Microfluidic enzyme assays

Abstract

The invention concerns an improved microassay device comprising a plurality of pillars made from, or coated with, a copolymer comprising, as copolymerized units, at least one polythiol monomer ('thiol monomer') and at least one polyalkene / alkyne monomer ('ene' monomer'), wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100:110, and more preferably is about 100:110 or 100:125; its method of fabrication; use of the device for assaying enzyme reactions, specifically for use with metabolic enzymes such as Cytochrome P450 enzymes or, indeed, any other enzyme(s) but particularly including those susceptible to deactivation by uncrosslinked thiol monomers; and a kit of parts including said device.

Description

[0001] 4908P WO As Filed

[0002] Microfluidic Enzyme Assays

[0003] Field of the Invention

[0004] The invention concerns an improved microassay device comprising a plurality of pillars made from, or coated with, a copolymer comprising, as copolymerized units, at least one polythiol monomer (‘thiol monomer’) and at least one polyalkene / alkyne monomer (‘ene’ or ‘yne’ monomer’), wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100:110, and more preferably is about 100:110 or 100:125; its method of fabrication; use of the device for assaying enzyme reactions, specifically for use with metabolic enzymes such as Cytochrome P450 enzymes or, indeed, any other enzyme(s) but particularly including those susceptible to deactivation by uncrosslinked thiol monomers; and a kit of parts including said device.

[0005] Background of the Invention

[0006] Adverse reactions to novel therapeutics tend to be characterised by inhibition of metabolic enzymes, such as for example the cytochrome P450 enzymes, and this inhibition can be reversible (competitive or non-competitive) or irreversible. Irreversible inhibition is particularly damaging to the individual as it results in sustained inactivation of the enzyme and even auto-immune reactions. It is therefore a necessary step in drug development to test novel therapeutics for any adverse metabolic effects with a view to eliminating those drugs that inhibit, particularly irreversibly, metabolic enzymes. To facilitate this process, it is highly desirable to use fast, reproducible and reliable in vitro assay techniques. Microfluidic (flow-through) assays (referred to herein as microassays) are particularly favoured because their scale enables reduction in the consumption of bioreagents and they can be automated to enable measurement of transient enzyme activities, both reproducibly and reliably. They are also favoured because they allow multiple interrogations simultaneously on a single sample source, thus enabling the effect of increasing / decreasing concentrations of novel therapeutics to be determined, time course of exposure to be assayed and association with other drugs to be assayed. Additionally, they also enable one to distinguish between reversible 4908P WO As Filed and irreversible enzyme inhibitor drugs, in a straightforward manner, as they allow for the test substance to be withdrawn from the feed and the enzyme activity to then be assayed again in the absence of the test substance to determine if enzyme activity has been restored, indicating whether any inhibition is reversible / irreversible. This cannot be done using static conditions where the enzymes are dissolved in a reaction mixture.

[0007] However, we have discovered that certain metabolic enzymes are susceptible to deactivation by the composition of the microassay device or microassay substances. Conventional microfluidic assays are performed using micropillars to which, typically, enzymes to be tested are tethered whilst a solution containing the novel therapeutic to be tested and a substrate and cofactor(s) for the enzyme flows thereby / thereover. The conversion of the substrate to a reaction product is then measured to determine the activity of the enzyme in the presence of the novel therapeutic and so provides a measure of the effect of the therapeutic on metabolic enzyme activity. Given the ability to assay the effects of multiple parameters simultaneously more than one novel therapeutic can be tested in this system and / or more than one enzyme can be tested simultaneously in this system at more than one therapeutic concentration. Additionally, the system can be used to determine whether any enzyme inhibition is reversible / irreversible. Thus, the use of microassays in this way represent powerful investigative tools.

[0008] It is therefore imperative that the polymer substrate to which the microfluidic assay components are attached is inactive or does not interfere with the activity of the enzyme to be investigated.

[0009] When assaying metabolic enzymes using fluidic devices, biological lipid membranes (microsomes) and membrane-bound proteins contained therein, such as enzymes, in particular cytochrome P450 (CYP) enzymes and UDP glucuronosyl transferase (UGT) enzymes, are tethered to micropillars using biotinylated fusogenic liposomes and a thiol-ene polymer micropillar surface coated with an avidin, such as streptavidin. This is illustrated in Figure 1 . Our 4908P WO As Filed work has shown that the CYP enzymes are prone to rapid inactivation when immobilized on such a thiol-ene polymer based micropillar reactor. In fact, we have discovered it is the uncrosslinked thiol monomer(s) that result(s) in CYP inactivation. It is thought that the root cause of the CYP inactivation is the leaching of uncrosslinked thiol monomers from the bulk polymer.

[0010] Notably, not all enzymes are affected in this way, indeed others have shown (Kiiski et al., Eur J Pharm Sci 2021 , 158, 105677 (9 pp)) microfluidic assays for UGTs using pillars comprising, as copolymerised units, pentaerythritol tetrakis (3-mercaptopropionate) (a tetrathiol monomer) and triallyl-1 ,3,5- triazine-2,4,6(1 H,3H,5H)-trione (a triallyl), in the fabrication of the microreactors, did not result in any kind of UGT enzyme inhibition at a drug testing concentration of up to 1 mM.

[0011] Therefore, there is a need for a new microassay device that overcomes the above-described problems. Our new device avoids or slows down the inactivation of the membrane-bound CYP enzymes during use, via adjustment of the thiol-ene / yne bulk polymer composition (stoichiometric ratio of thiol and ‘ene / yne’ monomers). Additionally, we have made further improvements that increase the reliability of both CYP and UGT assays such as adapting the polymer post-processing protocol and increase the sensitivity by changing the micropillar structure (diamond instead of round shape); these adaptions include optimizing the master (mold) microfabrication process for high quality diamond micropillar arrays, to improve immobilization of membrane-bound enzymes.

[0012] Statements of Invention

[0013] According to a first aspect of the invention there is provided a microassay device comprising a plurality of pillars comprising, as copolymerized units, at least one polythiol monomer (‘thiol monomer’) and at least one polyalkene / alkyne monomer (‘ene’ monomer’), wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 4908P WO As Filed

[0014] 100:105, preferably about 100:140 to 100:110, and more preferably is about 100:110 or 100:125.

[0015] As used herein, the term polythiol monomer refers to a compound comprising at least two, preferably at least three, and more preferably at least four thiol (- SH) functional groups.

[0016] In exemplary embodiments, the thiol monomer is a dithiol compound comprising two thiol functional groups. Suitable dithiol compounds include but are not limited to: 1 ,6-hexanedithiol; 2,5-dimercaptomethyl-1 ,4-dithiane; 2,3- dimercapto-1 -propanol; Benzene-1 ,2-dithiol; 1 ,8-octanedithiol; Ethylene glycol bis(3-mercaptopropionate).

[0017] In other exemplary embodiments, the thiol monomer is a trithiol compound comprising three thiol functional groups. Suitable trithiol compounds include but are not limited to: Trimethylolpropane tris(3-mercaptopropionate); Trimethylolpropane tris(3-mercaptoacetate); 2,3-(dimercaptoethylthio)-1- mercaptopropane.

[0018] In yet other exemplary embodiments, the thiol monomer is a tetrathiol compound comprising four thiol functional groups. Suitable tetrathiol compounds include but are not limited to: pentaerythritol tetrakis)3- mercaptopropionate) (PETMP); Pentaerythritol tetrakis(2-mercaptoacetate).

[0019] As used herein, the term polyalkene / alkyne monomer refers to a compound comprising at least two, and preferably at least three or four unsaturated groups selected from alkene (C=C) or alkyne groups (C C), or combinations thereof. Most preferably the unsaturated groups are alkene groups.

[0020] In exemplary embodiments, the ‘ene’ monomer is a diene compound comprising two alkene functional groups. Suitable diene compounds include but are not limited to: Tri(ethylene glycol) divinyl ether; Trimethylolpropane diallyl ether. 4908P WO As Filed

[0021] In other exemplary embodiments, the ‘ene’ monomer is a triene compound comprising three alkene functional groups. Suitable triene compounds include, but are not limited to, 1 ,3,5-Triallyl-1 ,3,5-triazine-2,4,6(1 H,3H,5H)-trione (TATATO); Trimethylolpropane-tri(norbom-2-ene-5-carboxylate;

[0022] Pentaerythritol-tri(norborn-2-ene-5-carboxylate).

