Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Feb 15.
Published in final edited form as: Anal Biochem. 2011 Oct 5;421(2):526–533. doi: 10.1016/j.ab.2011.09.028

Non-volatile copolymer compositions for fabricating gel element microarrays

Julia B Golova 1,*, Boris K Chernov 1, Alexander N Perov 1, Jennifer Reynolds 1, Yvonne L Linger 1, Alexander Kukhtin 1, Darrell P Chandler 1
PMCID: PMC3270149  NIHMSID: NIHMS329745  PMID: 22033291

Abstract

By modifying polymer compositions and cross-linking reagents, we have developed a simple yet effective manufacturing strategy for copolymerized three-dimensional gel element arrays. A new gel-forming monomer (2-(hydroxyethyl) methacrylamide; HEMAA) was used that possesses low volatility and improves the stability of copolymerized gel element arrays to on-chip thermal cycling procedures relative to previously used monomers. Probe immobilization efficiency within the new polymer was 55%, equivalent to that obtained with acrylamide (AA) and methacrylamide (MA) monomers. Non-specific binding of single stranded targets was equivalent for all monomers. Increasing cross-linker chain length improved hybridization kinetics and end-point signal intensities relative to N,N-methylenebisacrylamide (Bis). The new copolymer formulation was successfully applied to a model orthopox array. Because HEMAA greatly simplifies gel element array manufacture, we expect it (in combination with new cross-linkers described herein) to find widespread application in microarray science.

Keywords: Hydrogels, microarrays, copolymerization, biochip, hybridization

INTRODUCTION

Microarrays are widely used in scientific research and are now finding practical application in drug discovery, pathogen detection, and clinical diagnostics (1-7). Typically, oligonucleotide or protein microarrays are manufactured on a two-dimensional surface such as glass, plastic, or polymeric films (8-11). One of the limitations of two-dimensional arrays is the slow reaction kinetics and relatively low level of sensitivity, due in part to comparatively low immobilized probe density and probe proximity to the support surface that can affect the efficiency and kinetics of target capture (12-14). For these reasons, three-dimensional microarrays utilizing hydrogel matrices are attractive supports for nucleic acid and protein microarrays.

Gel element arrays formed via copolymerization (15, 16) have proven effective in nucleic acid and protein applications, including on-chip enzymatic assays (17-20). Copolymerized gel elements are formed by mixing biological probes with gel-forming reagents, depositing the mixture on a substrate, and simultaneously polymerizing the mixture to form the single gel elements. The resulting three-dimensional gel elements have a number of performance-enhancing properties relative to planar arrays, such as higher probe immobilization capacity; sterically-favorable spacing of immobilized molecules throughout the volume of the gel element; and an aqueous environment surrounding attached probes. The net result of these attributes is an increase in hybridization efficiency (21-23), with an early report claiming a 100-fold increase in detection sensitivity for gel element arrays compared to planar arrays (24).

The standard composition of copolymerized gel elements includes acrylamide (AA) or methacrylamide (MA) as the gel-forming monomer and N, N-methylenebisacrylamide (Bis) as a cross-linking reagent (25-27). A 5% AA gel is suitable for detecting and analyzing intact nucleic acid fragments of 100-150 nucleotides. However, efforts to increase the diffusion rate of large DNA fragments into gel elements by changing the pore size of the AA gel network leads to a decrease of gel mechanical stability, such that the resulting biochips are quite fragile and unsuitable for high-temperature experiments such as on-chip polymerase chain reaction (PCR) (20, 25). Several synthetic and naturally derived hydrogel polymer networks based on poly(ethylene) glycol, polypyrrol, chitosan, or polysaccharide have been proposed and tested over the last several years (28-31). However, none of these compositions has proven effective in the range of microarray applications described above, nor have they made the transition to high-throughput manufacturing. Thus, there is still a need to develop gel compositions that are conducive to high-throughput manufacture yet overcome some of the pore-size and thermostable limitations of AA or MA hydrogels used for microarray fabrication. The objective of this study was therefore to develop a new class of copolymers with enhanced thermo-mechanical stability and manufacturing properties relative to those based on AA or MA.

MATERIALS AND METHODS

Acrylamide, methacrylamide, N,N-methylenebisacrylamide, methacryloyl chloride, sodium sulfate, glycerol, ethanolamine, 1,3-diamino-2-hydroxypropane, 1,3-butadiene diepoxide, 2,2’-(ethylenedioxy)bis(ethylamine), trimethoxysilylpropyl methacrylate, N-hydroxysuccinimide, Tween™ 20, methanole, triethylamine, acetonitrile, tetrahydrofuran, dichloromethane, chloroform, acetone, and ammonium hydroxide were obtained from Sigma-Aldrich Chemical Co., Inc. (Milwaukee, WI) and used without further purification. Sodium chloride, dibasic sodium phosphate, and monobasic sodium phosphate (monohydrate) were purchased from Fisher Scientific Inc. (Fair Lawn, NJ). 40% AA/Bis solution (19:1) was purchased from Bio-Rad Laboratories (Hercules, CA). Ethanol was purchased from AAPER Alcohol and Chemicals Co. (Shelbyville, KY). 0.5 M EDTA (pH 8.0) and 20X SSPE buffer solution were purchased from Ambion Inc. (Austin, TX). Texas Red® sulfonyl chloride (TR, mixed isomers) and tetramethyl rhodamine-5 (and 6) isothiocyanate (TMR, mixed isomers) were purchased from Molecular Probes (Eugene, OR). Reagents for solid phase oligonucleotide synthesis were purchased from Glen Research (Sterling, VA). Acrylic glass slides for microarray fabrication were purchased from CEL Associates (Pearland, TX). Mass spectra were recorded using a Bruker BiFlex III MADLI-TOF mass spectrometer; NMR spectra were recorded using a Bruker Advance 400 MHz NMR spectrometer (Bruker Daltonics, Billerica, MA).

N-Hydroxysuccinimide methacrylate (NHS-MA)

Methacryloyl chloride (0.1 mol) in 25 ml of tetrahydrofuran was added drop-wise with stirring and cooling on ice to a solution of N-hydroxysuccinimide (0.1 mol) and triethylamine (0.1 mol) in 50 ml of tetrahydrofuran. After 1 hour of stirring at room temperature, the precipitate was removed and the filtrate evaporated under vacuum. The residue obtained was crystallized from methanol to give NHS-MA with 75% yield and a melting point of 102-104°C.

