RESEARCH PAPER
Non-denaturing low-temperature bonding of patterned poly(methyl methacrylate)enzymatic microreactors
Michelle D.Douma •Laurie Brown •
Terry Koerner •J.Hugh Horton •Richard D.Oleschuk
Received:9May 2012/Accepted:15July 2012/Published online:1September 2012ÓSpringer-Verlag 2012
Abstract A low-temperature solvent bonding system using methanol and water has been developed to bond poly(methyl methacrylate)(PMMA)microchips at 35°C.The substrate/cover plate adhesion strengths obtained with this bonding protocol peaked at 4,000kN/m 2for unmodified PMMA substrates.Nanoindentation measurements per-formed using atomic force microscopy revealed that only the first 30nm of the PMMA surface showed a decreased hardness following surface modification and solvent treat-ment of the PMMA surface allowing the channel architec-ture to be maintained.The low temperature utilized for bonding enabled both a temperature-robust and temperature-labile enzym
e to be facilely patterned prior to bonding with little-to-no loss in enzyme activity.Furthermore,the bond-ing methodology could be customized and used to fabricate an enzyme microreactor with pepsin (a pH,temperature and solvent sensitive enzyme).The enzyme microreactor per-formance was characterized by the longevity of the mic-roreactor,as well as the efficiency of the protein digest performed.Enzyme immobilized with WSC decreased over a period of days,whereas the enzyme immobilized with both WSC (water soluble carbodiimide and NHS (N -hydroxy-succinimide)remained active even after a month of use.Short 10mm column lengths with limited residence time provided high protein sequence coverage.
Keywords Microreactor ÁProteolytic digestion Á
Low temperature bonding ÁPolymemthylmethacrylate ÁSolvent bonding
1Introduction
The development of microfluidics in the 1990s has led to the creation of micro total analysis systems (l -TAS)(Harrison et al.1992,1993;Vilkner et al.2004).The first devices fabricated were glass chips,on which electropho-retic separations were performed (Effenhauser et al.1993;Seiler et al.1993).Since then,the field has grown rapidly and many common laboratory practices have been inte-grated on a mi
crofluidic platform.Microfluidics offers many advantages over conventional bench-top chemistry.The devices have high-throughput capabilities,require only minute amounts of sample,and analysis times are decreased (Dolnik et al.2000).
Initially,microfluidic devices were fabricated in glass or silicon because standard photolithography and chemical etching protocols had already been optimized for these materials in the microelectronics industry (Fan and Harrison 1994).Furthermore,the surface chemistry of glass and associated surface modifications has also been well studied,allowing facile modification for specific applications.Glass devices are,however,relatively expensive to fabricate because each device requires a photolithographic step,so other substrate materials have been explored,with polymers receiving the most attention.Polymer microstructures can be fabricated inexpensively without the use of a clean room using high-throughput techniques with methods such as injection molding (Heckele and Schomburg 2004),laser ablation (Roberts et al.1997),X-ray photolithography
Electronic supplementary material The online version of this article (doi:10.1007/s10404-012-1031-6)contains supplementary material,which is available to authorized users.
M.D.Douma ÁL.Brown ÁT.Koerner ÁJ.Hugh Horton ÁR.D.Oleschuk (&)
Department of Chemistry,Queen’s University,Kingston,ON K7L 3N6,Canada e-mail:oleschuk@chem.queensu.ca;Richard.Oleschuk@chem.queensu.ca
Microfluid Nanofluid (2013)14:133–143DOI 10.1007/s10404-012-1031-6
(Meng et al.2001),and hot embossing(Kricka et al.2002). After the microstructures have been fabricated,a cover plate must be bonded to the substrate to complete the capillary channel structure(Becker and Gartner2000).Several methods of bonding rigid polymers have been explored, including thermal bonding(Chen et al.2005b;Muck and Svatos2004),lamination(Metz et al.2001),adhesives (Schlautmann et al.2003),solvents(Hiratsuka et al.2004) and surface modifications(Behnisch et al.2001).Bonding of polymers is difficult because the changes to the surface must be closely controlled.Ideally,the bond must be able to withstand the requirements of the chemistry and attached instrumentation without delamination,the channel must not be clogged during bonding and the channel integrity,shape and structure,and surface chemistry must not be altered (Becker and Gartner2000).To date,it has been difficult to find a bonding protocol that fulfills each of these requirements.