[0023] In yet other exemplary embodiments, the ‘ene’ monomer is a tetraene compound comprising four alkene functional groups. Suitable tetraene compounds include but are not limited to: Pentaerythritol-tetra(norborn-2-ene- 5-carboxylate); Di(trimethylolpropane)tetra-(norborn-2-ene-5-carboxylate).

[0024] In yet other exemplary embodiments, the ‘ene’ monomer is substituted with a polyalkyne compound comprising two or more alkyne functional groups. Suitable diyne compounds include but are not limited to: 1 ,6-heptadiyne and 1 ,7-octadiyne.

[0025] As the skilled reader will readily appreciate, in a preferred embodiment the copolymer is formed via alkene / alkyne hydrothiolation reactions between the thiol functional groups of the thiol monomer(s) and the alkene / alkyne groups of the ‘ene / ’yne monomer(s). In preferred embodiments, the copolymer is formed by UV initiated free-radical addition.

[0026] In yet a further preferred embodiment of the invention the copolymer is an off- stoichiometry thiol-ene / yne polymer (OSTE) that does not include an epoxy component, often referred to as OSTE+.

[0027] We have determined that the use of polymers comprising a thiol : alkene / alkyne functional group ratio within the claimed range, especially with diamond-shaped pillars and a heat manufacturing technique that, postprocessing, removes uncrosslinked, typically excess monomers, by heating. These features improve enzyme assay sensitivity by approx. 19-fold, compared to the use of conventional microfluidic pillars of a round shape and 4908P WO As Filed having a 25% thiol excess (and without any post-processing heating step to remove uncrosslinked monomers).

[0028] In a preferred embodiment of the invention, said pillars are provided as an array and more preferably are diamond shape in cross-section, although conventional shaped pillars may be used. Ideally, the microassay device is of a conventional size and nature. This means it is typically 30-mm-long and 4 mm wide, it includes a sealed microchannel featuring an array of micropillars approx. 13 800 diamond micropillars as shown with exemplary dimensions in Figure 2. The nominal microchannel (and micropillar) height is 200 pm resulting in an approximate total internal volume of approx. 19 pL (which includes the micropillar-free triangular areas at the ends of the pillar array and the connecting channels to the inlet / outlet, but excluding the space occupied by the micropillars). Preferably, the pillar array is sealed with a planar cover layer.

[0029] In yet a further preferred embodiment, said pillars are functionalized by coating the alkene / alkyne-rich surfaces of same with an avidin, such as streptavidin. Ideally, this involves coating the micropillar arrays with biotin and then an avidin, such as streptavidin, by using biotin-(PEG)n-thiol, where n = 4-22, e.g. biotin-PEG4-alkyne (n=4xPEG) or biotin-PEG(1 kDa)-alkyne (n=22xPEG) and both work equally well, (ideally 0.01 to 1 mM, preferably 0.1 mM in ethylene glycol, ideally with 0.01 % to 5%, preferably 1 %, Igracure TPO-L as a photoinitiator). As acknowledged in the afore, the PEG chain can vary at least between four PEG units (ca. 200 Da) to 22 PEG units (ca. 1 kDa). The avidin, ideally streptavidin, is used at about 0.5 - 5000 pg / mL, preferably 0.5 pg / mL streptavidin in PBS. As shown in Figure 1 , this results in the production of micropillars that can bind biotinylated lipid membranes such as microsomes (microsomes are artificial vesicles derived from pieces of endoplasmic reticulum (ER) formed during cell homogenization) or supersomes (recombinantly expressed drug metabolizing enzyme reagents, consisting typically of microsomes prepared from insect cells infected with a virus engineered to express a human CYP isoform). 4908P WO As Filed

[0030] Reference herein to microsomes includes, without limitation, reference to human liver microsomes or human intestinal microsomes and recombinant human liver or intestinal enzymes (typically produced in insect cells and often referred to as supersomes) based thereon.

[0031] In yet a further preferred embodiment of the invention, said microassay device comprises pillars having alkene / alkyne-rich surfaces that are functionalized with an avidin and also having bound thereto biotinylated lipid membranes such as microsomes / supersomes, such as biotinylated commercial human liver or intestinal microsomes. Yet more ideally still, this functionalized and microsome-loaded microassay device is provided in a freeze-dried form.

[0032] Advantageously, we have discovered that the use of the claimed pillar arrays enables us to lyophilize the immobilized biomaterial, inside the microfluidic device, and then re-solubilization of the immobilized biomaterial after a period of storing (at room temperature) - following the freezing - without any disadvantages results. Indeed, we have discovered that after freeze-drying the microassay device of the invention, it can be stored at room temperature for long periods (beyond that of the conventional 2-week period for prior art devices). Indeed, conventional microassay devices have a shelf life of approx. 2 weeks (i.e., in the fridge in wet state). Based upon comparisons with our alternative technology (PCT / EP2024 / 067617) we expect to be able to successfully store our microassay device, after freeze-drying, for at least nine months without a substantial loss of enzyme activity. It would therefore appear that our microassay device, as well as preventing / minimizing enzyme inactivation, also enables long term storage of microfluidic devices loaded with microsomes / supersomes.

[0033] According to a further aspect of the invention there is provided a method for the manufacture of a microassay device including a plurality of pillars for measuring the activity of at least one metabolic enzyme comprising: 4908P WO As Filed i) making or coating said plurality of pillars from or with a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about is from about 100:150 to 100:105, preferably about 100:140 to 100: 110, and more preferably is about 100: 110 or 100: 125; and ii) coating the alkene / alkyne-rich surfaces of the pillars of part i) with an avidin.

[0034] According to a further aspect of the invention there is provided a method for the manufacture of a microassay device including a plurality of pillars for measuring the activity of at least one metabolic enzyme comprising: i) making or coating said plurality of pillars from or with a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about is from about 100:150 to 100:105, preferably about 100:140 to 100: 110, and more preferably is about 100: 110 or 100: 125; ii) coating the alkene / alkyne-rich surfaces of the pillars of part i) with an avidin; and iii) loading the avidin coated pillars of part ii) with biotinylated lipid membranes containing the enzyme to be tested.

[0035] In a preferred method of the invention, part i) involves making or coating said pillars with a copolymer comprising, as copolymerized units, at least one tetrathiol compound and at least one triene compound. In a particularly preferred method of the invention, the tetrathiol compound is PETMP and the triene compound is TATATO.

[0036] In a preferred method of the invention, part ii) of the method involves coating the pillars with biotin-PEGn-thiol (such as 0.1 mM biotin-PEG(1 kDa)-thiol in ethylene glycol). Ideally, said biotin-PEGn-thiol reagent solution comprises a photoinitiatior such as Igracure TPO-L (ideally, within the range of 0.01 % to 4908P WO As Filed

[0037] 5%, preferably 1 %), then the method involves exposing the biotinylated pillars to an avidin (ideally, 0.5 pg / mL streptavidin in PBS).

[0038] In a further preferred, but optional, method of the invention, part i) of the method is followed by heating the device, ideally to 70-110°C for 1 -3h, preferably to 100°C for approx. 2h. This post-processing of the pillars by heat was added to further remove any uncrosslinked monomers before functionalizing the pillars with avidin and the biotinylated lipid membranes.

[0039] In part iii) the biotinylated lipid membranes, containing the enzyme to be tested, are prepared by exposing microsome or supersomes to biotin-containing fusogenic liposomes (b-FL), these b-FL are prepared from commercially available lipids e.g., via Avanti Polar Lipids. Preferably, to prepare unilamellar vesicles for merging the biotin tag to the microsomes or supersomes and attaching them to the avidin coated pillars, the multilamellar liposome mixture is extruded through a polycarbonate membrane (pore size 100 nm) 51 times (suggested supplier protocol) using a benchtop extruder (Avanti Polar Lipids). More preferably still, to prepare biotinylated lipid membranes such as microsomes or supersomes, equal volumes of the b-FL dispersion (2 mg / mL total lipid) and the lipid membrane stock, such as human liver microsomes (HLM), (20 mg / mL total protein) are mixed and incubated at 37°C for 15 min to transfer the biotin tag to the HLMs via spontaneous fusion. These biotinylated lipid membranes containing the enzyme to be tested are then incubated in the device and the device is stored in the fridge until used or freeze-dried for longer storage at room temperature. Alternatively, it is possible to use the ability of the liposomes and the lipids membranes to spontaneously fuse in a different way: by producing the b-FL as afore and then incubating the b-FL product with the avidin coated pillars and then exposing these now pillar tethered liposomes to the lipid membranes containing the enzymes to be tested. This device can then be stored as afore: stored in the fridge until used or freeze-dried for longer storage at room temperature. 4908P WO As Filed

[0040] In an alternative aspect, or a yet a further preferred, method of the invention, particularly where diamond shaped pillars are to be used, the above method may be preceded by the manufacture of the microassay device, and this comprises: a) creating a photolithographic master cast, such as a Sll-8 master cast, by spin coating a first layer of photoresist on an untreated wafer, preferably a silicon wafer, and then soft baking, preferably at about 65°C for about 10 min, then at about 95°C for about 40 min to yield ca. 100-pm-thick layer, b) repeating step a); and c) exposing the product of step b) to UV light for about 30s.