2-(Hydroxyethyl)-methacrylamide (HEMAA)

A solution of NHS-MA (1.83 g, 10 mmol) in 25 ml of acetonitrile was added to solution of ethanolamine (0.67 g, 11 mmol) in 50 ml of acetonitrile. The reaction mixture was stirred for 20 min at room temperature, filtered and then concentrated under vacuum. The residue was purified by column chromatography on silica gel by applying a gradient of 0-5% of acetone in dichloromethane. The final product was obtained with 81% yield (1.05 g). FAB-MS; 130.1 (M+H)+. Calc. for C6H11NO2: MW 129.04. 1H NMR (400 MHz, CDCl3): δ(ppm) 6.54 (s 1H, NH); 5. 69 (s, 1H, (H3C)C=CHaHb); 5.31 (s, 1H, (H3C)C=CHaHb), 3.70 (t, J=5.1 Hz, 2H, CH2OH); 3.43(dt, J=5.3Hz, 5.1Hz, 2H, NHCH2); 3.33 (s, 1H, OH), 1.92 (s, 3H, CH3).

N,N’-Dimethacryloyl-1,3-diaminopropan-2-ol (DMDAP)

A mixture of 1,3-diamino-2-hydroxypropane (0.9 g, 10 mmol), NHS-MA (3.84g, 21 mmol) and triethylamine (2.02 g, 20 mmol) was dissolved in 50 ml of acetonitrile and stirred at room temperature for 1.5 hours. The solvent was removed under vacuum, and the crude material was partitioned between water and ether (1:2; 3 × 20 ml). Organic phases were combined, concentrated under vacuum, and purified by silica gel column chromatography by applying a gradient of 0-10% acetone in chloroform. The final product was obtained with 82% yield (1.87g). EAB-MS: 228.1 (M+2H)+. Calc. for C11H18N2O3: MW 226.08.

N,N-Dimethacryloyl -1,4-diaminobutane-2,3-diol (DMDAB)

1,3- butadiene diepoxide (4.3 g, 0.05 mol) was placed in a two-necked flask equipped with a condenser, dropping funnel and magnetic stirrer. Ammonium hydroxide (15 ml of 32% water solution) was added and the contents stirred at room temperature. After 3 hours, the reaction mixture was evaporated to dryness, and then the oil residue was co-evaporated with ethanol. The product obtained was reconstituted in a solution of ethyl alcohol (150 ml), triethylamine (10.1 g, 0.1 mol), and methacrylic anhydride (16.0 g, 0.103 mol) and stirred at room temperature for 5 hours. The solution was then concentrated to dryness under reduced pressure. Chloroform (200 ml) was added to the residue and the resulting mixture was washed with water (50 ml), dried under sodium sulfate, concentrated under vacuum and purified by silica gel column chromatography by applying a gradient of 0-10% methanol in chloroform to give DMDAB (as an oil) with a yield of 15% (3.9 g). FAB MS: 252.1 (M+H)+. Calc. for C12H20N2O4: MW 256.08.

Synthesis of 3’-metacrylamido-modified oligonucleotides

Oligonucleotide synthesis was carried out on an Applied Biosystems 394 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA) at 0.5 μmol scale according to the manufacturer's recommended protocol, and using commercial β-cyanoethyl phosphoramidites. Methacrylamido-modifier-CPG was used as a support for solid phase oligonucleotide synthesis to prepare probes containing a methacrylic function at their 3’-ends. Cleavage from the solid support and deprotection of 3’-methacrylated oligonucleotides was accomplished with concentrated aqueous ammonium hydroxide for 8-12 hours at 55°C. 5’-DMTr-protected 3’-methacrylated oligonucleotides were purified by reverse-phase (RP) HPLC. The 5’-DMTr-protective group was removed with 80% acetic acid (10 min) and 3’-methacrylated oligonucleotides were isolated by RP HPLC. HPLC-purified oligonucleotides were evaporated to dryness (CentiVap concentrator, Labconco, Kansas City, MO), reconstituted in 500 μl of Milli-Q water, and quantified by UV adsorption (UV/VIS Spectrophotometer Lambda Bio 10, Perkin Elmer, Boston, MA). Thereafter, oligonucleotides were normalized in Milli-Q water to a final concentration of 2 mM and stored at -4°C until use. 5’-Cy-3-labeled and 3’-methacrylated oligonucleotides were synthesized by standard solid phase phosphoramidite chemistry according to the manufacturer's recommendations using methacrylamido-modifier-CPG and commercial Cy3 phosphoramidite (Glen Research, Sterling, VA), and purified by reverse-phase HPLC after the deprotection procedure.

Preparation of copolymerization mixtures

Gel-forming monomers and cross-linking reagents utilized in this study are listed in Table 1. Standardized pre-polymer mixtures contained gel-forming monomer (0.47 M); cross-linker reagent (0.0235 M); glycerol (65% v/v), and 0.2 M sodium-phosphate buffer (pH 7.2), and are modeled after recipes originally developed for 4% MA monomer. Aliquots of normalized oligonucleotide solutions were dried in vacuum and then re-dissolved in pre-polymer mixture to achieve a final concentration of 0.25 mM. The pre-polymer solutions containing oligonucleotides were then placed in a 384-well microtiter plate (20 μl per well; Genetix, New Milton, UK) for printing.

Table 1.

Oligonucleotide probes and targets.

Probes Sequence (5’ to 3’)
P1 ACACTGTTGTAGCAAATAGG-Met
P2 ACGACTGATTTGCATTCCGG-Met
Q GATGATGATGATGATGATGA-Met
C Cy3-TTGTGGTGGTGGTGTGGTGG-Met
C1 Cy3-TTTTTTTTTT-Met
NS NNNNNNNNNNNNNNNNNNNN-Met
Targets
T1 TR-T20-CCTATTTGCTACAACAGTGT-T14
T2 TMR-TTATACCACTCTGCAACGTGTCGCAGCGCTGGAACGTTCCGGAATGCAAATCAGTCGTCACTCACTGGTTTCATCATATCTGGCGTTAATGCGTT
QC Cy3 -TCATCATCATCATCATCATC
Open in a new tab

Met = -OCH2CH(CH2OH) [NH(CH2)5HNOCCH(CH3)=CH2]

TR = Texas red

TMR = Tetramethyl rhodamine

NS = nonsense oligonucleotide containing a mixture of all possible 20-mers

Microarray fabrication

Microarrays were printed on CEL Associates acrylic glass slides (Pearland, TX) essentially as described previously (20). Copolymerization mixtures were printed with a QArray2 arrayer (Genetix, New Milton, UK) and 150 μm blunt pins. For AA copolymer solutions, printed arrays were incubated for 2 hours in an airtight container with 2 to 4 ml of a mixture that included all of the components of the pre-polymer solution except the oligonucleotides (in what follows, this step is referred to as “restoration” - see Results). After restoration, the slides were placed in an airtight cassette equipped with quartz windows and polymerized for 30 min in a nitrogen atmosphere under a Thermo Spectronic Model XX-15A UV lamp with 312 nm tubes. Microarrays printed with HEMAA-based compositions were polymerized without the restoration procedure. Finally, the slides were washed in 1X SSPE - 0.05% SDS buffer for 40 min at 40°C, thoroughly rinsed with MilliQ water, and air dried.

