicator dye) is estimated as

h²/ (2D) . As a result the gas diffusion in the 45, 90 and 250 μπι high chambers is limited on time scales of 0.48, 1.9 and 14.9 sec, respectively. Since all subsequent oxygen measurement dynamics in respective chambers are significantly above the calculated diffusion time scales, it can be assumed that the oxygen level determined at the inner wall represents the mean value inside the chamber.

[0087] Example 2: Oxygen scavenging in OSTEMER microchambers

[0088] A material capable of rapid and efficient molecular oxygen uptake is highly desired for anoxic in vitro cell cultures. One concept to achieve low oxygen

concentrations in microchambers is the integration of oxygen scavenger compounds into the bulk material which avoids the addition of such possibly harmful molecules directly into the broth or culture medium (Park JH, et al . American Journal of Physiology - Cell Physiology.

1999;277 (6) :C1066-C74) . In this section parameters influencing the oxygen scavenging kinetics in thiol-ene- epoxy based microchambers are investigated.

[0089] First the influence of surface-to-volume ratio on DO scavenging was investigated. Therefore, devices with identical lateral side geometry and varying heights of 45, 90 and 250 μπι were fabricated and polymerized at 37°C for 35 h with immobilized indicator beads (Fig. 4A) . In the 45 μπι high chamber a t9o-value of 5.6 min was

measured, while a doubled increase in height (and volume) resulted in less than three times prolonged tgo of

15.6 min. A further increased chamber height of about 5.5 times (250 μπι) resulted in 8.7 times increased tgo of 48.9 min. In table 1 the geometrical factors of the chambers are listed with the corresponding tgo values. Table 1. Oxygen scavenging in OSTEMER chambers with varying heights. Experimentally obtained tgo values represent the time until the oxygen tension reaches 20 hPa.

[0090] Interestingly, the t90 values are inversely

proportional to the surface-to-volume ratios. For

instance, the increase of the surface-to-volume ratio to 8.7 (comparing h=250 to 45 μπι) results in an increase of the t9o-value ratios (compare h=45 to 250 μπι) by the same factor. This quantitative result points towards a surface reaction of oxygen. Such reaction could most likely be explained by oxidation of unreacted thiol-groups at the surface. However, the sulfhydryl group density at the thiol-ene surface (without the addition of epoxy) is approximately 20 nnr² after the first curing step, according to Lafleur JP, et al . (Analyst.

2013; 138 (3) : 845-9) . Neglecting that the addition of epoxy monomers would reduce the SH-surface density, as well as a second curing step would do, a total oxygen uptake capacity of 1.25*10⁹ for the 45 μπι high chamber was calculated. Comparing that with the total amount of oxygen molecules in the chamber, which is approximately 4.8*10¹⁴, amounts to 5 orders of magnitude the SH-groups present at the surface. Consequently the SH-groups at the surface cannot be solely attributed to the oxygen

depletion, but this quantitative evidence points towards significant bulk uptake of oxygen. [0091] To investigate whether the indicator dye alters its functionality as a consequence of surface immobilization, DO measurements have been performed in microchambers with a 250 μπι thick oxygen permeable PDMS sheet as a bottom lid. With this setup no DO depletion was detected during the first hour of measurement. This finding is in

accordance with a previous study where the functionality of the indicator dye covalently bonded to thiol-groups, was demonstrated (Koren K, et al . Sensors and Actuators B, Chemical. 2012 ; 169 (5) : 173-81 ) .

[0092] During DO measurements the highly reactive singlet oxygen (¹02) is produced by photo-irradiation of the indicator and subsequent collisional quenching with oxygen (Ho RYN, et al . Overview of the Energetics and Reactivity of Oxygen. In: Foote CS, Valentine JS,

Greenberg A, Liebman JF, editors. Active Oxygen in

Chemistry. Dordrecht: Springer Netherlands; 1995. p. 1- 23) . To exclude that the luminescent-based oxygen sensing influences the oxygen depletion in microchambers a comparative analysis with varying sample times

(excitation intervals) was performed. It revealed that neither sample intervals (0.25 sec compared to 5, 10 and 60 min) nor ambient light influences the DO scavenging (data not shown) . Therefore, it can be concluded that singlet oxygen produced as a consequence of indicator irradiation is not directly involved in the scavenging reaction .

