Final Report: Microarray System for Contaminated Water Analysis

EPA Grant Number: R830420C004
Subproject: this is subproject number 004 , established and managed by the Center Director under grant R830420
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: Center for Environmental and Energy Research (CEER)
Center Director: Earl, David A.
Title: Microarray System for Contaminated Water Analysis
Investigators: Cardinale, Jean , DeRosa, Rebecca
Institution: Alfred University
EPA Project Officer: Lasat, Mitch
Project Period: September 1, 2003 through August 31, 2005
RFA: Targeted Research Center (2004) Recipients Lists
Research Category: Hazardous Waste/Remediation , Targeted Research


Enzyme-linked immunosorbent assays (ELISA) can be used to detect biomolecules in solution. Current methods of microbial detection depend mainly on cultivation of indicator organisms or genetic (PCR) detection methods, both of which have significant drawbacks. An alternate method for detection, ELISA, relies on immunological detection of organisms. The traditional ELISA employs a polystyrene wellplate, which has a few drawbacks: the production of large amounts of polymer and biologically contaminated waste, low throughput, and the need for a large volume of expensive, high-purity antibodies and consumable reagents. Recently, there has been an effort to modify the technique to minimize waste and consumables by miniaturizing the assay and developing what has now been termed a microarray. There is interest to develop a substrate which covalently binds protein, thus providing the potential for a stringent and reusable system.

Micro reactor, micro array, and microfluidic technologies are all examples of miniaturized assaying techniques. One of the main issues in micro technologies is surface and interfacial response. The microarray typically employs a flat glass substrate and, similar to the ELISA, can capture a biomolecule, such as an antigen, with an adsorbed antibody. One way to optimize microarray sensitivity is to increase the binding concentration of the capture antibody by physicochemical tailoring of the substrate. Most of the research in this field is devoted to chemically modifying the surface to create a permanent binding site for the primary antibody.

Physical and/or chemical mechanisms can be used to increase the adsorption of proteins to a surface. By increasing the substrate surface area, there is potential for more protein to attach. A simple method to produce a favorable environment for protein adsorption is to create a functionalized surface that will covalently interact with the protein. Common methods for attachment use noncovalent interactions, often due to the hydrophobic adsorption. Noncovalent interaction does not guarantee permanent immobilization of the protein, leaving the protein susceptible to removal during washing steps. Therefore, it may be advantageous to use a covalent attachment method. To this end, a physicochemical glass modification was investigated to increase the protein binding. Initially, a hydrophilic surface was created using a water-plasma treatment, which created a strong reaction with a silane. Physical modification of the surface was performed using a photo-hydrolytic treatment using UV exposure in water at ambient temperatures.

This study was directed toward creating a reusable, miniaturized multianalyte device (microarray) that can be used as a sensor for detection of multiple water-microbial contaminants—essentially a miniature ELISA system. The focus of research for this project was on immunoassays and an antibody:antigen arrays. ELISA analysis is becoming more and more widely used in many fields, due to the ease of use over traditional analytic methods. Areas which are seeing an increase in ELISA analysis include pesticide and chemical analysis of environmental samples and human biomonitoring, as well as pathogen and toxin detection in recreational and drinking water, and in food preparation. While this miniature ELISA system was developed for water analysis, the technology has enormous potential in many areas, such as medical diagnostics, detection of biowarfare agents, and drinking water analysis, and it may be adapted for any ELISA system in order to reduce waste and improve detection capabilities. In addition, since the reusable microarray requires fewer reagents and has higher detection sensitivity at lower concentrations of antigens, contact with hazardous materials by personnel involved in the assaying is likewise reduced.

The objectives of this project were:

  1. To develop and optimize a novel method for glass derivitization followed by antibody absorption to the glass surface in order to develop a multiplexed “ELISA-on-a-chip” system for antigen analysis.
  2. To adapt this novel ELISA system for pathogen analysis in local recreational waters.

This system has the potential to fulfill a demand for faster and more cost effective means of testing for multiple pathogens in environmental samples, and due to a change in the materials and process, reduce the waste stream generated from traditional ELISA analysis.

Summary/Accomplishments (Outputs/Outcomes):

We used commercially available Corning 2947 (soda-lime-silicate [SLS]) and Corning Microarray Technology 1737 (proprietary alkaline earth aluminoborosilicate [1737]) slides as the substrates. Chemical modification was investigated by treating both glass compositions with two silane compositions (3-aminopropyl-triethoxysilane [APS], and 3-mercaptopropyl-trymethoxysilane [MPTS]), resulting in a functional terminus that covalently binds antibody and provides more stringent binding between substrate and antibody.