Probe immobilization efficiency

Probe immobilization efficiency was quantified by printing arrays with 5’-Cy3-(T)10- Met-3’. Fluorescent measurements were first taken from polymerized arrays and before the slide washing procedure using a custom-built microscope described (32) and Microchip Imager software (Aurora Photonics Inc., Lake Barrington, IL). The calibration control for the imager was a Flour-Ref 2273-G slide (Ted Pella Inc, Redding, CA). For each gel drop, the fluorescent signal intensity was calculated by integrating the pixel intensities corrected for the average local background over the entire area of the gel-element image. After capturing an initial array image, the arrays were then washed with 1X SSPE-0.05% SDS at 45°C for 45 min, rinsed with Milli-Q water, dried with compressed air, and re-imaged. The washing procedure was repeated until the signal from dried gel elements remained unchanged. Probe immobilization efficiency was then calculated as the average percentage of residual fluorescence after washing procedure over all replicate arrays (n = 15 replicates of each probe per array; one array per slide; four slides total). We note here that because of the low intensity the incident laser light, we do not observe any evidence for photobleaching upon repeated measurements of the same microarray.

Thermal stability

The thermal stability of new copolymers was tested on replicate arrays containing 100 gel elements of each copolymer formulation, under 4 different (commercially available) buffer/enzyme combinations and varied temperature regimes. The most vigorous thermal cycling regime included an initial hot start at 80°C for 2 min and 93°C for 3 min, followed by 100 cycles of [92°C, 45 sec; 59°C, 90 sec; 72°C, 60 sec] and a 3 min, 72°C final extension. Durability was defined as the ability of gel drops to remain affixed to the substrate after thermal cycling and subsequent rinses in wash buffer and water, and was deduced by visual examination of the arrays under a microscope before and after the thermal cycling procedure.

Synthetic DNA targets

Fluorescently labeled 50- and 100-mer targets were synthesized by standard solid phase phosphoramidite chemistry on an Applied Biosystems 394 DNA/RNA Synthesizer (Applied Biosystems Inc., Foster City, CA). For incorporating Texas Red (TR) and tetramethyl rhodamine (TMR) fluorescent labels, 5’-amino-oligonucleotides were synthesized with 5’-amino modifier from Glen Research (Glen Research, Sterling, VA). After the standard deprotection procedure, 5’-amino-oligonucleotides were purified by reverse-phase HPLC. To incorporate Texas Red (TR), 5 OU of purified 5’-amino-oligonucleotide was dissolved in 50 μl of 0.1 M NaHCO3, and Texas Red ® sulfonyl chloride (0.01 mg μl-1 in acetonitrile) was added in two portions of 50 μl each. The labeling reaction was carried out at 4°C for 2 hours. To incorporate tetramethyl rhodamine (TMR) fluorescent label, 5 OU of purified 5’- aminated oligonucleotide was dissolved in 50 μl of acetonitrile-triethylamine-Milli-Q water (3:0.5:1), and tetramethyl rhodamine-5(6)-isothiocyanate (0.01 mg μl-1 in acetonitrile-triethylamine-MQ water (3:0.5:1)) was added in two portions of 50 μl each. The TMR labeling reaction was carried out at room temperature for one hour. Texas Red and TMR-labeled products were precipitated by adding 1.5 ml 1% LiClO4 in acetone, followed by centrifugation (13,000 RPM, 5 min, centrifuge Galaxy 16 DH, Eppendorf, Westbury, NY). The resulting precipitates were dissolved in 500 μl of Milli-Q water and purified by RP HPLC to give 3.5 - 4.0 OU of labeled oligonucleotides. 5’-Cy-3 labeled oligonucleotide was synthesized by standard solid phase phosphoramidite chemistry using commercial Cy3™ phosphoramidite (Glen Research, Sterling, VA), and purified by reverse- phase HPLC after the deprotection procedure.

PCR amplicons

DNA amplicons were prepared by asymmetric PCR using a set of TaqMan primer sequences and DNA templates kindly provided by Dr. M. Sofi Ibrahim (United States Army Medical Research Institute of Infectious Disease (USAMRIID)). The asymmetric PCR reaction (25 μl final volume) contained 1x PCR buffer (Invitrogen Platinum Kit, Invitrogen Corporation, Carlsbad, CA), 3 mM MgCl2, 7 μM of each dNTP, 7 μM of Cy3-dUTP (Amersham Biosience, GE Healthcare, Piscataway, NJ), 0.04 μM of forward primer, 0.2 μM of reverse primer, 2 pg of template DNA, and 2U of Platinum® Taq DNA Polymerase (Invitrogen Corporation, Carlsbad, CA). Amplification was carried out in DNA Engine Dyad thermal cycler (Bio-Rad Laboratories, Inc., Hercules, CA) and included an initial denaturation at 95°C for 2 min, 44 cycles of [15 sec at 94°C, 1 min at 60°C], and final extension at 72°C for 5 min. Amplified PCR fragments were separated from unincorporated Cy3-dUTP using Centri-Sep columns as per the manufacturer's instructions (Princeton Separation, Inc., Adelphia, NJ).

Hybridization experiments

Kinetic hybridization experiments with synthetic target oligonucleotides (T1 and T2, Table 1) were carried out in frame-seal incubation chambers (Bio-Rad, Hercules, CA) under glass cover slips. Hybridization buffer contained 0.01 M sodium-phosphate (pH 7.2), 1 M NaCl, 1 mM EDTA, 0.1% (w/v) Tween-20 and 10 fmol μl-1 of the synthetic 50-mer target or 50 fmol μl-1 of the 100-mer target. Hybridization was carried out at room temperature for up to 24 hours; and repetitive detection of fluorescent signals was performed with fully assembled hybridization chambers using the custom-built microscope described above. Average, background-corrected hybridization signals were calculated as described above (n = 15 replicate elements per target).