[0093] To investigate the involvement of unknown chemical additives (e.g. stabilizers, initiators) from the

commercial OSTEMER 322-40 product to the DO depletion in microchambers, OSTE+ thermoset was prepared in-house as a reference material. Therefore, after the first UV-curing step the thermosets were both cured for 20 h at 70°C and subsequently filled with ddH20. After 30 min DO

concentration was determined and resulted in 7 hPa and 42 hPa, for OSTEMER and OSTE+ respectively. Although DO content was lower in michrochambers made of the

commercial thermoset, in OSTE+ microchambers made from monomers the DO concentration was significantly reduced as well. This result indicates that the oxygen scavenging mechanism cannot be solely attributed to possible

additives inside the OSTEMER, although their involvement cannot be excluded.

[0094] To likewise investigate the involvement of possible OSTEMER leachables in DO scavenging microchambers were baked for 22 h at 110°C and subsequently washed for

7 days in ddH20 at 70°C. DO concentrations were monitored for 2 h prior and after the washing step while no

significant difference was detected (data not shown) .

This result indicates that OSTEMER leachables do not affect the DO scavenging in microchambers.

[0095] Next it was investigated whether DO scavenging is altered during continuous, long-time perfusion of the microchamber . Therefore, a 45 μπι high chamber polymerized over night at 37 °C was connected via oxygen permeable tubes to a peristatic pump and dd¾0 from an open

reservoir was constantly perfused at a high flow rate of 200 μΐ/min. To measure DO scavenging, the pump was stopped and the connecting tubes were clamped at

different time points, while perfusion was restarted after full DO scavenging. Fig. 4B shows 5 scavenging cycles during 18 days of perfusion, while 1^st day and 18^th day represents the scavenging prior and 18 days after perfusion, respectively. Comparing the scavenging cycles at day 1 and 18, surprisingly a lower tgo value was measured after 18 days of perfusion then prior perfusion, 29.7 min and 58.6 min, respectively. Furthermore, it can be seen that the scavenging rate does not correlate with the perfusion duration. Therefore it can be concluded that DO scavenging in OSTEMER microchambers does not alter neither saturation occurs after several days of perfusion .

[0096] Next the influence of long-term storage under ambient environment was investigated. A microchamber stored for

5 months at room temperature revealed a tgo value of about

6 h, showing that the oxygen scavenging mechanism

prevails even after several months of storage. [0097] The crosslinking degree of the final polymer and hence conversation of functional groups depends among other parameters on the curing temperature and curing duration. To first investigate the influence of

polymerization temperature on DO scavenging, OSTEMER microchambers were backed at different temperatures and following DO was monitored over time (Fig. 4C) . For polymerization at room temperature, a rapid decrease of DO was monitored resulting in a tgo value of 15.3 min, while polymerization at 70°C resulted in an increased t go value of 73 min. Further elevated curing temperatures highly decreased DO scavenging in the microchambers, where for instance 72 h after dd¾0 injection still a DO of 40 hPa was measured in microchambers cured at 110°C. Next the influence of polymerization time on DO

scavenging was investigated at a curing temperature of 150°C (Fig. 4D) . Baking of the thermoset for 7 min a t go of 15.4 min was measured, while polymerization for 20 min resulted in a t go of 118.5 min and for 6 h a tgo of 26.3 h. These findings show that DO depletion highly depends on the polymerization temperature as well as polymerization duration, those points towards oxidation of unreacted groups in the material.

[0098] Example 3: FTIR spectroscopy of the partially-cured polymer

[0099] In complex polymers such as the ternary thiol-ene- epoxy network which is based on polyfunctional monomers, a full polymerization cannot be achieved, and hence accessible groups may remain in the bulk. Unreacted epoxy groups which could not react with thiols can

homopolymerize or internally cyclizate. In contrast, thiols are effective hydrogen donors and hence oxygen scavengers. To consequently investigate the amount of unreacted thiol groups FTIR spectroscopy was applied.