[0041] In this preferred method of the invention, a photomask comprising the pillar array is applied to the product of part b), ideally by vacuum suction, prior to UV exposure.

[0042] Reference herein to UV exposure is ideally to collimated UV light.

[0043] In the preferred preceded method, step a) is repeated once and so the product of step b) comprises thus building the coating up in two steps, whereby two nominally ca. 100 pm layers are created (instead of one-step spin coating of 200-pm-thick layer). In this method, the soft bake is performed separately for the first layer before spin coating of the next layer, and for the second layer the soft bake is performed before UV lithographic patterning (of both layers simultaneously). As a result, the apparent layer thickness uniformity is improved from 6.7% RSD to 0.9% RSD (n=12 data points around the master wafer, as illustrated in Figures 3b and c).

[0044] In an alternative preferred preceded method, step a) comprises building a single 200-pm-thick layer which is then exposed to a separate edge bead removal (EBR) step by spraying acetone to the edge of the silicon substrate (UD-3b dispenser, Laurell Technologies; 800 rpm / 15 s) after spin coating of the SU-8 layer.The afore method creates the master cast (step i of Table 1 ). 4908P WO As Filed

[0045] Casting of the polymethylsiloxane (PDMS) molds (step ii of Table 1 ) is performed by mixing the base elastomer and a curing agent in a ratio of about 10:1 (w / w), degassing in vacuum for about 30 min, and pouring the mixture onto the photolithographic master cast, such as a Sll-8 master cast, prior then curing by heat (using standard protocol of about 70°C for at least 8 hours), as per the standard protocol.

[0046] In yet a further aspect of the invention there is provided a method for measuring the activity of a metabolic enzyme comprising use of the device according to the invention.

[0047] In yet a further aspect of the invention there is provided a method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100: 150 to 100: 105, preferably about 100: 140 to 100: 110, and more preferably is about 100:110 or 100:125 that are functionalized with an avidin and also have bound thereto biotinylated lipid membranes, such as microsomes / supersomes, containing the enzyme to be assayed, to a sample comprising a substrate and / or cofactor for said enzyme; ii) exposing the microassay device of part i) to a therapeutic to be tested; iii) measuring the activity of said enzyme during steps i) and ii); iv) comparing the measured activity in part ii) and with that of the enzyme when not exposed to said therapeutic in part i) and, where the activity is reduced when the therapeutic is present, concluding the therapeutic has an inhibitory effect on the activity of said enzyme. 4908P WO As Filed

[0048] In yet a further aspect of the invention there is provided a method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100: 150 to 100: 105, preferably about 100: 140 to 100: 110, and more preferably is about 100:110 or 100:125 that are functionalized with an avidin, to biotinylated lipid membranes or biotinylated microsomes / supersomes, containing the enzyme to be assayed, to produce a loaded microassay device; ii) exposing the loaded microassay device of part i) to a sample comprising a substrate and / or cofactor for said enzyme; iii) exposing the loaded microassay device of part ii) to a therapeutic to be tested; iv) measuring the activity of said enzyme during ii) and iii); v) comparing the measured activity in part iv) of the enzyme when exposed to said therapeutic and when not exposed to said therapeutic and, where the activity is reduced when exposed to said therapeutic, concluding the therapeutic has an inhibitory effect on the activity of said enzyme.

[0049] In yet a further aspect of the invention there is provided a method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to 4908P WO As Filed alkene / alkyne functional groups in said copolymer is from about 100: 150 to 100: 105, preferably about 100: 140 to 100: 110, and more preferably is about 100:110 or 100:125 that are functionalized with an avidin, to biotinylated lipid membranes; ii) exposing the microassay device of part i) to lipid membranes or microsomes or supersomes, containing the enzyme whose activity is to be measured, to produce a loaded microassay device; iii) exposing the loaded microassay device of part ii) to a sample comprising a substrate and / or cofactor for said enzyme; iv) exposing the loaded microassay device of part iii) to a therapeutic to be tested; v) measuring the activity of said enzyme during iii) and iv); vi) comparing the measured activity in part v) of the enzyme when exposed to said therapeutic and when not exposed to said therapeutic and, where the activity is reduced when exposed to said therapeutic, concluding the therapeutic has an inhibitory effect on the activity of said enzyme.

[0050] In either method for measuring the activity of a metabolic enzyme the said pillars are round or diamond-shaped in cross-section but diamond-shaped in cross-section is preferred. Additionally, or alternatively, the microassay device is provided in a freeze-dried form.

[0051] According to a yet further aspect of the invention there is provided a kit of parts comprising: i) a microassay device comprising a plurality of pillars made from or coated with a copolymer comprising, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional 4908P WO As Filed groups in said copolymer is from about is from about 100:150 to 100: 105, preferably about 100: 140 to 100: 110, and more preferably is about 100:110 or 100:125, and are functionalized by the attachment of avidin thereto; and ii) biotinylated lipid membranes; and / or iii) microsomes / supersomes containing the enzyme to be assayed; and / or; iv) biotinylated microsomes / supersomes containing the enzyme to be assayed.

[0052] In a preferred kit of parts one or more of i), ii), iii) or iv) is / are freeze dried. Additionally, or alternatively, said pillars are diamond-shaped in cross-section.

[0053] Preferred features of each aspect of the invention may be as described in connection with any of the other aspects.

[0054] Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of the words, for example “comprising” and “comprises”, mean “including but not limited to” and do not exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

[0055] All references, including any patent or patent application, cited in this specification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. Further, no admission is made that any of the prior art constitutes part of the common general knowledge in the art.

[0056] Other features of the present invention will become apparent from the following examples. Generally speaking, the invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including the accompanying claims and drawings). Thus, features, integers, characteristics, 4908P WO As Filed compounds or chemical moieties described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein, unless incompatible therewith.

[0057] Moreover, unless stated otherwise, any feature disclosed herein may be replaced by an alternative feature serving the same or a similar purpose.

[0058] The Invention will now be described by way of example only with reference to the Examples below and to the following Figures wherein:

[0059] Figure 1. Upper: Shows a schematic illustration of the biotinylation of microsomes / supersomes with the help of biotinylated fusogenic liposomes (b- FL) to yield biotinylated microsomes / supersomes that can be immobilized on avidin-coated surfaces. Middle: Schematic illustration of the functionalization workflow to obtain avidin-coated micropillars. In this illustration, the diagram refers to alkene-rich [C=C] polymers, by referencing a double bond on the micropillar surface. Lower: Schematic illustration of the microfluidic (flow- through) pillar array (device) and the setup used in enzyme activity determination of enzymes incorporated in microsomes / supersomes, such as cytochromes P450 (CYP) and UDP-glucuronosyl transferases (UGT).

[0060] Figure 2. Upper: Shows the layout of the micropillar reactor and the exemplary critical dimensions (all in mm). Middle: Shows the layout of each type of pillar array with dimensions (all in mm): round (initial), small diamonds, and wide diamonds. Lower: Shows exemplary scanning electron micrographs of each type of pillar array.

[0061] Figure 3. Shows visualizations of (a) the edge bead resulting from spin coating of thick SU-8 layers and its impact on (b) the layer thickness (non)uniformity. Impact of the application of (c) the two-step spin coating protocol (yielding highly uniform / favourable results) and (d) the EBR step on the layer thickness uniformity. Comparison of feature resolution of (e) the initial spin coating and UV curing process which results in rounding of the comers of the diamond- 4908P WO As Filed shaped micropillars and (f) our new process employing EBR step and a vacuum applied photomask, resulting in sharp comers.