Microarray hybridizations with asymmetric DNA amplicons (150-180 nucleotides) were carried out in 25 μl reaction chambers (Grace Bio-Labs, Inc., Bend, OR), except the hybridization buffer was 1 M guanidine thiocyanate, 0.05 M HEPES, 0.05 M EDTA, 3.3 mg ml-1 BSA and 10 μl of purified PCR amplicon. Hybridizations were carried out on a MJ Research PTC-200 DNA Engine Dyad thermal cycler using a thermal profile of 3 min at 95°C followed by 3.5 hours at 55°C. After removing hybridization chambers, arrays were washed in 1X SSPE-0.1% Triton X-100 buffer for 10 min at room temperature, followed by two quick rinses in Milli-Q water. Microarrays were then dried with compressed air and imaged as described above. Total hybridization signal was extracted from each image using a fixed circle algorithm and signal intensity averaged over all replicate probes. The average signal to noise ratio (SNR) was then calculated as the mean of the average fluorescence signals divided by the mean of the average background signal from gel elements containing the nonsense probe.

RESULTS

Monomer volatility

Our initial indications that AA and MA based copolymers can be problematic for manufacturing scale up were observable deviations in drop morphology, poor adherence to the substrate during thermal cycling and large deviations in end-point signal intensities. These observations and signs of gel element deterioration were more pronounced with longer time lapses between printing and UV-induced photopolymerization. For AA gel elements (and no restoration procedure), deterioration became obvious if the polymerization was delayed by approximately 20 min. For the MA arrays, the characteristic time for gel element degradation was even shorter, in that only traces of gel could be observed on the substrates with a 20-min delay between printing and polymerization (not shown). We hypothesized that such inconsistency in polymerization results (and resulting gel element hybridization behavior) might be caused by the evaporation of AA and MA monomers during and after the printing procedure. If true, monomer evaporation would alter the fundamental properties of the gel elements in uncontrollable ways during array manufacture.

To test this hypothesis, gel element arrays were fabricated from 4% MA-Bis and 4.75% AA-Bis compositions at 20:1 and 19:1 molar ratios, respectively, using oligonucleotide P1 (n = 8 replicates per array, three arrays per slide; 24 total replicates per slide). Microarrays were subjected to UV photopolymerization at 2, 10, and 20 min intervals after printing, but without applying the restoration procedure described in the Methods. Arrays were then hybridized with 50-mer target T1 (at 10 fmol μl-1) for 12 hours and images captured every 15 minutes. As shown in Figure 1, there is a noticeable decrease in end-point signal intensity (approximately 20% reduction) for both AA and MA copolymers when subjected to a 20 min and 10 min delay between printing and photopolymerization, respectively, and observation that is consistently seen across multiple print runs (not shown). Importantly, the MA copolymer was essentially nonfunctional if the time delay before photopolymerization increased to 20 minutes, because the gel elements were too thin to generate a detectable signal. For these reasons, AA and MA gel element arrays require a post-polymerization restoration, which may or may not be a scalable manufacturing process.

Figure 1.

Figure 1

Open in a new tab

Hybridization of target T1 on (A) acrylamide, (B) methacrylamide and (C) custom-synthesized 2-(hydroxyethyl) methacrylamide (HEMAA) gel element arrays polymerized without delay, 10 min and 20 min after printing. No restoration step was applied prior to photo-polymerization. Image acquisition time = 100 msec per frame.

In order to overcome this fundamental manufacturing and performance challenge, we therefore synthesized the non-volatile monomer 2-(hydroxyethyl) methacrylamide (HEMAA). In this case, the 20 min delay between printing and photopolymerization resulted in little to no difference in hybridization kinetics and end-point product accumulation (5% reduction in signal intensity). The lower signal intensity in HEMAA relative to AA and MA in Figure 1 may be related to hydrogel network structure that might affect target accessibility. For example, it is known that 2-(hydroxyethyl) methacrylate (HEMA), a monomer with a very similar structure to HEMAA, results in highly ordered structures that exhibit higher mechanical strength than either AA or MA (33-35). It has also been reported that including a hydroxyl-containing group into the copolymerization mixture accelerates the monomer conversion and polymerization rates (36). Under identical polymerization conditions, then, we expect a higher rate of HEMAA incorporation into the hydrogel and the formation of a tighter network structure than what is otherwise obtained with AA or MA. This hypothesis is supported by data in Figure 2, where HEMAA-Bis gel element arrays (20:1) were manufactured at 0.47 M, 0.37 M, and 0.27 M composition and subject to the same time-course hybridization experiment shown in Figure 1. In this case, lowering the HEMAA monomer concentration significantly improved hybridization kinetics and end-point product accumulation relative to the 0.47 M HEMAA copolymer (approximately 100% gain in signal intensity), although end-point signal intensity still fell short of the AA and MA monomers. For these reasons, we developed a number of additional cross-linkers to modify the pore structure and hybridization kinetics of HEMAA (below). Nevertheless, the fact that the HEMAA-based arrays provided strong hybridization signals without the restoration shows that this monomer is less volatile than AA or MA and can therefore be manufactured with greater control over polymerization behavior. HEMAA therefore has certain, practical advantages for gel element array production scale-up.

Figure 2.

Figure 2

Open in a new tab

Effect of HEMAA monomer concentration on target T1 hybridization kinetics for 2 min, 10 min and 20 min time lags between printing and UV-induced photo-polymerization. No restoration step was applied prior to photo-polymerization. Image acquisition time = 100 msec per frame.

Probe immobilization efficiency

Copolymerization mixtures of AA, MA, and HEMAA-Bis (0.47 M monomer with monomer-crosslinker molar ratio = 20:1) were fabricated with oligonucleotide C1 (n = 24 gel elements per target for each copolymerization composition; AA and MA arrays included the restoration step as described in the Methods) and drop diameters measured with a tool-makers microscope. Average droplet diameters were 120, 150, and 135 microns, respectively, with a 5% relative standard deviation for AA and MA and a 3% relative standard deviation for HEMAA. These data indicate that the HEMAA polymer, without any restoration procedure, provides gel element morphological reproducibility equivalent to that of AA and MA polymers (where a restoration step is included in the manufacturing procedure).