Therefore thin OSTMER sheets were polymerized at

different temperatures (37°C for 16 h, 85°C for 1 h and 130°C for 16 h) and analysed using ATR-FTIR. In Fig. 5A the full spectra of three samples of each polymer is presented. The thiol band (2600-2550 cnr¹) is shown in Fig. 5B and is highest for the samples cured at lowest temperature (37 °C) . At elevated curing temperatures the thiol-peak decreases but still remains prominent. To exclude that the decrease in the thiol-peak is arising due to an altered polymer to ATR-crystal distance, OH and CH stretching band areas are shown as a reference (Fig. 5C) . Since these reference band areas do not decrease but even slightly increase the influence of contact

alterations to the measurement can be excluded. Those it could be shown that the amount of unreacted thiol groups decrease with increasing curing temperature and still remain at curing temperatures as high as 130°C. Comparing this result with Fig. 4C shows that the decreased DO scavenging effect in microchambers baked at high

temperatures is due to the decreased amount of thiols available to react with the dissolved molecular oxygen.

[0100] Example 4: Anaerobic bacteria germination

[0101] For antibiotic developments animal challenge studies are used to test the impact of bacterial infection.

Therefore, a fast and effective germination of spores is highly desired to reduce preparation time and costs. In particular anaerobic bacteria germination constitutes an elaborative preparation process from deoxygenation of media (overnight) to germination (several days) . Hence, it was investigated whether thiol-ene-epoxy based

microchambers could serve as a device for rapid

germination. To first show that germination from the anaerobic bacteria strain C. difficile is significantly increased in growth medium with a low DO concentration, spores were inoculated in centrifuge tubes either filled with fully oxygenated medium (Fig. 6A left) or medium incubated overnight in an anaerobic jar with L-cysteine supplemented (Fig. 6A right) . Incubation for 23 h clearly revealed that C. difficile spores preferably germinate in medium with a low DO concentration. Next, spores were injected in microchambers and incubated at 37 °C

overnight. A concentration of 1.33*10⁷ CFU/ml was

inoculated in chambers polymerized at RT and 150°C. Fig. 6B-D shows bright field microscopic images of the microchambers 23h post-inoculation. In microchambers polymerized at room temperature bacteria were clearly visible all over (Fig. 6B) , while in chambers polymerized at high temperature only several bacteria were observed. To investigate whether a higher amount of bacteria could be obtained with the use of deoxygenated medium, spores were diluted in low DO medium and inoculated in

microchambers polymerized at room temperature. Fig. 6D shows that no significant increase of bacteria is

visible, indicating that DO scavenging of the material is sufficient for germination in anaerobic environment.

These results indicate that microchambers made from

OSTEMER thermoset and polymerized at low temperatures are suitable to be used for germination of anaerobic bacteria spores. The DO concentration in the growth medium is sufficiently scavenged to promote germination. An aspect which will be investigated in future concerns the

bacterial adhesion to surfaces, those it is challenging to harvest bacteria from the microchambers after

germination .Example 5: In vitro ischemia model of the blood-brain barrier

03] Another import application of low-level DO cell culture is concerned with in vitro ischemia models of the BBB . The integrity of BBB is mainly determined by

endothelial tight junctions, which are multiprotein complexes consisting of transmembrane proteins and cytosolic proteins like zonulaoccluden (ZO)-l. To show that chambers made of low-temperature cured thiol-ene- epoxy thermoset are well suitable for studying ischemic stroke, mouse cerebral endothelial cells (cerebEND) were cultivated and exposed to oxygen-glucose deprivation (OGD) . The OGD treatment was performed in microchambers using glucose-, and serum-free medium (aglycemia) . Prior cell experiments DO was determined inside the closed chamber, reaching <10 hPa after 15 min of incubation at 37°C (data not shown) . Next, cells are grown on gelatine coated glass substrates (0 1.2 cm) until reaching

confluence (6-7 days) and subsequently sealed with a 1.1 mm thick lid made of UV-cured OSTEMER. Since OGD induces loss of cellular barrier function and

consequently changes cell morphology as well as tight- junction expression, cells are investigated using F-actin and ZO-1 staining (Fig. 7A) . First, cells grown under standard culture conditions (5% CO2, 95% humidified air, 37 °C) show typical cytoskeleton development exhibiting elongated and parallel aligned stress fibres. Similarly, tight- unction associated ZO-1 molecules are expressed homogeneously at cellular boundaries. In contrast, 1 h exposure to OGD already shows altered cellular

morphology, characterized by astringent cytoskeleton visible in f-actin immunofluorescence staining. In addition tight junctions (ZO-1) rearrange and are denser concentrated at cellular boundaries. Next, the exposure of the endothelial cells to OGD was extended to 4 h.