[0062] Figure 4. Shows the combined impact of bulk polymer composition on cumulative CYP enzyme activity. A Bulk polymer composition of 25% molar excess of alkene / alkyne (allyl), pillar shape (wide diamond), and postprocessing by heat (100°C, 2h) on the cumulative (1 h) CYP enzyme activity of immobilized HLMs compared with bulk polymer composition of 25% molar excess of thiols, round pillar shape, and no post-processing). B Bulk polymer composition of 25% molar excess of alkene / alkyne (allyl), pillar shape (wide diamond), and post-processing by heat (100°C, 2h) on the cumulative (1 h) CYP enzyme activity of immobilized HLMs compared with bulk polymer composition of 25% molar excess of thiols, round pillar shape, and no postprocessing [as per A] as well as including an alternative comparison using a bulk polymer composition with a 10% molar excess of thiols, wide diamond pillar shape, and identical post-processing. The data was collected in flow- through conditions at a flow rate of 5 pL / min and averaged from n=7 (new protocol) or n=4 (initial protocol) chips.

[0063] Figure 5. Shows impact of postprocessing conditions on the cumulative (1 h) CYP enzyme activity of immobilized HLMs in flow-through conditions at a flow rate of 5 pL / min. Two temperature (70°C and 100°C) and two heating times (2h vs. 4h) were compared. Temperatures >130°C resulted in mechanical damage to the bulk polymer. Times >4h did not improve the performance. Data was acquired using a thiol-rich formulation, but a similar trend (although of a different magnitude) is assumed using an allyl-rich formulation in respect of any residual thiol monomers. In addition, the postprocessing will eliminate uncrosslinked alkenes / alkynes preventing and reducing any other unwanted effects of extra monomers in the reaction solution.

[0064] Figure 6. Shows the impact of bulk polymer composition (25% vs. 10% molar excess of alkenes / alkynes (allyls) on the cumulative (1 h) enzyme activity of immobilized HLMs in flow-through conditions at a flow rate of 5 pL / min: (A) 4908P WO As Filed

[0065] CYP activity, (B) UGT activity. The data is averaged from n=7 (25% molar excess) or n=8 (10% molar excess) chips.

[0066] Figure 7. Shows impact of the pillar array design on the cumulative (1 h) CYP enzyme activity of immobilized HLMs in flow-through conditions at a flow rate of 5 pL / min. Data was acquired using a thiol-rich formulation, but a similar trend is assumed in respect of allyl-rich formulation due to maximizing the surface- to-volume ratio of the micropillar arrays and beneficial flow dynamics.

[0067] Figure 8. Shows the transient marker metabolite concentrations in the effluents of the microfluidic devices, implemented using the optimized protocol (two different alkene / alkyne [allyl-rich] compositions) and functionalized with biotinylated human liver microsomes, for selected CYP (upper) and UGT (lower) marker reactions in flow-through assays performed at 5 pL / min as described in Table 4, demonstrating good stability over time: (A) CYP activity using a polymer with either a 25% or 10% allyl excess monomer (and including post-processing heating), in comparison to the initial protocol (a polymer with 25% thiol monomer molar excess, round pillars, no post-processing). (B) UGT activity using a polymer with either a 25% or 10% allyl excess monomer (and including post-processing heating).

[0068] The data is averaged from n=7 (+25% allyls), n=8 (+10% allyls) or n=4 (+25% thiols).

[0069] MATERIALS & METHODS

[0070] 2.1.1 Microreactor design

[0071] The conventional microreactor design comprises of a 30-mm-long and 4 mm wide, sealed microchannel featuring an array of ca. 14 400 round-shaped micropillars and ca. 360 semicircular micropillar structures on the channel walls (Figure 2). The nominal microchannel (and micropillar) height is 200 pm resulting in approximate total internal volume of ca. 25 pL, which includes the micropillar-free triangular areas at the ends of the pillar array and the connecting channels to the inlet / outlet, and excluding the space occupied by the micropillars. The pillar array is sealed with a planar cover layer. 4908P WO As Filed

[0072] Alternatively, new microreactor designs were used comprising micropillar arrays of the same size and about same total micropillar amount (precisely 13 840 and 346 half-pillars on the walls) as the initial design, but with diamondshaped, instead of round, micropillars. The impact of micropillar shape on the CYP enzyme activity level is compared between the initial round shape (0 50 pm) and two different diamond shapes, including ‘small diamonds’ with equal perimeter to that of the round pillars (Figure 2), and ‘wide diamonds’ with ca. 43% increase in perimeter (and thus, ca. 33% in the total surface-over-volume ratio) compared with that of the round pillars (Figure 2). The approximate free internal volumes of the microchannels incorporating ‘small diamond’ and ‘wide diamond’ pillar arrays are ca. 22 pL and 19 pL, respectively.

[0073] 2.1.2 Microfabrication protocol

[0074] Microfabrication of the thiol-ene based micropillar reactors, essentially, comprises a four-step protocol including:

[0075] (i) UV-lithography based fabrication of the initial Sll-8 masters;

[0076] (ii) casting of the PDMS molds (soft lithography) for the micropillar and cover layers;

[0077] (iii) UV replica-molding; and

[0078] (iv) bonding of the thiol-ene based micropillar and cover layers by lamination. The comparison of the initial and re-designed microfabrication protocols is described in Table 1.

[0079] Table 1. Comparison of the critical steps of initial (starting point) and redesigned (new) microfabrication protocols. 4908P WO As Filed

[0080] With reference to Table 1 , (step i) for fabrication of the photolithographic master cast according to the invention, the photolithographic polymer is spin coated on an untreated silicon wafer. The spin coating involves either spin coating two layers each: at 500 rpm / 40s, then 2500 rpm / 30s; 100 pm per layer thick, and baking at 65°C / 10 min, then 95°C / 40 min

[0081] OR spin coating one layer: 500 rpm / 40s, then 1150 rpm / 30s +800 rpm / 15s; 200 pm layer thick (and undertaking edge bead removal with acetone) and baking 65°C / 25 min, then 95°C / 110 min.

[0082] Either method produces a ca. 200-pm-thick layer, which is then exposed through a plastic photomask, attached to the photoresist layer using vacuum suction, under collimated UV light for 30 s (OAI LS 30 / 5, OAI Instrument; nominal intensity of 40 mW / cm2).

[0083] After UV exposure, the master is post exposure baked (for example 95°C / 40min). 4908P WO As Filed

[0084] The exposed microstructures are then developed in propylene glycol methyl ether acetate for 20 min with stirring, rinsed with isopropanol, and dried with nitrogen gas, followed by a conditioning bake (95 °C / 30 min).

[0085] When fabricating the masters using diamond-shaped micropillar arrays, the initial (standard) single spin coating protocol yields insufficient feature resolution, which necessitates customization of the spin coating, softbake and UV exposure steps of the standard master fabrication process, whereas the later parts (post exposure bake and conditioning bake) follow the standard procedure. The impact of the spin coating protocol on the quality of the diamond-shaped micropillar arrays is described in section 2.2.1.

[0086] Casting of the PDMS molds (step ii) is performed by mixing the base elastomer and the curing agent in an approximate ratio of 10:1 (w / w), degassing in vacuum for about 30 min, and pouring the mixture onto the Sll-8 master(s) prior to curing according to standard protocol (e.g., 70°C, overnight).

[0087] UV replica molding of the thiol-ene polymer layers (step iii) is prepared by mixing the PETMP (tetrathiol) and TATATO (triene) monomers in the claimed molar ratio (with respect to free thiol and ‘ene’ functional groups) and the mixture is poured onto the PDMS mold, degassed in vacuum, and cured under UV for 5 min (Dymax 5000-EC Series UV flood exposure lamp, Dymax Corporation; nominal power 225 mW cm’2). In the conventional protocol, the molar ratio of thiol over ‘ene’ functional groups was 125:100. In our new protocol, the molar ratio is optimized to obtain alkene / alkyne-rich bulk composition with minimal impact of free monomers. In fact, a post-processing by heat step was additionally used as part of the process to remove uncrosslinked monomers before bonding. Both micropillar and cover layers (with inlet / outlet holes) are prepared in the same manner. The impact of monomer ratio (bulk composition) and post-processing conditions on the CYP activity and stability are elaborated in section 2.2.2. 4908P WO As Filed

[0088] The bonding (step iv) is performed according to the prior art protocol, or standard protocol, ideally, by preheating the micropillar and cover layers before lamination and UV exposure (Table 1 ).