It is already known that oligonucleotide probe immobilization efficiency in 4-5% AA and MA gel elements is approximately 50% (20, 26). To evaluate the efficiency of probe immobilization and the effect of the time delay between printing and photopolymerization events in 0.47 M HEMAA-Bis (20:1) gel elements, three replicate arrays (containing n = 8 replicates of probe C1 per array) were manufactured on the same slide with a time delay of 60 and 90 min between the first, second, and third array, respectively. Arrays were processed without any restoration procedure and photopolymerized immediately after all three replicate arrays were printed (resulting in an approximately 90 min maximum time delay for the first printed array per slide). Regardless of the time delay between printing and UV polymerization, the average immobilization efficiency was 54-55% (Figure 3), consistent with the efficiency observed previously for AA and MA arrays polymerized using the same protocol after the restoration. It is therefore possible to achieve immobilized probe concentrations on par with the AA or MA reference polymers.

Figure 3.

Figure 3

Open in a new tab

Oligonucleotide probe immobilization efficiency in HEMAA-based gel elements as a function of the time delay between printing and photopolymerization. After printing and photopolymerization, gel element arrays were briefly rinsed with water, air dried, and analyzed on a fluorescent microscope. Thereafter, arrays were washed in 0.1X SSPE, 0.5% SDS for 40 min at 45°C with constant agitation, rinsed with water, air dried, and re-imaged. Probe immobilization efficiency is calculated as the average residual fluorescence after the washing procedure over replicate arrays (n = 15 probes per array, one array per slide, 5 slides).

HEMAA thermal stability

The thermo-mechanical stability of HEMAA-based gel elements was evaluated in a series of thermal cycling experiments. The thermal cycling regime included an initial denaturation step at 95°C for 2 min followed by 50 to 100 cycles of [94°C for 15 sec, 60°C for 1 min], and final step at 72°C for 5 min with four different enzyme/buffer combinations. Regardless of the number of thermal cycles (between 50 and 100) or enzyme/buffer combination, we observed no detached or damaged gel elements and no discernable deviation in shape or size. These results indicate that a 0.47 M HEMAA-Bis copolymer is sufficiently durable to withstand extreme hybridization temperatures and on-chip thermal cycling procedures under a range of buffer and pH conditions.

Enhancing hybridization kinetics by modifying gel element pore structure

HEMAA monomer solved the fundamental challenge of monomer volatility during array manufacture and was shown to withstand rigorous thermal cycling procedures. In order to address pore size limits on target accessibility and enhance nucleic acid hybridization behavior in HEMAA relative to what is shown in Figures 1 and 2, we therefore synthesized a series of new cross-linkers that are compatible with UV-induced copolymerization methods (Table 2). AA was used as the reference monomer since it demonstrated the most rapid kinetics and highest end-point signal intensity in prior experiments (Figure 1 and 2). All of the cross-linkers possess the same potential functionality (number of polymerizable groups) Bis. At the same time, variations in the length of linker groups between unsaturated C=C bonds provide the opportunity to change the porosity of the gel network.

Table 2.

Monomers and cross-linkers used for manufacturing gel element microarrays.

Abbreviation Chemical Name Structure
AA Acrylamide CH2=CHCONH2
MA Methacrylamide CH2=C(CH3)CONH2
HEMAA 2-(hydroxyethyl) methacrylamide CH2=C(CH3)CONHCH2CH2OH
Bis N,N-methylenebisacrylamide CH2=CHCONHCH2NHCOCH=CH2
DMDAP N,N’-dimethacryloyl-1,3-diaminopropan-2-ol CH2=C(CH3)CONHCH2CH(OH)CH2NHCOC(CH3)=CH2
DMDAB N,N’-dimethacryloyl-1,4-diaminobutane-2,3-diol CH2=C(CH3)CONHCH2CH(OH)CH(OH)CH2NHCOC(CH3)=CH2
EDEMA 2,2-(ethylenedioxy)bis(ethylenemethacrylamide) CH2=C(CH3)CONHCH2CH2OCH2CH2OCH2CH2NHCOC(CH3)=CH2
Open in a new tab

Gel element arrays were fabricated with an equimolar concentration of monomer and a 20:1 molar ratio of monomer to cross-linker. A series of kinetic hybridization experiments were performed with synthetic 100-mer target T2, which is similar in size to a typical asymmetric PCR amplicon. Taking into account the data shown in Figure 1, AA arrays were subjected to the restoration procedure to avoid the loss of gel element functionality, whereas HEMAA arrays were produced without the restoration. The data in Figure 4 clearly show an interdependence between hybridization kinetics, end-point signal intensity and linker length, as 2,2-(ethylenedioxy)bis(ethylenemethacrylamide) (EDEMA) and N,N’-dimethacryloyl-1,4-diaminobutane-2,3-diol (DMDAB) (both with the longest distances between unsaturated C=C groups; Table 1) resulted in the most rapid and sensitive signal response curves; after 90 min of hybridization the fluorescent signal intensity for AA-EDEMA was 16 times higher than AA-Bis and HEMAA-EDEMA was 3 times higher than HEMAA-Bis. These results also demonstrate that the kinetics of hybridization in HEMAA gel elements is similar to AA (i.e. predictable and consistent with the principles of retarded diffusion), and that longer cross-linkers improve hybridization kinetics and end-point signals relative to conventional Bis-formulations (regardless of monomer). Both the increase in hybridization rate and non-equilibrium end-point signal intensities afforded by the alternative cross-linkers are important parameters for utilizing gel element arrays for either on-chip PCR experiments (37) or as a rapid diagnostic test.

Figure 4.

Figure 4

Open in a new tab

Kinetics of non-equilibrium hybridization of 100-mer target T2 on gel element arrays manufactured with a 20:1 molar ratio of monomer to cross-linker. Monomer and cross-linker abbreviations are as described in Table 2. Trend lines were added in Microsoft Excel.