Recently,there have been several groups that have published PMMA bonding methods that use tempe
ratures lower than the glass transition temperature of PMMA. These include UV/ozone(Tsao et al.2007;Devoe and Tsao 2009)(25,50,90and110°C but prior to bonding polymer pieces were dried in a vacuum oven at85°C for2h), ultrasonicfield(Li et al.2009)(immersed in solvent at50 or60°C),carbon dioxide,(Lu et al.2007)low-molecular-weight PMMAfilm(Li et al.2005)(95°C for20min), or solvent bonding with such solvents as cyclopenta-none(Chen et al.2005a)(75–85°C at1–2N/cm2for 10–15min),acetic acid(Chen et al.2005a)(75–85°C at 1–2N/cm2for10–15min),isopropyl alcohol(Cha et al. 2006)(60°C at1.3bar for10min)or a mixture of1,2-dichloroethane and ethanol(Lin et al.2007).Although all these techniques lower the temperature required for PMMA bonding,the temperature for all cases except that of1,2-dichoroethane and ethanol or carbon dioxide-assisted bonding can still be too high for non-denaturing protein conditions.Work by Lin et al.(2007)using1,2-dichloro-ethane and ethanol allows bonding of PMMA at room temperature;however,these conditions have not been applied to microfluidic devices with enzyme immobilized on the surface.Another method using carbon dioxide-assisted bonding was performed at37°C and only showed a10%reduction in bovine serum albumin content unlike thermal bonding which showed a50%reduction(Lu et al. 2007).ELISA was also used to determine the activity of devices containing immobilized lysozyme and prepared with carbon dioxide-assisted bonding.Enzyme activity was maintained by more than95%using their method in comparison with less than90%using thermal bonding.
In this work,PMMA enzyme microreactors have been fabricated with both pepsin and trypsin immobilized onto the channels of a simple microchip.Trypsin is a widely used enzyme that is very robust to analysis conditions and cleaves specifically at the C-terminus of lysine and arginine residues except if followed by proline and thus provides known peptide sequences.Polymer-based microfluidic reactors that employ trypsin have found utility in a number of proteomic applications(Liuni et al.2010;Landreh et al. 2011;Rob et al.2012).On the other hand,pepsin is a less robust enzyme and its non-specific cleavage of proteins results in less predictable peptide sequences.Despite the potential of non-specific cleavage sites,pepsin is routinely used in hydrogen–deuterium(H–D)exchange studies.A non-denaturing low-temperature solvent bonding protocol was developed(below37°C)to ensure that the enzyme is not denatured during bonding and remains active in the final device.By combining enzyme immobilization tech-niques with the new solvent assisted bonding protocol, enzymes can be immobilized on the chip in selected regions prior to polymer microchip fabrication to create patterned enzyme micro reactors.Patterning prior to bond-ing eliminates the difficulties associated with patterning surface functionality following cover plate substrate adhesion.The solvent-assisted bonding system has been characterized with tensile strengths and provided bond strengths that exceed those obtained with conventional thermal bonding.Nanoindentation measurements are used to probe the depth of material softening.Minimal soft-ening ensures that channel architecture is preserved fol-lowin
g bonding.Enzyme activity following bonding is characterized by digestion efficiencies of a model protein (cytochrome c).The necessity of low-temperature solvent bonding conditions is demonstrated by fabricating an enzyme micro reactor with a thermally robust enzyme, trypsin,and a thermally labile enzyme,pepsin using standard thermal bonding conditions,and low-temperature solvent bonding conditions.Pepsin is permanently dena-tured at high pH and temporarily deactivated in presence of sufficient amounts of organic solvents.Thermal bonding provides a functional trypsin microreactor; however,a similarly prepared pepsin micro reactor was non functional.
The low-temperature solvent bonding conditions enable a functional pepsin micro reactor to be fabricated and should provide a more universal bonding scheme as the solvent conditions can be easily tailored for the use with pH-sensitive enzymes such as pepsin.