[0089] 2.1.3 Microreactor functionalization with human liver microsomes

[0090] The microreactor functionalization comprises two steps as previously described (Figure 1 ): (i) biotinylation of commercial human liver microsomes (Corning® Gentest 20-Donor Pool, BD Biosciences) with the help of biotinylated fusogenic liposomes, and (ii) sequential coating of the micropillar arrays with biotin-PEG1 kDa-thiol, streptavidin, and biotinylated HLM. In an alternative protocol, biotin-PEG1 kDa-thiol can be replaced by biotin-PEGn- thiol, where n=4-22. In yet another alternative protocol, HLMs can be replaced by human intestinal microsomes or human recombinant CYP / UGT enzymes (supersomes).

[0091] Preparation of the biotinylated fusogenic liposomes preparation and fusion with lipid membranes (step i): The lipids used for preparation of biotin-containing fusogenic liposomes (b-FL) are from Avanti Polar Lipids and include 1 ,2-dioleoyl-3-trimethylammonium-propane (chloride salt) (DOTAP), 1 ,2-dioleoyl-sn-glycero-3-phophoethanolamine (DOPE), 1 ,2-dioleoyl-sn- glycero-3-phophoethanolamine-N-(Cap biotinyl) (sodium salt) (biotin-cap- DOPE), and 1 ,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt) (Lissamine Rhodamine B-DOPE).

[0092] The b-FLs are prepared by mixing stock solutions (each in chloroform) of DOPE (10 mg / mL), DOTAP (10 mg / mL), biotin-cap-DOPE (10 mg / mL), and Lissamine Rhodamine B-DOPE (1 mg / mL) in a lipid mass ratio of 1 : 1 : 0.1 : 0.05, respectively.

[0093] After mixing, the bulk solvent is evaporated under a stream of nitrogen. To remove residual solvent, the lipid mixture is kept in vacuum for 2 h. Next, the dry lipid film is solvated in PBS to yield a total lipid concentration of 2 mg / mL and vortexed for 1 h at room temperature. To prepare unilamellar vesicles, the 4908P WO As Filed multilamellar liposome mixture is extruded through a polycarbonate membrane (pore size 100 nm) 51 times using a benchtop extruder (Avanti Polar Lipids).

[0094] For HLM biotinylation, equal volumes of the b-FL dispersion (2 mg / mL total lipid) and the HLM stock solution (20 mg / mL total protein) are mixed and incubated at 37°C for 15 min to transfer the biotin tag to the HLMs via spontaneous fusion.

[0095] Alternatively, the biotinylated liposomes are exposed to streptavidin functionalized pillars where they are retained and so available for spontaneous fusion when subsequently exposed to microsomes / supersomes.

[0096] Alternatively, human intestinal microsomes (HIM), typically sold as a stock of 10 mg / mL total protein) or human recombinant enzymes (supersomes) can be / are used instead of HLMs, and so reference above to HLM above is substituted for reference to HIM or supersomes.

[0097] Functionalization of the micropillar reactors (step ii) is performed by coating the alkene / alkyne-rich surfaces first with biotin-PEG1 kDa-thiol (0.1 mM in ethylene glycol, optionally but preferably, also with 1 % Igracure TPO-L as a photoinitiator) and using photopolymerization under UV-LED (A = 365 nm) for 1 min. Next, the microreactor is sequentially rinsed with methanol and deionized water (ca. 2 mL each), filled with Alexa Fluor® labeled Streptavidin (0.5 pg / mL in PBS) and incubated at room temperature for 40 min. Next, the micropillar channel is rinsed with ca. 2 mL of PBS and filled with the b-HLM suspension (10 mg / mL total protein) and incubated at 4°C overnight (or until used, but no longer than two weeks at maximum unless the device is freeze- dried).

[0098] In an alternative, biotin-PEGn-thiol (n=4-22) can be used instead of biotin- PEG1 kDa-thiol.

[0099] 2.1.4 Determination of the CYP activity in flow-through conditions 4908P WO As Filed

[0100] Before use, the microreactors featuring immobilized biotinylated microsomes were equilibrated to room temperature and rinsed with ca. 2-5 mL of fresh PBS (to remove any excess of unimmobilized microsomes). Next, the microreactor is assembled with a syringe pump with the help of nanoport fluidic connectors and capillaries and placed on a PID-controlled heating block at 37°C. The CYP activities of the b-HLMs immobilized were assessed by infusing a feed solution into the microreactor at a constant flow rate (here, 5 pL / min) and collecting fractions of the effluent (a 50 pL) every 10 min with the help of CMA 470 refrigerated fraction collector (Harvard Apparatus). The feed solution contains a CYP2C9-spesific, prelum inescent model substrate (Luciferin-H, 200 pM) and a CYP cosubstrate, NADPH (1 mM), in PBS (pH 7.4). After the experiment, the fractions collected from the microreactor are mixed with 50 pL of a detection reagent (P450-Glo Assay, Promega), incubated for 20 min, and analysed offline for luminescence (arising from the Luciferin metabolite produced in CYP2C9 reaction) using Varioskan LUX well plate reader (Thermo Scientific).

[0101] The enzyme activity (in arbitrary units) is determined by subtracting the background (feed solution) luminescence from the collected (effluent) fractions to quantitate the concentration of the reaction product (Luciferin metabolite) of the marker molecule (Luciferin-H) in each collected fraction. The cumulative enzyme activity is calculated by summing the so obtained metabolite amounts (in arbitrary units) from six subsequent fractions (a 10 min) for each replicate chip. When initial number of samples of a given set is four or higher, results deviating by >1.25 (standard deviations) from mean are excluded as outliers and not included in the final calculated mean + / - standard deviation.

[0102] The enzyme kinetics (moles per time) can be calculated by quantitation the metabolite concentration with the help of a Luciferin (metabolite standard) calibration curve and subsequently multiplying the metabolite concentration with the flow rate in order to proportion the metabolite yield to the reaction (residence) time. 4908P WO As Filed

[0103] The different experimental conditions used to assay different metabolic enzymes besides CYP2C9 described herein, are shown in Table 4.

[0104] 2.1.5 Determination of the UGT activity in flow-through conditions

[0105] To assess the UGT activity, the microreactors were prepared as described in section 2.1.4, but the feed solution contained a non-specific UGT model substrate (8-hydroxyquinoline, 50 pM) as well as a UGT co-substrate, UDPGA (1 mM), in 0.1 M Tris buffer with 5 mM MgCI2 (pH 7.5). The fractions (100 pL) were collected every 20 min and mixed with 10 pL of 4 M HCIO4 before analysis by fluorescence (ex. 245 nm, em. 475 nm) using Varioskan LUX well plate reader (Thermo Scientific). The subsequent calculation of the enzyme activity follows the same principle as described in 2.1.4.

[0106] 2.1.6 Determination of other enzyme activities in flow-through conditions

[0107] Different metabolic enzymes besides CYP2C9 and UGT can be measured in the same way by replacing the isoenzyme-specific marker molecule in the feed and adjusting its concentration based on marker-specific enzyme affinity (KM). The flow-through setup enables one to use any known CYP or UGT specific marker molecule known in the prior art, including those recommended by the regulatory authorities (e.g., FDA Drug Development and Drug Interactions | Table of Substrates, Inhibitors and Inducers - http s: / / www.fda.gov / drugs / drug- interactions-labeling / drug-development-and-drug-interactions-table-substrates-inhibitors-and- inducers or the EMA equivalent Guideline on the investigation of drug interactions, https: / / www.ema.europa.eu / en / documents / scientific~guideline / guideline~ investigation-drug-interactions-revision-1 en.pdf).

[0108] 2.2 Results

[0109] 2.2.1 Impact of our new master fabrication protocol on the quality of the micropillar structures

[0110] Microfabrication of thick (here, nominally 200 pm) and high aspect ratio microstructures (here, ca. 4:1 for the micropillar arrays) using nonstandard methodology. 4908P WO As Filed

[0111] The spin coating and soft bake steps critically affect the layer thickness and uniformity (i.e. , the micropillar height uniformity), because of the so-called edge bead effect. Edge bead refers to a build-up of the coated photoresist (Sll-8) at the edge of the (silicon) substrate, as illustrated in Figure 3a. During the softbake, the edge bead transforms into an uneven Sll-8 layer thickness around the silicon wafer (Figure 3b), which further impairs the lithographic patterning resolution. In the new microfabrication protocol, the edge bead effect is minimized by spin coating the Sll-8 100 photoresist onto the silicon wafer in two steps, each nominally ca. 100 pm layer (instead of one-step spin coating of 200-pm-thick layer). In this case, the soft bake is performed separately for the first layer before spin coating of the next, and for the second layer before UV lithographic patterning (of both layers simultaneously). As a result, the apparent layer thickness uniformity is improved from 6.7% RSD to 0.9% RSD (n=12 data points around the master wafer, as illustrated in Figures 3b and c). Alternatively, a separate edge bead removal (EBR) step can be adapted as part of the master fabrication protocol by spraying acetone to the edge of the silicon substrate (UD-3b dispenser, Laurell Technologies; 800 rpm / 15 s) after spin coating of the Sll-8 layer. In our process, application of the EBR step to the one-step spin coating protocol improves the apparent layer thickness uniformity from the initial 6.7% RSD to 4.7% RSD (n=12 data points around the master wafer, as illustrated in Figures 3b and d. In addition to EBR, the way in which the photomask is attached to the soft-baked photoresist layer before UV curing has an impact on the feature resolution, which is additionally illustrated in Figures 3e (poorer resolution, initial process) and 3f (better resolution, new process comprising vacuum suction).