Results clearly show the disruption of the confluent endothelial monolayer, exhibiting inter-endothelial gap formation as well as further increased concentration of ZO-1 molecules at the cellular boundaries. To investigate how cells react to aglycemia at normoxic conditions, cerebENDs were cultured in chambers without the OSTEMER lid and incubated at 37 °C humidified and fully-oxygenated atmosphere. Images shown in Fig. 7 (bottom row) clearly reveal that neither morphology nor tight-j unction

distribution visibly changes in the case of nutrient deprivation. Comparing the results with a previously published study the effect of OGD on tight junction disruption is much higher pronounced in this work

(Neuhaus W, et al . Frontiers in Cellular Neuroscience . 2014; 8: 352) . This could be explained by the longer duration cells were exposed to anoxia due to rapid DO scavenging of the thermoset material. With our method hypoxia is reached already 15 min after exposure while in a hypoxic incubator the DO concentration at the bottom of the culture medium (where cells adhere) needs several hours to equilibrate to the desired level (Fernandes TG, et al. Stem Cell Research. 2010 ; 5 ( 1 ) : 76-89 ) . Another reason for the occurrence of inter-endothelial gap formations could be the absolute absence of DO (anoxia) in thiol-ene-epoxy chambers, while in the previous study the experiments were performed at approximately 10 hPa DO (hypoxia) .

[0104] It was reported that mediators like vascular

endothelial growth factor (VEGF) and glutl (a main glucose transporter) are upregulated in the endothelium during ischemic stroke and are involved in BBB disruption (Engelhardt S, et al . Fluids Barriers CNS . 2015; 12 ( 1 ) : 1- 16) . To further characterize the cell responses to reduced oxygenation, mRNA expression of the HIF-1 target genes VEGF and glutl were analyzed by qPCR. Results shown in Fig. 7 reveal VEGF upregulation by 361% in OGD, while aglycemia causes a 176% higher VEGF expression compared to normoxia. Similarly, glutl was upregulated by 14% under aglycemia, while 4 h of GOD exposure increased the glutl amount by 85%. This results qualitatively

correspond with a previous study, where cerebEND cells were exposed to 6h of OGD (Neuhaus W, et al . Neuroscience Letters. 2012 ; 506 ( 1 ) : 44-9 ) . However, glutl was

significantly higher upregulated in the previous study during OGD exposure.

[0105] Conclusions

[0106] Fast and effective dissolved oxygen scavenging is highly desired for cultivation of cells under ultra-low- oxygen tension. Here it is shown that a simple and efficient method to generate low oxygen levels in micro- sized devices. Dissolved oxygen is rapidly scavenged by the preferably partially-cured thiol-ene-epoxy polymer, generating hypoxic conditions 5.6 min after media

injection. It was shown that the scavenging rate can be adjusted by the polymerization temperature and time.

Moreover, it was found no saturation of the scavenging effect 18 days after constant water perfusion. FTIR spectroscopy revealed that even after polymerization at 130°C unreacted thiol-groups are present in the material, thus indicating that oxidation of the sulfur-groups causes the oxygen scavenging from the fluid. Practical applications of the oxygen-scavenging method for studying cell cultures under anoxic conditions are demonstrated. It was shown that the germination of obligate anaerobe bacterial spores C. difficile in anoxic microchambers , where bacteria were readily identified 23h post- inoculation, while with the use of standard anaerobic jars a minimal time of 48h is needed. Moreover, it was successfully demonstrated that the scavenging method can be applied for studying cellular and molecular mechanisms involved in BBB disruption during ischemia. BBB

endothelial cells cultured in anoxic microchambers, loose their native morphology and tight- unction formations, as well as over-express permeabilizing factors such as VEGF and glutl, both previously shown to be involved in BBB disruption .