Functional hybridizations with PCR amplicons

The effectiveness of HEMAA gel elements for detecting asymmetric PCR amplicons is illustrated in Figure 5 for a model orthopox array. Both AA and HEMAA arrays were manufactured with identical molar ratios of monomer to (Bis) cross-linker. The HEMAA arrays were polymerized 60 min after printing (no restoration step), whereas AA arrays included a restoration step. The hybridization profile for both monomers was very specific for camel pox, cow pox and vaccinia amplicons, with no detectable cross-hybridization to non-target probes at SNR >1.1. Importantly, empty gel elements prepared from HEMAA and AA demonstrate equivalent and very low background (at the level of the empty gel elements, nonsense probe gel elements, and bulk substrate), indicating that the nucleic acid targets and/or Cy3 labeled primers do not non-specifically react with HEMAA. These results also demonstrate the functional efficacy of HEMAA relative to the AA reference polymer for nucleic acid targets prepared by asymmetric PCR.

Figure 5.

Figure 5

Open in a new tab

Hybridization with single-stranded pox amplicons. Target amplicons are indicated above each bar graph. The signal to noise ratio (SNR) is calculated as the average probe signal (n = 12 over 3 replicate arrays) divided by the average dN20 signal (n = 12 over 3 replicate arrays). Image acquisition time = 10 sec.

DISCUSSION

Copolymerized gel element arrays are typically manufactured with AA or MA hydrogel supports because of their mechanical stability, low non-specific absorption and high probe immobilization capacity. During the course of our studies and attempts to scale up production, however, we discovered that both of these monomers were overly volatile, such that the manufacturing process required a lengthy restoration step that is intended to make up for the monomer loss over the time between drop deposition and UV photopolymerization (Figure 1). Unfortunately, the restoration procedure significantly complicates manufacturing scale up. This fundamental challenge to gel element array production and reproducibility was resolved by using HEMAA in lieu of AA or MA.

It is known that AA hydrogels have restricted permeability for high-molecular weight analytes. The impact of gel permeability on within-gel diffusion is compounded by the kinetic process of retarded diffusion that is inherent to gel element arrays (38-40). Different approaches have certainly been used to increase diffusion of targets to immobilized probes and enhance hybridization kinetics, such as electric field directed hybridization (41), microagitation (42), and microfluidic chaotic mixing (43). None of these methods, however, overcomes the basic constraint of polymer pore size. And while it is well known that the porosity of a hydrogel network can be increased by varying the ratio of monomer and cross-linker in the copolymerization mixture, by using cross-linkers other than Bis, or changing polymerization conditions (44, 45), increasing the pore size of AA reduces its mechanical durability and makes it susceptible to damage and deformation at even moderate temperatures (20, 25), limiting its utility for stringent hybridization experiments or more rigorous procedures such as on-chip thermal cycling. MA has therefore been substituted for AA as a gel-forming monomer, leading to gel elements with relatively high porosity and similar thermal stability compared to AA (17, 20). MA, however, is even more volatile than AA (Figure 1), which makes it an even more challenging monomer for manufacturing scale up and quality control.

Several other attempts have been undertaken to overcome the limiting diffusion of macromolecules within gels by incorporating hydrophilic groups into the structure of gel-forming monomers or by using charged polysaccharides, such as chitosan (31, 45). The “sugar polyacrylate” hydrogel described by Soto et al., for example, has pore sizes > 500 nm, it is mechanically robust and stable at high temperatures (30). However, there remain several significant limitations to these formulations for manufacturing scale-up and widespread application. For example, the fundamental repeat unit described by Sotto et al. (6-acryloyl-β-O-methylgalactopyranoside) is produced by enzymatic acrylation of the sugar. The resulting hydrogels are prepared as continuous slabs from a copolymerization mixture that also includes N-(3-aminopropyl) methacrylamide hydrochloride and Bis. The chemically-initiated polymerization should be carried out under nitrogen, overnight, by pre-incubating gel slabs with bis-succinimidile suberate, and then incubating with the spotted solution of amino-modified oligonucleotides. In short, sugar polyacrylates are not conducive to manufacturing scale up.