[0112] On the basis of these results, the master fabrication protocol was customized by adapting the two-step or two-layer spin coating protocol (Table 1 ) to ensure sufficiently high feature resolution in diamond-shaped micropillar arrays.

[0113] 2.2.2 Impact of bulk polymer composition and post-processing on CYP activity in flow-through experiments 4908P WO As Filed

[0114] Functionalization of the micropillar reactors with microsomes (e.g., human liver or intestine) is based on specific, multi-step protocol built on thiol-ene polymer. In the first step, a biotin-PEGn-thiol (here, biotin-PEG1 kDa-thiol) is covalently bound to the free surface alkenes / alkynes of an alkene / alkyne-rich thiol-ene polymer (here, +25% or +10% molar excess of alkene / alkyne (allyl) over thiol functional groups). Additionally, the use of off-stoichiometric (alkene / alkyne- rich) bulk composition safeguards against any uncrosslinked tetrathiol (PETMP) monomers into the microchannel inactivating the immobilized CYP enzymes (residing on the cytosolic side of the membrane) in an irreversible manner. However, it can result in uncrosslinked alkene / alkyne monomers in the microchannel which can have unwanted effects.

[0115] Accordingly, an additional post-processing protocol is used as part of the microfabrication of thiol-ene microreactors (Table 1 ) to ensure elimination of any uncrosslinked monomers from the bulk. The post-processing is performed between the thiol-ene replica-molding and the bonding steps for functionalizing the pillar surfaces, by heat-treating the cured thiol-ene replicas at temperatures (70 - 100°C for 2h or 4h (last step of Table 1 ) prior to functionalization and enzyme immobilization.

[0116] The combined impact of these manufacturing changes, including adaptation of an alkene / alkyne-rich (allyl-rich) composition and post-processing by heat (100°C, 2h) as well as use of the wide diamond pillar shape (instead of round pillars, Figure 2) on the cumulative (1 h) CYP activity is demonstrated in Figure 4A. Here, as a first example, we compare a thiol-rich composition (+25 mol-% excess of thiols over allyls), because we have shown uncrosslinked thiol monomers inhibit CYPs; with an alkene / alkyne-rich (allyl-rich) composition, where, logically, the amount of uncrosslinked thiol monomer is minimal, after the initial UV curing, but the postprocessing by heat will nevertheless help eliminate any trace level thiols, as well as any excess of uncrosslinked alkene / alkyne monomers. Compared with the initial protocol, these amendments increase the cumulative (1 h) CYP activity by a factor of >19. 4908P WO As Filed

[0117] Additionally, Figure 4B shows a comparison where the thiol-rich composition of A and the allyl-rich composition of A are compared with an alternative technology using a minimal amount (+10%) thiol excess and a manufacturing protocol including heat post-processing of diamond shaped pillars. It can be seen that use of the allyl-rich formulation increases the cumulative CYP activity by ca. one-third and reduces assay-to-assay variation from ca. 45% (n=4) to ca. 20% (n=6) (Figure 5).

[0118] The impact of polymer post-processing temperature (70 or 100°C) on the model CYP enzyme activity is illustrated in Figure 5, suggesting that the higher post-processing temperature more effectively vaporizes any uncrosslinked monomers during post-processing prior to functionalization, which prevents any uncrosslinked monomers from interfering with the activity of the enzyme to be measured. Temperatures >130°C resulted in mechanical damage to the bulk polymer (qualitative finding).

[0119] The impact of heating time (2h vs. 4h) on the model CYP enzyme activity is illustrated in Figure 5. This data suggests that extending the heat-treatment time beyond 2h is unnecessary and does not result in substantial improvement in the performance of the microreactor.

[0120] On the basis of these results, it can be concluded that heat treatment of the replica-molded thiol-ene parts at 100°C for 2h is the overall best protocol for maintaining the enzyme activity and material stability (affecting bonding).

[0121] 2.2.3 Impact of alkene / alkyne (allyl) excess in the bulk polymer composition on the CYP and UGT activities in flow-through experiments

[0122] The impacts of the alkene / alkyne excess in the bulk polymer composition on the model CYP and UGT enzyme activities are illustrated in Figure 6. The results suggest that reducing the excess of allyl functional groups in the bulk from initial +25 mol-% to about +10 mol-% does not have a statistically significant impact (p>0.05) on the cumulative (1 h) CYP activity (Figure 6A) or 4908P WO As Filed the cumulative (1 h) UGT activity (Figure 6B). However, reducing the molar excess of allyls in the bulk polymer from 25% to 10% may result in a higher variation between replicate chips, and an apparently lower average activity (likely due to less functional groups available for initial biotin-PEG1 kDa-thiol binding). Moreover, increasing the allyl excess >50% will yield mechanically too rigid and stiff microstructures, which compromises the bonding of the micropillar and cover layers (qualitative finding), thus indicating that the preferred allyl excess is about 25 mol-%.

[0123] With our invention the relative enzyme activity (decay) is identical to that of inherent CYP decay (also observed in static enzyme incubations) and features a somewhat linear decline over time with half-life of 70-100 min.

[0124] 2.2.4 Impact of micropillar shape on the CYP activity in flow-through experiments

[0125] Maximizing the surface-to-volume ratio of the micropillar arrays provides additional means to further increase the apparent ‘concentration’ of the immobilized microsomes, and thus, the initial CYP enzyme activity level. From the microfabrication perspective, the density of the micropillar array (ca. 120 pillars / mm2) cannot be much increased without compromising the replicamolding process (qualitative finding). However, the total surface area can be impacted by altering the micropillar shape.

[0126] The impact of the micropillar shape on the model CYP enzyme activity is illustrated in Figure 7, indicating that changing the micropillars from the initial round shape to the diamond shape, both featuring the same pillar perimeter (identical surface area), already increases the cumulative (1 h) CYP activity by about 40% (small diamonds, Figure 7). Such increase is likely associated with the different fluid dynamics in each type of pillar array, as the diamond shape effectively eliminates the ‘silent (no flow) zone’ behind the micropillars (in flow direction), which is characteristic for the round shape pillars. 4908P WO As Filed

[0127] Increasing the perimeter (surface area) of the (wide / elongated) diamond arrays by about 43% further increases the cumulative (1 h) CYP activity by about 70% (wide diamonds, Figure 7) compared with round pillars.

[0128] When using the optimum bulk polymer composition (thiol : alkene 100:125 (PETMP:TATATO), Figures 4 & 6) and post-processing conditions (100°C / 2h, Figure 5), and the wide diamond pillar shape (Figure 7), the combined increase in the cumulative (1 h) CYP activity compared with the initial protocol (ca. 0.19E+05 a.u.) is ca. 19-fold (Figure 6, ca. 3.71 E+05 a.u.) for the preferred functionalization protocol using biotin-PEG(1 kDa)-thiol, streptavidin, and biotinylated human liver microsomes (HLMs). When using the alternative protocol with biotin-PEG4-thiol, streptavidin, and biotinylated HLMs, the increase compared with the initial / conventional protocol is likely to be comparable.

[0129] Finally, the validity of the optimized protocol for measuring transient liver microsomal enzyme activities (here, CYP and UGT) is demonstrated under flow-through conditions, indicating good stability of the enzymatic marker activities over time (Figure 8).