Claims

CLAIMS :

1. Method for creating hypoxic and/or anoxic conditions within a microfluidic device comprising the step of introducing a fluid into a microfluidic device, wherein said microfluidic device comprises at least one inlet for introducing a fluid into said device, at least one outlet for removing a fluid from said device and at least one chamber fluidly connected to said at least one inlet and/or said at least one outlet, wherein at least one inner wall of the at least one chamber comprises an oxygen scavenging material, preferably on its surface .

2. Method according to claim 1, wherein the oxygen scavenging material comprises at least one free -SH (thiol) group and/or at least one R-C-S-C-R' (thioether) group.

3. Method according to claim 1 or 2, wherein the at least one inner wall comprises a polymer comprising free -SH groups and/or R-C-S-C-R' groups.

4. Method according to any one of claims 1 to 4, wherein the at least inner wall comprises a thiol-ene, thiol-vinyl, thiol- allyl, thiol-alkene, thiol-yne, thiol-maleimide, thiol- isocyanate and/or thiol-epoxy based polymer.

5. Method according to claim 4, wherein the thiol-ene based polymer is OSTEmer 322, OSTEmer 220, OSTEmer 325, OSTEmer 324 or NOA 81.

6. Method according to any one of claims 1 to 5, wherein the at least inner wall comprises a polymer which is obtainable by at least partially polymerizing a multifunctional thiol with a multifunctional allyl (vinyl) and/or multifunctional an isocyanate and optionally an multifunctional epoxy monomer .

7. Method according to claim 6, wherein the multifunctional allyl is selected from the group consisting of 1 , 3 , 5-triallyl- 1 , 3 , 5-triazine-2 , 4 , 6 ( 1H, 3H, 5H) -trione, Glyoxal bis (diallyl acetal), 1, 1, 2, 2-Tetraallyloxyethane, Vinyl Ether, Allyl Ether, Propenyl Ether, Allyl Triazine, Allyl Isocyanurate, Alkene, Acrylate, Unsaturated Ester, Maleimide, Methacrylate, Acrylonitrile, Styrene, Diene and N-Vinyl Amide.

8. Method according to claim 6, wherein the multifunctional thiol is selected from the group consisting of pentaerythritol tetrakis ( 3-mercaptopropionate ) , Alkyl 3-mercaptopropinate, alkylthioglycolate and Alkyl thiol.

9. Method according to claim 6, wherein the epoxy monomer is selected from the group consisting of D.E.N. 431 Epoxy

Novolac, Bisphenol A diglycidyl ether and Tris(2,3- epoxypropyl) Isocyanurate.

10. Method according to any one of claims 1 to 9, wherein the at least one inlet, the at least one outlet and the at least one chamber are fluidly connected via at least one channel.

11. Method according to any one of claims 1 to 10, wherein the fluid is an aqueous fluid.

12. Method according to any one of claims 1 to 11, wherein the fluid comprises a cell culture medium.

13. Method according to any one of claims 1 to 12, wherein the fluid comprises cells, preferably animal and/or microbial cells .

14. Method according to claim 13, wherein the animal cells are mammalian cells, more preferably human or mouse cells.

15. Method according to claim 13 or 14, wherein the animal cells are selected from the group consisting of brain

endothelial cells, enterocytes.

16. Method according to claim 13, wherein the microbial cells are bacterial cells, preferably anaerobic or facultative anaerobic bacterial cells.

17. Method according to claim 13 or 16, wherein the microbial cells are selected from the group consisting of bacterial cells of the class Clostridia, preferably of the genus

Clostridium, more preferably Clostridium difficile.

18. Method according to any one of claims 1 to 17, wherein the microfluidic device is used for cultivating cells under hypoxic or anoxic conditions comprising the step of incubating the cells in a microfluidic device comprising at least one chamber .

19. Method according to any one of claims 1 to 18, wherein the microfluidic device is used to monitor the influence of hypoxic or anoxic conditions on the growth or biochemical behaviour of cells.

20. Method according to any one of claims 1 to 18, wherein the microfluidic device is used to monitor the influence of cells of different sources to each other or of substances on the growth of cells under hypoxic or anoxic conditions.