Primary and secondary antibodies and antigens were obtained and prepared for use in testing the substrates. Control assays were conducted with Escherichia coli in a traditional ELISA method to confirm specificities of the antibodies.

The substrate surfaces were treated with 10 procedures to optimize antibody covalent binding. The procedures were combinations of the following steps: water-plasma treatment, photo (UV)-hydrolytic treatment (deionized [DI] water), silane treatment, and photo (UV)-hydrolytic-co-silane treatment. Contact angle measurements were taken immediately after each step to characterize the glass treatment. Surface phases and topography of the treated slides were analyzed using atomic force microscopy (AFM). We used X-ray photoelectron spectroscopy (XPS) to characterize the elemental composition of the surface. The slides were printed with antibodies specific to indicator organisms (E. coli) and treated according to ELISA protocol. Direct microarray and ELISA assays were prepared and tested.

Microarray performance was evaluated by means of a fluorophore standard curve. A standard intensity curve for the fluorophore was calculated to correlate the intensity of antibody microarray features to a known concentration of deposited antibody within a detectable range. The dynamic range of the curve was determined to be the linear portion between the lowest detection value (concentration of 2 μg/mL) and the saturation point (60,000 fluorescence units). The fluorescence units are assumed to be directly proportional to the amount of antibody on the surface within the dynamic range.

The treated slides were arrayed with FITC- conjugated antibody and washed with a PBS/Blotto solution to determine binding efficiency of antibody to slide surface. The results of binding efficiency tests are shown below (Figure 1).

Figure 1. Binding Efficiency of ug/mL IgG GtxRb (FITC) on Treated SLS and 1737 Slides.

    Figure 1. Binding Efficiency of μg/mL IgG GtxRb (FITC) on Treated SLS and 1737 Slides. Standard deviation bars are equal above and below the bar. Values are the average of two slides.

Of the two silane treatments tested, APS-treated slides resulted in higher fluorescence intensity, correlating to more binding capacity, than the MPTS treatments. Specifically, the difference between treatments D and E (Figure 1, shown above) indicates that APS results in nearly three times as much fluorescence intensity as MPTS and would therefore have a greater potential to bind more antigen and be able to detect lower concentrations of antigens.

The effects of photo-hydrolytic treatment on binding efficiency in treatment D indicated a significant increase in antibody adsorption compared to treatment A (plasma, APS). The difference between the treatments is likely due to the increase in surface area caused by photo-hydrolytic treatment, as shown with AFM analysis. The intensity difference between treatments B (plasma, MPTS) and E is insignificant within the standard deviation, suggesting that the increase in surface roughness after photo-hydrolytic treatment does not have a positive impact on antibody binding when MPTS is used. Treatment F showed an increase in antibody adsorption compared to treatment A ; however, it was no greater than treatment D. The differences between treatments D and F are insignificant, due to differences in standard deviations. For this reason, both were chosen as the optimum treatments, due to the high antibody intensity.

Variations were insignificant between treatments A and B, limiting any definitive conclusions to be drawn about the effects of glass composition on antibody concentration. The difference in fluorescence intensity between compositions is distinct for treatments after photo-hydrolysis. The SLS slides showed a higher binding intensity than the 1737 slides, suggesting that the increase in surface area after photo-hydrolytic treatment, attributed to higher leaching behavior of SLS slides due to UV treatment, resulted in an increased antibody binding concentration.

A comparison of antibod y binding efficiency of the commercial systems with SLS slides Treatment D (photo-hydrolytic, APS) and Treatment F (photo-hydrolytic-co-APS) is shown below (Figure 2). The commercial slides performed comparably to the treated slides, within the standard deviation.

Figure 2. Binding Efficiency of 5 ug/mL IgG GtxRb (FITC) on Most Efficient SLS Treatment D [Photo Hydrolytic, APS] and F [Photo-Hydrolytic-co-APS], Compared to Commercial Slides.

    Figure 2. Binding Efficiency of 5 μg/mL IgG GtxRb (FITC) on Most Efficient SLS Treatment D [Photo Hydrolytic, APS] and F [Photo-Hydrolytic-co-APS], Compared to Commercial Slides. Standard deviation lines are equivalent above and below the bar.