2Experimental
2.1Materials and chemicals
Poly(methyl methacrylate)(PMMA)polymer sheets (150915091.5mm and150915092mm)were
purchased from Warehoused Plastic Sales Inc.(Toronto, ON,Canada)and cut to size before use.Fused
silica cap-illary with a polyimide coating and an outer diameter of 150l m and an inner diameter of75l m was purchased from PolyMicro Technologies(Phoenix,AZ,USA) N-hydroxysuccinimide(NHS),and trifluoroacetic acid (TFA)were purchased from Aldrich Chemicals(Milwaukee, WI,USA).N-(3-dimethylaminopropyl)-N’-ethylcarbodi-imide(WSC)and a-cyano-4-hydroxycinnamic acid(HCCA) were acquired from Fluka(Buchs,Switzerland).Trypsin (bovine13,000units/mg),pepsin(porcine3,600units/mg) and cytochrome c(horse heart)were obtained from Sigma (Oakville,ON,Canada).TRISÁHCl,2-propanol(ACS reagent grade)and methanol(ACS reagent grade,Ultrapure grade)were purchased from Fisher Scientific(Fair Lawn,NJ, USA).Sulfuric acid(ACS reagent)and acetic acid were purchased from Fisher Scientific(Nepean,ON,Canada). Ammonium bicarbonate,2,5-dihydroxybenzoic acid (DHB),sodium chloride(ACS reagent),sodium phosphate monobasic,magnesium dichloride hexahydrate and calcium chloride dihydrate(ACS reagent)were acquired from Sigma (Oakville,ON,Canada).Deionized water(18.2M X)was purified using a MilliQ apparatus(Millipore,Bedford,MA, USA).
2.2Enzymatic patterning on PMMA substrates
Poly(methyl methacrylate)substrates(1.5mm thick)were embossed using an electroplated nickel master with dimensions1009100mm(Tecan,Dorset,UK)with a single straight channel design measuring
85mm long, 100l m wide and20l m deep.Briefly,the PMMA and nickel master were brought into conformal contact,aligned and placed in a HEX-01hot embosser(Jenoptik Micro-technik,Jena,Germany).The embossing chamber was closed,evacuated and heated to115°C and an embossing force of10,000N applied for10min.Following stamping, the temperature of the tool and substrate was decreased to 85°C and then the chamber was brought to atmospheric pressure and opened.The nickel master and PMMA sub-strate were then carefully demolded by hand.The PMMA substrate and cover plates were cut to size,and2mm holes were drilled in the cover plates in predefined locations to act as access reservoirs to the channels.As an alternative, larger channel diameters of150l m could be fabricated by embossing a fused silica capillary with an outer diameter of150l m into the PMMA substrate using a temperature of115°C with a force of2,000N for600s.The capil-lary could then be removed to leave a150-l m channel diameter in the PMMA.In this case the cover plate did not have the requirement of having access holes because fused silica capillary pieces with an outer diameter of 150l m and an inner diameter of75l m could be placed at the entrance and exit of the channel to define a specific channel length while allowing the microfluidic chip to be connected to a syringe pump.The embossed PMMA substrates were immersed in3M H2SO4at60°C for 20min.Acid treatment was used to hydrolyze ester linkages to carboxylic acids on the surface of the poly-mer.The modified PMMA was then rinsed with copious amounts of water,followed by rinsing with2-
propanol and drying with a stream of nitrogen.PDMS pieces were cut to create solution reservoirs2mm wide and10mm or80mm in length.These were positioned over the channel region and pressed against the PMMA substrate plates by hand until a watertight seal was formed (Fig.1).Two different methods of enzyme immobiliza-tion were used.reactor debug mode is enabled
2.3N-(3-dimethylaminopropyl)-N0-ethylcarbodiimide
(WSC)immobilization method
A solution(approximately40l L for the10mm reactor and320l L for the80-mm reactor)containing1.0mg/mL of aqueous N-(3-dimethylaminopropyl)-N0-ethylcarbodi-imide(WSC)was added to the PDMS reservoirs and allowed to react for2h at room temperature.The PMMA was then rinsed with water and2-propanol and dried with nitrogen.The PDMS pieces were replaced,to limit the amount of protein solution required,in the same positions on the surface of the PMMA,and a10-mg/mL trypsin solution in50mM NH4HCO3(pH7.4),10mM NaCl and 10mM MgCl2was added to the solution reservoirs.The reservoirs were covered with parafilm and stored at4°C for24h.After that time,the PDMS sample reservoirs were removed and the PMMA was rinsed gently with copious amounts of water to remove any adsorbed enzyme.