2-(hydroxyethyl) methacrylate (HEMA) is a monomer used for hydrogel microcapsulation of pharmaceuticals (33-35). It has also been shown that HEMA can be used successfully for the preparation of hydrogels by UV-initiated graft copolymerization with poly(ethylene glycol) (46). These mixed hydrogel matrices demonstrate discrete size-selectivity molecular penetration into the gel network, and show very low levels of non specific binding. Based on these data, we synthesized and used the analogue compound 2-(hydroxyethyl) methacrylamide (HEMAA) as a monomer for manufacturing hydrogel microarrays. We reasoned that HEMAA would have lower volatility than AA or MA and higher chemical stability over a wide range of pH and temperatures than HEMA because HEMAA has an amide chemical bond instead of an ester (47). These hypotheses were correct, as the volatility and thermal cycling stability data showed that HEMAA monomer is less volatile than both AA and MA and is mechanically stable under all tested thermal cycling conditions. In all kinetic and functional hybridization experiments, HEMAA gel element arrays behaved in similar fashion as AA or MA. The porosity of the HEMAA hydrogel network and resulting end-point signal intensities were readily increased by replacing Bis with one of the newly synthesized cross-linkers (Table 2 and Figure 4; this study). The combination of HEMAA monomer and new cross-linker regents therefore afford new opportunities to tailor gel element properties for general microarray science and specific nucleic acid diagnostic applications where manufacturability and quality control are primary technical drivers.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge Dr. Sofi Ibrahim (USAMRIID) for access to cloned DNA targets and PCR primer sequences. This work was supported in part by Grant Number R01 AI059517 from the National Institutes of Health (NIH). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the views of the NIH.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.Campas M, Katakis I. DNA Biochip arraying, detection and amplification strategies. Trends Anal. Chem. 2004;23:49–62. [Google Scholar]
  • 2.Goldsmith ZG, Dhanasekaran N. The microrevolution: applications and impacts of microarray technology on molecular biology and medicine (review) Int. J. Mol. Med. 2004;13:483–495. [PubMed] [Google Scholar]
  • 3.Lettieri T. Recent application of DNA microarray technology to toxicology and ecotoxicology. Environ. Health Persp. 2006;114:4–9. doi: 10.1289/ehp.8194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Stoevesandt O, Taussig MJ, He M. Protein microarrays: high-throughput tools for proteomics. Expert Rev. Proteomics. 2009;6:145–157. doi: 10.1586/epr.09.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Stoll D, Templin MF, Bachmann J, Joos TO. Protein microarrays: applications and future challenges. Curr.Opin. Drug Discov. Devel. 2005;8:239–252. [PubMed] [Google Scholar]
  • 6.Stoughton RB. Application of DNA microarrays in biology. Annu. Rev. Biochem. 2005;74:53–82. doi: 10.1146/annurev.biochem.74.082803.133212. [DOI] [PubMed] [Google Scholar]
  • 7.Jayapal M, Melendez AJ. DNA microarray technology for target identification and validation. Clin. Exp. Pharmacol. Physiol. 2006;33:496–503. doi: 10.1111/j.1440-1681.2006.04398.x. [DOI] [PubMed] [Google Scholar]
  • 8.Barbulovic-Nad I, Lucente M, Sun Y, Zhang M, Wheeler A, Bussmann M. Bio-microarray fabrication techniques. Crit. Rev, Biotechnol. 2006;26:237–259. doi: 10.1080/07388550600978358. [DOI] [PubMed] [Google Scholar]
  • 9.Hardiman G. Microarray platforms – comparisons and contrasts. Pharmacogenomics. 2004;5:487–502. doi: 10.1517/14622416.5.5.487. [DOI] [PubMed] [Google Scholar]
  • 10.Heller MJ. DNA microarray technology: devices, systems, and applications. Annu. Rev. Biomed. Eng. 2002;4:129–153. doi: 10.1146/annurev.bioeng.4.020702.153438. [DOI] [PubMed] [Google Scholar]
  • 11.Fixe F, Dufva M, Telleman P, Christensen CBV. Functionalization of poly(methyl methacrylate) (PMMA) as a substrate for DNA microarrays. Nucl. Acids. Res. 2004;32:e9. doi: 10.1093/nar/gng157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Dorris D, Nguyen A, Gieser L, Lockner R, Lublinsky A, Patterson M, Touma E, Sendera T, Elghanian R, Mazumder A. Oligodeoxyribonucleotide probe accessibility on a three-dimensional DNA microarray surface and the effect of hybridization time on the accuracy of expression ratios. BMC Biotechnology. 2003;3:6–17. doi: 10.1186/1472-6750-3-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Peterson AW, Heaton RJ, Georgiadis RM. The effect of surface probe density on DNA hybridization. Nucl. Acids Res. 2001;29:5163–5168. doi: 10.1093/nar/29.24.5163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Schepinov MS, Case-Green SC, Southern EM. Steric factors influencing hybridization of nucleic acids to oligonucleotide arrays. Nucl. Acids Res. 1997;25:1155–1161. doi: 10.1093/nar/25.6.1155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Prudnikov D, Timofeev E, Mirzabekov A. Immobilization of DNA in polyacrylamide gel for the manufacture of DNA and DNA-oligonucleotide microchips. Anal. Biochem. 1998;259:34–37. doi: 10.1006/abio.1998.2620. [DOI] [PubMed] [Google Scholar]
  • 16.Rehman FN, Audeh M, Abrams ES, Hammond PW, Kenney M, Boles TC. Immobilization of acrylamide-modified oligonucleotides by co-polymerization. Nucl. Acids Res. 1999;27:649–655. doi: 10.1093/nar/27.2.649. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Dykova VI, Dementieva EI, Zubtsov DA, Galanina OE, Bovin NV, Rubina AY. Hydrogel glycan microarrays. Anal. Biochem. 347:94–105. doi: 10.1016/j.ab.2005.09.009. 200. [DOI] [PubMed] [Google Scholar]
  • 18.Mikhleikin AI, Surzhikov SA, Zasedateleva OA, Vasiliskov VA, Pan'kov V, Grechishnikova IV, Kisselev LL, Zasedatelev AS. An RNA microchip containing immobilized oligoribonucleotides with protective groups at 2’-O-positions. BioTechniques. 2008;44:77–83. doi: 10.2144/000112677. [DOI] [PubMed] [Google Scholar]
  • 19.Rubina AY, Dementieva EI, Stomakhin AA, Darii EL, Pan'kov SV, Barsky VE, Ivanov SM, Konovalova EV, Mirzabekov AD. Hydrogel-based protein microchips: manufacturing, properties, and applications. BioTechniques. 2003;34:1008–1022. doi: 10.2144/03345rr01. [DOI] [PubMed] [Google Scholar]
  • 20.Rubina AY, Pan'kov SV, Dementieva EI, Pen'kov DN, Butygin AV, Vasiliskov VA, Chudinov AV, Mikheikin AL, Mikhailovich VM, Mirzabekov AD. Hydrogel drop microchips with immobilized DNA: properties and methods for large scale production. Anal. Biochem. 2004;325:92–106. doi: 10.1016/j.ab.2003.10.010. [DOI] [PubMed] [Google Scholar]