[0130] Testing of the inhibitory effects of therapeutics using the microassay device

[0131] The feed solution contains a CYP marker substrate (isoenzyme-specific) and a cofactor (NADPH), and the possible inhibitor (the test pharmaceutical). To measure the inhibitory impacts of the test pharmaceutical, its concentration in the feed is gradually increased, during the course of the assay, and the enzyme activity measured based on the conversion of the marker substrate into its specific (marker) metabolite. To measure the inhibition mechanism, i.e. , whether the inhibitory impact is reversible or irreversible, the pharmaceutical is eventually excluded from the feed and the recovery of the inhibited enzyme activity is measured (based on the marker substrate conversion). The enzymatic reaction time in the flow-through system is controlled by the flow rate (and the internal volume of the microfluidic device), which defines the 4908P WO As Filed residence time of the feed solution inside the microfluidic device. In the inhibition assays, the flow rate is kept constant over time and adjusted so that the enzyme kinetics of the marker reaction follows first order kinetics (i.e., marker metabolite concentration in the effluent is linearly proportional to the residence time).

[0132] As the invention facilitates the measurement of enzyme activities under flow- through conditions, the system most importantly enables examination of not only the inhibitory concentrations of pharmaceuticals on the given enzyme activities (particularly the CYPs are prone to inhibition by pharmaceuticals) but also the inhibition mechanisms (whether reversible or irreversible) in a manner that is not feasible for current state-of-the-art static in vitro assays.

[0133] Furthermore, by excluding the marker substrate and the co-factor from the feed, which are necessary for monitoring the enzyme activities, but including the therapeutic / pharmaceutical, it is also possible to use the same microfluidic device for assessment of the nonspecific binding of the given therapeutic / pharmaceutical to the immobilized microsomes in order to determine so called free ‘unbound fraction’ (fu). Besides the inhibitory constants (IC50) and inhibition mechanism, the fu is another critical factor to correct for the in vitro metabolic clearance rate (CLint) of pharmaceuticals, in the absence of cofactor triggered metabolism. The current equivalent state-of- the-art static assay for determining the fu is a static assay, called rapid equilibrium dialysis.

[0134] 3. CONCLUSIONS

[0135] Overall, the reported adjustments to the microfabrication protocol, including customization of the master fabrication: optimization of the thiol-ene ratio of the polymer composition; production of wide diamond-shaped pillars and postprocessing of the thiol-ene replicas by heating, the model CYP2C9 activity increases by a factor of >19 compared with the initial protocol (25% molar excess of thiols, round pillars and no post-processing heating) Figure 4. 4908P WO As Filed

[0136] Preparing the functionalized microfluidic devices for freeze-drying

[0137] After incubating the microsome-coated devices in fridge (2-6°C) for 5 or 7 days, the microfluidic devices are rinsed and filled with 10% (wA / ) sucrose solution (a cryoprotection reagent) and stored overnight before loading into the freezedryer (FTS LyoStar II freeze dryer (SP Industries, Inc., Stone Ridge, NY).

[0138] The freeze-drying protocol

[0139] The freeze-drying protocol comprises four steps (Table 3). The optimized conditions used in each step are given in Table 3. After freeze-drying, the microreactors are kept at room temperature, protected from light and moisture, until use.

[0140] Further adjustment and optimization of these conditions may have an impact on the overall performance (recovered enzyme activity), but the method presented herein provides the evidence of the feasibility of freeze-drying to lyophilization of immobilized microsomes in order to extend the shelf life (storing stability) of the microfluidic devices compared with storing in wetted state in fridge.

[0141] Table 3. The freeze-drying protocol used in cryopreservation experiments enabling long-term storage.

[0142] After manufacture and functionalization, the microfluidic devices are stored in a ‘wet’ state in a fridge, i.e., after immobilization of the pre-biotinylated microsomes and until use (enzyme activity testing). In this manner, the CYP 4908P WO As Filed and UGT activities are at their highest about 7-9 days after storage, after which the activities start to decline and eventually fade within ca. 3 weeks. The impact of storing time has been evaluated using Luciferin-H (CYP2C9-specific marker, section 2.1.4) and 8-hydroquinoline (nonselective UGT marker, Section 2.1.5) as marker substrates representative of the CYP and UGT activities.

[0143] By using the freeze-drying protocol presented above, the microfluidic devices can be stored at room temperature in dry state until use, and the storing time can be extended to several months. The impact of freeze-drying on the recovered enzyme activities was assessed with three similarly prepared batches of microfluidic devices, the first one of which had been stored for >12 months, the second for >9 months, and the third for approx. 6 months at this point. At best, recovered enzyme activities similar to control activity can be achieved even after 9 months of storing in room temperature after freezedrying. The impact of storing time on the recovered enzyme activities has also been evaluated using Luciferin-H (CYP2C9) and 8-hydroxyquinoline (UGT) as marker substrates representative of the CYP and UGT activities.

[0144] DETERMINATION OF THE STABILITY OF CYP AND UGT ACTIVITIES DURING USE

[0145] The human liver microsomes (HLMs) incorporate several different CYP (cytosolic) and UGT (luminal) enzyme isoforms. The activity of selected CYP isoforms, known to be most important to elimination (metabolism) of active pharmaceutical ingredients, can be determined using enzyme specific marker substrates (Table 4). In addition, the liver microsomal UGT enzyme activities can be evaluated using the nonselective UGT marker substrate, 8- hydroxyquinoline, known to metabolize via several UGT isoforms (Table 4) or any other UGT-selective substrate. Furthermore, the microsomal CYP and UGT enzyme activities in the intestine can be evaluated using the same marker substrates, but replacing human liver microsomes with human intestinal microsomes in the functionalization protocol. 4908P WO As Filed

[0146] Before use, the microfluidic devices incorporating the immobilized microsomes are equilibrated to room temperature and rinsed with approx. 2-5 mL of fresh run buffer. Next, the microfluidic device was coupled to a pressure-induced pumping system with the help of nanoport fluidic connectors and capillaries and placed on a heating block or a heated chip housing at 37°C. The CYP and UGT activities of the immobilized microsomes are assessed by infusing a feed solution containing the enzyme-specific marker substrate and the co-substrate in the run buffer, at a constant flow rate. The enzyme activities are quantified by analyzing the effluent of the microfluidic devices for the metabolites produced from the marker enzyme substrates by the microsomal enzymes.

[0147] The effluent can be collected and appropriately fractionated using an automated fraction collector (here, CMA 470, Harvard Apparatus), and the samples are prepared for analysis, as indicated in Table 4. Using the assay protocols described herein, enzyme stability for selected CYP and UGT enzymes can be measured, as exemplified in Figure 8.

[0148] Table 4. Examples of experimental conditions for determination ofCYP and

[0149] UGT enzyme activities. 4908P WO As Filed

[0150] Experiments using the new microassay device to determine enzyme inhibition and its mechanism (reversible vs. irreversible) To assess the inhibitory effects of therapeutic drugs on CYP activities in the immobilized microsomes, the microassays can be undertaken using the described protocol (Table 4). The transient CYP enzyme activities are assessed similar to enzyme stability during use, except for the fact that the test substance (therapeutic) is introduced to the microassay step-wise, increasing concentration (typically 0.01 to 100 pmol / L). Typically, three fractions were collected before introducing the therapeutic, and thereafter three fractions at each concentration level. The first fraction always represents the intermediate state, whereas the next two represent the equilibrium of the new condition. After collecting three fractions at the highest concentration, the test substance is withdrawn from the feed solution to assess if the inhibited enzyme activity recovers (reversible inhibition) or persists (irreversible inhibition).

Claims

4908P WO As FiledCLAIMS1. A microassay device comprising a plurality of pillars made from or coated with a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer (‘thiol monomer’) and at least one polyalkene / alkyne monomer (‘ene / yne’ monomer’), wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100:110, and more preferably is about 100: 110 or 100: 1252. The microassay device according to claim 1 wherein the thiol monomer is a dithiol, trithiol or a tetrathiol compound comprising two, three or four thiol functional groups.

3. The microassay device according to claim 1 wherein the polythiol compound is selected from the group comprising: 1 ,6-hexanedithiol;2.5-dimercaptomethyl-1 ,4-dithiane; 2,3-dimercapto-1 -propanol;Benzene-1 ,2-dithiol; 1 ,8-octanedithiol; Ethylene glycol bis(3- mercaptopropionate);Trimethylolpropane tris(3-mercaptopropionate); Trimethylolpropane tris(3-mercaptoacetate); 2,3-(dimercaptoethylthio)- 1 -mercaptopropane; pentaerythritol tetrakis)3-mercaptopropionate) (PETMP); and Pentaerythritol tetrakis(2-mercaptoacetate) .

4. The microassay device according to any one of claims 1 - 3 wherein the polyalkene / alkyne monomer comprises a compound comprising at least two, and preferably at least three or four alkene (C=C) or alkyne (C=C) functional groups.