PDMS Reservoir/Mask
A
B
2.4WSC/N-hydroxysuccinimide(NHS)
immobilization method
In the second method,a solution of1.0mg/mL aqueous WSC and0.1mg/mL N-hydroxysuccinimide(NHS)was added to the PDMS solution reservoirs and reacted at room temperature for2h.The PMMA substrates were then rinsed with water and2-propanol and dried with nitrogen. The PDMS solution reservoirs were replaced and a10mg/ mL trypsin solution in50mM NH4HCO3(pH7.4),10mM NaCl and10mM MgCl2or a5-mg/mL pepsin solution in 1%acetic acid was used tofill the reservoirs.The reser-voirs were sealed with parafilm and stored at4°C for24h followed by gentle rinsing with water or1%acetic acid to remove the pepsin.An acidified rinse solution was used due to the pH sensitivity of Pepsin.
2.5Low-temperature bonding with immobilized
enzyme
Poly(methyl methacrylate)cover plates were sonicated in a 1:12-propanol and water solution for10min and dried with nitrogen.For the tryptic microreactors,eight drops (approximately250l L)of a97%methanol(ACS grade) and3%water solution were evenly spread over the stamped PMMA substrate.Bonding solution was applied to the substrate and the cover plate and substrates brought into conformal contact.The two polymer layers were inspected to ensure the solution had evenly wet the interface.The cover plate was aligned and placed in conformal contact with the substrate and placed within the embossing chamber.Alternatively,when peptic microreactors were prepared the conditions were modified to address the pH sensitivity of the pepsin becomes irreversibly inactivated above a pH of5).Solvent bonding conditions consisted of:97%methanol and3%of a1M H2SO4 solution.The chamber was closed,a touch force of250N was applied for30s and the tool and substrate were brought to35°C.A force of3,000N was then applied for 1min.For the100-l m-wide920-l m-deep channels,a one-piecefinger-tightfitting(Upchurch Scientific,Oak Harbour,WA,USA)wasfixed to one of the access holes of the PMMA microchip with5min epoxy(Lepage,Taylor, MI,USA).This step was unnecessary when the larger channel diameters of150l m were used.Capillary lengths (*20cm)were positioned at the entrance and exit of the channel to define the channel length,followed b
y bonding of the cover plate.For tryptic microreactors,the channels were rinsed with50mM TRISÁHCl(pH7.4)after bonding and stored at4°C in a50mM TRISÁHCl(pH7.4)and 10mM CaCl2solution.To prevent drying of the enzyme during storage,the microchip access holes were sealed with parafilm.Prior to use,the PMMA microfluidic channels were rinsed with50mM TRISÁHCl(pH7.4),followed by 50mM NH4HCO3(pH7.4)at aflow rate of5.0l L/min for 10min using a syringe pump(Harvard Apparatus,St. Laurent,PQ,Canada).With peptic microreactors,the channel was rinsed with1%acetic acid solution and also stored at4°C using the same solution.
2.6Thermal bonding of PMMA substrates
The temperature stability of immobilized pepsin and trypsin was tested using thermal bonding enzyme immobilized substrates.Conditions consisted of105°C with500N of applied force for300s.The activity of the enzyme was ascertained by sequence coverage(see below)both before and after thermal bonding conditions were applied.
2.7Measurement of cover plate adhesion strength
The tensile strength of the bond between the sulfuric acid modified or unmodified PMMA substrates and unmodified cover plates was determined using an Instron3369tensile tester(Instron,Toronto,ON,C
A).Two pieces of PMMA (7092091.5mm)were bonded with the aid of a solvent solution consisting of97%methanol and3%water.These PMMA substrates were bonded with known cross-sectional areas ranging between1.0and3.0cm2.The samples were clamped to the tensile tester with grips pulled away from each other at a rate of1.3mm/min and set114mm apart initially(ASTM2004).The force at which the bonded PMMA failed was measured.These values were divided by the cross-sectional area to obtain the tensile strength of adhesion.Each value reported is the average of at leastfive samples per cross-sectional area,and three cross-sectional areas were tested for each method of bonding.