  • 21.Khrapko KR, Lysov YP, Khorlin AA, Ivanov IB, Yershov GM, Vasilenko SK, Florentiev VL, Mirzabekov AD. A method for DNA sequencing by hybridization with oligonucleotide matrix. DNA Sequence. 1991;1:375–388. doi: 10.3109/10425179109020793. [DOI] [PubMed] [Google Scholar]
  • 22.Preininger C, Sauer U, Obersriebnig S, Trombitas M. Signal enhancement of protein chips using 3D-materials for immobilization. Int. J. Envir. Anal. Chem. 2005;85:645–654. [Google Scholar]
  • 23.Vasiliskov AV, Timofeev EN, Surzhikov SA, Drobyshev AL, Shick VV, Mirzabekov AD. Fabrication of microarray of gel-immobilized compounds on a chip by copolymerization. BioTechniques. 1999;27:592–606. doi: 10.2144/99273rr06. [DOI] [PubMed] [Google Scholar]
  • 24.Guschin D, Yershov G, Zaslavsky A, Gemmel A, Shick V, Prudnikov D, Arenkov P, Mirzabekov A. Manual manufacturing of oligonucleotide microchips to enhance sequencing by hybridization. Anal. Biochem. 1997;250:203–211. doi: 10.1006/abio.1997.2209. [DOI] [PubMed] [Google Scholar]
  • 25.Barsky VE, Kolchinsky AM, Lysov YP, Mirzabekov AD. Biological microchips with hydrogel-immobilized nucleic acids, proteins, and other compounds: properties and applications in genomics. Mol. Biol. 2002;36:437–455. [PubMed] [Google Scholar]
  • 26.Brueggemeier SB, Wu D, Kron SJ, Palecek SP. Protein-acrylamide copolymer hydrogels for array based detection of tyrosine kinase activity from cell lysates. Biomacromolecules. 2005;6:2765–2775. doi: 10.1021/bm050257v. [DOI] [PubMed] [Google Scholar]
  • 27.Elghanian R, Xu Y, McGowen J, Seithoff M, Liu CG, Winick J, Fuller N, Ramakrishnan R, Beuhler A, Johnson T, Mazumder A, Brush CK. The use and evaluation of 2 + 2 photoaddition in immobilization of oligonucleotides on a three dimensional hydrogel matrix. Nucleosides Nucleotides Nucl. Acids. 2001;20:1371–1375. doi: 10.1081/NCN-100002558. [DOI] [PubMed] [Google Scholar]
  • 28.Grosejean L, Cherif B, Mercey E, Poget A, Levy Y, Marche PN, Villiers MB, Livache T. A polypyrrole protein microarray for antibody-antigen interaction studies using a label-free detection process. Anal. Biochem. 2005;347:193–200. doi: 10.1016/j.ab.2005.09.033. [DOI] [PubMed] [Google Scholar]
  • 29.Gurevitch D, Dong XF, Pircher TJ, Matsumoto SS, Roycroft P, Tsinberg P, Falcovitz YH, Hahn S. A novel three-dimensional hydrogel-based microarray platform. JALA. 2001;6:87–91. [Google Scholar]
  • 30.Soto CM, Patterson CH, Charles PT, Martin BD, Spector MS. Immobilization and hybridization of DNA in sugar polyacrylate hydrogel. Biotechnol. Bioeng. 2005;92:934–942. doi: 10.1002/bit.20665. [DOI] [PubMed] [Google Scholar]
  • 31.Yi H, Wu L-Q, Summner JJ, Gillespie JB, Payne GF, Bentley WE. Chitosan scaffolds for biomolecular assembly: coupling nucleic acid probes for detecting hybridization. Biotechnol. Bioeng. 2003;83:646–652. doi: 10.1002/bit.10712. [DOI] [PubMed] [Google Scholar]
  • 32.Chandler DP, Alferov O, Chernov B, Daly DS, Golova J, Perov A, Protic M, Robison R, Schipma M, White A, Willse A. Diagnostic oligonucleotide microarray fingerprinting of Bacillus isolates. J. Clin. Microbiol. 2006;44:244–250. doi: 10.1128/JCM.44.1.244-250.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Lester CL, Smith SM, Colson CD, Guymon CA. Physical properties of hydrogels synthesized from lyotropic liquid crystalline templates. Chem. Mater. 2003;15:3376–3384. [Google Scholar]
  • 34.Liu RH, Yu Q, Beebe DJ. Fabrication and characterization of hydrogel based microvalves. J. Microelectromech. Sys. 2002;11:45–53. [Google Scholar]
  • 35.Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27–46. doi: 10.1016/s0939-6411(00)00090-4. [DOI] [PubMed] [Google Scholar]
  • 36.Lizotte JR, Long TE. Stable free radical polymerization kinetics of alkyl acrylate monomers using in situ FTIR spectroscopy: influence of hydroxyl-containing monomers and additives. Macromol. Chem. Phys. 2004;205:692–698. [Google Scholar]
  • 37.Pemov A, Modi H, Chandler DP, Bavykin S. DNA analysis with multiplex microarray-enhanced PCR. Nucl. Acids Res. 2005;33:e11. doi: 10.1093/nar/gnh184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Livshits MA, Mirzabekov AD. Theoretical analysis of the kinetics of DNA hybridization with gel-immobilized oligonucleotides. Biophys. J. 1996;71:2795–2801. doi: 10.1016/S0006-3495(96)79473-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Sorokin NV, Chechetkin VR, Livshits MA, Vasiliskov VA, Turygin AY, Mirzabekov AD. Kinetics of hybridization on the oligonucleotide microchips with gel pads. J. Biomol. Struct. Dynam. 2003;21:279–288. doi: 10.1080/07391102.2003.10506923. [DOI] [PubMed] [Google Scholar]
  • 40.Zubtsov DA, Ivanov SM, Rubina AY, Dementieva EI, Chechetkin VR, Zasedatelev AS. Effect of mixing on reaction-diffusion kinetics for protein hydrogel-based microchips. J. Biotechnol. 2006;122:16–27. doi: 10.1016/j.jbiotec.2005.08.010. [DOI] [PubMed] [Google Scholar]
  • 41.Edman CF, Raymond DE, Wu DJ, Tu E, Sosnowski RG, Butler WF, Nerenberg M, Heller MJ. Electric field directed nucleic acid hybridization on microchips. Nucl. Acids Res. 1997;25:4907–4914. doi: 10.1093/nar/25.24.4907. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Toegl A, Kirchner R, Gauer C, Mikkelsen JC. Enhancing results of microarray hybridization through microagitation. J. Biomol. Tech. 2003;14:197–204. [PMC free article] [PubMed] [Google Scholar]
  • 43.Liu J, Williams BA, Gwirtz RM, Wold BJ, Quake S. Enhanced signals and fast nucleic acid hybridization by microfluidic chaotic mixing. Angewandte Chemie International Edition. 2006;45:3618–3623. doi: 10.1002/anie.200503830. [DOI] [PubMed] [Google Scholar]
  • 44.Patras G, Qiao GG, Solomon DH. Novel cross-linked homogeneous polyacrylamide gels with improved separation properties: investigation of the cross-linker functionality. Electrophoresis. 2001;22:4297–4302. doi: 10.1002/1522-2683(200112)22:20<4303::AID-ELPS4303>3.0.CO;2-E. [DOI] [PubMed] [Google Scholar]
  • 45.Righetti PG, Brost BCW, Snyder RS. On the limiting pore size of hydrophilic hydrogels for electrophoresis and isoelectric focusing. J. Biochem. Biophys. Methods. 1981;4:347–363. doi: 10.1016/0165-022x(81)90075-0. [DOI] [PubMed] [Google Scholar]
  • 46.Andersson A, Larsson A, Ekblad T, Leidberg B. Gradient hydrogel matrix for microarray and biosensor application: an imaging SPR study. Biomacromolecules. 2009;10:142–148. doi: 10.1021/bm801029b. [DOI] [PubMed] [Google Scholar]
  • 47.Nishiyama N, Suzuki K, Yoshida H, Teshima H, Nemoto K. Hydrolytic stability of methacrylamide in acidic aqueous solution. Biomaterials. 2004;25:965–969. doi: 10.1016/s0142-9612(03)00616-1. [DOI] [PubMed] [Google Scholar]

RESOURCES