5. The microassay device according to claim 4 wherein the polyalkene / alkyne compound is selected from the group comprising:1 .3.5-Triallyl-1 ,3,5-triazine-2,4,6(1 H,3H,5H)-trione, TATATO.Tri(ethylene glycol) divinyl ether; Trimethylolpropane diallyl ether; 1 ,3,5- Triallyl-1 ,3,5-triazine-2,4,6(1 H,3H,5H)-trione (TATATO);Trimethylolpropane-tri(norborn-2-ene-5-carboxylate; Pentaerythritol- tri(norborn-2-ene-5-carboxylate);Pentaerythritol-tetra(norborn-2-ene-5-354908P WO As Filed carboxylate); andDi(trimethylolpropane)tetra-(norborn-2-ene-5- carboxylate); 1 ,6-heptadiyne; 1 ,7-octadiyne.

6. The microassay device according to any one of claims 1 -5 wherein the copolymer is an off-stoichiometry thiol-ene / yne polymer (OSTE) that does not include an epoxy component.

7. The microassay device according to any one of claims 1 -6 wherein said pillars are provided as an array and are diamond shape in crosssection.

8. The microassay device according to claim 7 comprising approx. 13 800 diamond-shaped in cross-section micropillars.

9. The microassay device according to any one of claims 1 - 8 wherein said pillars are functionalized by coating the alkene / alkyne-rich surfaces of same with an avidin.

10. The microassay device according to claim 9 wherein the pillars are coated with biotin-PEGn-thiol / alkene / alkyne and then an avidin.11 . The microassay device according to any one of claims 1 - 10 wherein the pillars are also coated with biotinylated lipid membranes containing the enzyme to be assayed.

12. The microassay according to claim 11 wherein the lipid membranes are microsomes or supersomes.

13. The microassay according to claim 11 or claim 12 wherein the lipid membranes are derived from liver or intestinal tissue.

14. The microassay device according to any one of claims 1 - 13 wherein the device has been freeze-dried.

15. A method for the manufacture of a microassay device including a plurality of pillars for measuring the activity of at least one metabolic enzyme comprising: i) making or coating said plurality of pillars from or with a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about364908P WO As Filed100:140 to 100:110, and more preferably is about 100:110 or 100:125; and ii) coating the allene / alkyne-rich surfaces of the pillars of part i) with an avidin.

16. The method according to claim 15 wherein step i) involves making said pillars from, or coating said pillars with, the tetrathiol compound PETMP and the triene compound TATATO.

17. The method according to anyone of claims 15 - 16 wherein step ii) of the method involves coating the pillars with a biotin-PEGn-thiol.

18. The method according to claims 17 wherein said biotin-PEGn-thiol reagent solution also comprises a photoinitiatior such as Igracure TPO- L.

19. The method according to anyone of claims 15 - 18 wherein step i) is followed by: heating the device to 100°C for approx. 2h before step ii) is performed.

20. The method according to any one of claims 15 - 19 wherein step ii) is followed by: iii) loading the avidin coated pillars with biotylinated lipid membranes or biotylinated lipid microsomes / supersomes containing the enzyme to be tested.

21. The method according to claim 20 wherein the biotinylated lipid membranes or microsomes or supersomes containing the enzyme to be tested are prepared by exposing said membranes microsomes or supersomes to biotin-containing fusogenic liposomes (b-FL).

22. The method according to any one of claims 15 - 21 wherein the biotinylated lipid membranes containing the enzyme to be tested are microsomes derived from liver or intestinal tissue or supersomes.

23. The method according to any one of claims 15 - 22 wherein the device has been freeze-dried.

24. The method according to any one of claims 15 - 23 wherein the method is preceded by the manufacture of the microassay device, and this comprises:4908P WO As Filed a) creating a photolithographic master cast by spin coating a first photoresist layer on an untreated wafer, preferably a silicon wafer, and then soft baking, preferably at about 65°C for about 10 min, then at about 95°C for about 40 min to yield ca. 100-pm-thick layer, b) repeating step a); c) optionally, applying a photomask, comprising the pillar array, to the product of part b); and d) exposed the product of step b) or c) to UV light.

25. The method according to claim 24 where said photomask is applied to said product of part b) by vacuum suction.

26. The method according to any one of claims 15 - 23 wherein the method is preceded by the manufacture of the microassay device, and this comprises: a) creating a photolithographic master cast by spin coating a layer on an untreated wafer, preferably a silicon wafer, and then soft baking, preferably at about 65°C for about 25 min, then at about 95°C for about 110 min to yield ca. 200-pm-thick layer; b) exposing said layer of part a) to a separate edge bead removal (EBR) step by spraying acetone to the edge of the silicon wafer; and c) exposing the product of step b) to UV light for about 30s.

27. A method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100: 110, and more preferably is about 100: 110 or 100: 125 that are functionalized with an avidin and also have bound thereto biotinylated lipid membranes, such as microsomes / supersomes, containing the enzyme to be assayed, to a sample comprising said enzyme cofactor and / or an enzyme substrate;4908P WO As Filed ii) exposing the microassay device of part i) to a therapeutic to be tested; iii) measuring the activity of said enzyme during steps i) and ii); and iv) comparing the measured activity in part ii) with that of the enzyme when not exposed to said therapeutic in part i) and, where the activity is reduced when the therapeutic is present, concluding the therapeutic has an inhibitory effect on the activity of said enzyme.

28. A method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100: 110, and more preferably is about 100: 110 or 100: 125 that are functionalized with an avidin, to biotinylated lipid membranes or biotinylated microsomes / supersomes, containing the enzyme to be assayed, to produce a loaded microassay device; ii) exposing the loaded microassay device of part i) to a sample comprising a substrate and / or cofactor for said enzyme; iii) exposing the loaded microassay device of part ii) to a therapeutic to be tested; iv) measuring the activity of said enzyme during steps ii) & iii); and v) comparing the measured activity in parts iv) of the enzyme when not exposed to said therapeutic and when exposed to said therapeutic and, where the activity is reduced when exposed to said therapeutic, concluding the therapeutic has an inhibitory effect on the activity of said enzyme.

29. A method for measuring the activity of a metabolic enzyme comprising: i) exposing a microassay device comprising a plurality of pillars having alkene / alkyne-rich surfaces, made from a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the394908P WO As Filed ratio of thiol to alkene / alkyne functional groups in said copolymer is from about 100:150 to 100:105, preferably about 100:140 to 100: 110, and more preferably is about 100: 110 or 100: 125 that are functionalized with an avidin, to biotinylated lipid membranes; ii) exposing the microassay device of part i) to lipid membranes or microsomes or supersomes, containing the enzyme whose activity is to be measured, to produce a loaded microassay device; iii) exposing the loaded microassay device of part ii) to a sample comprising a substrate and / or cofactor for said enzyme; iv) exposing the loaded microassay device of part iii) to a therapeutic to be tested; v) measuring the activity of said enzyme during steps iii) and iv); and vi) comparing the measured activity in part v) of the enzyme when exposed to said therapeutic and when not exposed to said therapeutic and, where the activity is reduced when exposed to said therapeutic, concluding the therapeutic has an inhibitory effect on the activity of said enzyme.

30. The method according to any one of claims 27 - 29 wherein said pillars are diamond-shaped in cross-section.31 . The method according to any one of claims 27 - 30 wherein said device has been freeze-dried.

32. The method according to anyone of claims 27 - 31 wherein the enzyme substrate and / or cofactor is omitted in part i) of claim 27 or part ii) of claim 28 or part iii) of claim 29 so that the enzyme is only exposed to the therapeutic to be tested, to obtain a measure of nonspecific binding of said therapeutic to the lipid membranes in order to determine ‘unbound fraction’ (fu).

33. A kit of parts comprising: i) a microassay device comprising a plurality of pillars made from or coated with a copolymer comprising or consisting of, as copolymerized units, at least one polythiol monomer and at least one polyalkene / alkyne monomer, wherein the ratio of thiol to404908P WO As Filed alkene / alkyne functional groups in said copolymer is from about 100: 150 to 100: 105, preferably about 100: 140 to 100: 110, and more preferably is about 100: 110 or 100: 125 and are functionalized by the attachment of an avidin thereto; and ii) biotinylated lipid membranes; and / or iii) microsomes / supersomes containing the enzyme to be assayed; and / or; iv) biotinylated microsomes / supersomes containing the enzyme to be assayed.

Images