2.8Nanoindentation measurements on the surface
of PMMA
Poly(methyl methacrylate)samples were cut to109 1092mm and left unmodified or treated with3M sul-furic acid at60°C for20min.Nanoindentation measure-ments were performed by immersing the PMMA substrates in2mL of the solvent mixture(3%water,97%metha-nol)at35°C for1min.Each sample was analyzed within 20min of preparation.Nanoindentation experiments were performed using either a Hysitron Triboscope TS1(Hysi-tron Inc.,Minneapolis,MN,USA)and a scanning probe microscope(SPM)(Nanoscope E,Digital Instruments, Santa Barbara,CA,USA)equipped with a1-l m conical diamond tip p
robe or a Hysitron nanoidenter with a1-l m 60°conical tip with an AFM integrated within the indenter to image the indents with a higher resolution than the indenter tip.
The SPM was used to image the surface topography of the polymer prior to,and after indentation to determine the integrity of both the polymer surface and the tip indenta-tion.A typical nanoindentation data curve and an illustra-tion of the indentation geometry with all pertinent parameters are given in Fig.2.
The Hysitron was calibrated using a fused quartz crystal to obtain the empirically determined constants C0,C1,C2, C3relating the contact area of the indenter,A,and the contact depth,h c.The two are expressed in terms of a polynomial:
Aðh cÞ¼C0h2cþC1h cþC2h1=2cþC3h1=4cþÁÁÁ
Loads ranging from10to1,000l N were repeated at least three times to determine the coefficients for this equation.Loads ranging from10to200l N were performed on the PMMA samples.Each force was applied to the PMMA at least three times.The hardness (H)was calculated as follows:
H¼P max Aðh cÞ
using Hysitron software(Klapperich et al.2001).
2.9On chip enzymatic digests
Cytochrome c was chosen as a model protein to test the enzymatic microreactors as it is easy to digest and contains no disulfide linkages that require reduction/alkylation to unfold the protein prior to enzymatic digestion.For tryptic microreactors,protein solutions(5l M in50mM NH4HCO3solution,pH7.4)were prepared and diluted3:2 with methanol to assist in protein denaturation(ACS grade) and pumped through the PMMA enzyme microreactors at a flow rate of0.5l L/min,producing an on chip residence time of approximately20s for the smaller channel diam-eter(100l m)and3min for the larger channel diameter (150l m)when an80-mm enzymatic channel bed was fabricated.When peptic microreactors were used,5l M cytochrome c in a100-mM sodium phosphate monobasic solution was used or a mixture of5l M cytochrome c in a 9:1mixture of1%acetic acid solution and methanol for the longevity experiments.The appropriate solution was then passed through the chip at0.5l L/min with a syringe pump.All digested solutions were collected and analyzed off-line by Matrix-assisted laser desorption ionization-time-of-flight mass spectrometry(MALDI-TOF MS).
2.10MALDI-TOF MS analysis of enzyme digests
A Voyager DE STR MALDI time-of-flight mass spec-trometer(Applied Biosystems,Foster City,CA,USA)was used to analyze all the collected protein digests except for the digest from the peptic microreactor fabricated using the low-temperature bonding technique.The latter protein digest was analyzed by a QSTAR XL Quadrupole time of flight mass spectrometer instrument equipped with a MALDI source(Applied Biosystems,MDS SCIEX,Con-corde,ON,Canada).A two-layer HCCA matrix sample prep(Dai et al.1999)was employed for analysis with the Voyager MALDI instrument,while a DH
B dried
drop Fig.2A schematic diagram of a nanoindentation curve with all necessary nomenclature for the determination of hardness and reduced modulus.a Shows the indentation under maximum load.
b Shows the an image of the nanoindentation produced while testing surface hardness.
c Shows a typical nanoindentation curve detailing the values obtaine
d from th
e unloading curve(Klapperich et al.2001; VanLandingham et al.2001).h
f is defined as thefinal displacement. P max-Maximum pressure at which the unloadin
g cycle begins. S-contact stiffness from whic
h modulus is obtained=initial slope of the unloading curve
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论