Direct Binding Assay

Treated slides were tested for direct binding of E. coli (cryptic cat:GFP) with an adsorbed anti-E. coli capture agent. Unfortunately, none of the treatments resulted in any discernable binding, as there was no observed fluorescence after incubation with E. coli. The fluorescence of the E. coli (cryptic cat:GFP) was confirmed prior to each incubation round, indicating that the lack of detection was not due to the quenching of the GFP. To ensure the anti-E. coli was not denatured before arraying on the slide, a standard ELISA test was conducted using multiwell plates. The results from the traditional ELISA confirmed that binding of the antibody and antigen occurred. Because the same protocol was followed for the traditional ELISA and the treated slides, it indicates that there is limited or no interaction between the surface-bound capture antibody and the E. coli. A plausible reason would be improper immobilization of the bound antibody. Inactivity of the antibody could be due to a side-on orientation in which the molecule is flat on the surface. This orientation would prevent interaction with an antigen, thus disabling the system.

Additionally, there is the possibility that the potential binding sites for the antigen (i.e., the number of adsorbed anti-E. coli) are too few to provide sufficient detection in the scanning system. We determined the dimensions of the antibody and potential contact with the surface—the theoretical value of the available binding sites. The number of antibody arrayed onto the surface was considerably less than the theoretical amount which could be bound. Therefore, by not maximizing the amount of antibody within the feature, the number of antigen that it can capture is limited, and the detection of low antigen concentrations is improbable.


Glass Characterization. SLS and 1737 slides were treated and characterized using contact angle, AFM, and XPS. Water- plasma treatment effectively removed organics and introduced a hydrophilic surface ideal for silane deposition. Photo-hydrolytic treatment increased the surface roughness and the hydrophobic behavior, potentially due to environmental contamination. Silane functionalization increased the hydrophobic behavior, due to the introduction of an organized hydrocarbon. Contact angle and XPS analysis highlighted surface differences between treatment D (photo-hydrolytic, APS) and treatment F (photo-hydrolytic-co-APS), proving that the UV was altering the silane, resulting in decreased nitrogen and carbon concentrations on the surface. There was little difference in contact angle behavior of antibody between effects of the glass compositions for slides not photo-hydrolytically treated. The photo-hydrolytic effect did show a greater change of surface roughness in SLS than in 1737 slides, due to ion leaching and the subsequent increase in surface area and molecular rearrangement.

Microarray Preparation. Binding efficiency tests were performed to test the antibody immobilization after washing with PBS/Blotto and a sodium dodecyl sulfate (SDS) detergent rinse. PBS/Blotto washes determined that treatment D [photo-hydrolytic, APS] maintained nearly 100% of the initially arrayed antibody; however, multiple rinsing steps indicated a 30% loss of antibodies which were not sufficiently bound to the surface. An SDS wash concluded that much of the protein was adsorbed to the surface by hydrophobic forces, as witnessed by high percent loss of antibody for most treatments. Treatment F (photo-hydrolytic-co-APS) showed the least amount of protein loss, maintaining 87% of the antibody on the surface, suggesting that stronger covalent interaction is occurring for this treatment. Tests also indicated that binding efficiency of t reatment D (photo-hydrolytic, APS) and F (photo-hydrolytic-co-APS) was comparable to commercial slides; however, commercial slides showed less variation in fluorescent intensity. The overall binding efficiency of the treatments was dependent on the hydrophobic behavior of the surface and the silane treatment. Treatment D (photo-hydrolytic, APS) exhibited the most hydrophobic surface and the highest binding efficiency. This corroborates the fact that hydrophobic interactions are significant in the adsorption of the antibody to the surface.

Direct Binding Assay. Direct binding assays did not indicate an interaction between the capture anti-E. coli and E. coli (cryptic cat:GFP). A traditional ELISA confirmed that the anti-E. coli and E. coli were interacting when adsorbed to a well-plate, indicating that the capture antibody was improperly immobilized on the glass surface such that antigen binding did not occur.

The current work established the use of a physicochemical method to develop a substrate for an ELISA. To fully characterize the surface, comprehensive XPS analysis of the treatments, including depth profiling, should be performed. This would provide more information to correlate the dependence of antibody binding efficiency to a functionalized surface. Additionally, experiments should be performed that focus on attaining proper immobilization of the molecule, such that it would be a sufficient to support antigen attachment. Correct immobilization of the antibody on the substrate would allow for retesting of the direct binding and testing of a multianalyte system.


Crowther JR. The ELISA Guidebook. Humana Press, 2000.

Holland L. A review of plasma process studies. Surface Technology 1980;11:145-169.

Zhou X. Effect of plasma treatment on soda-lime-silicate glass surfaces. Masters Thesis. Alfred University, 1990.

Escoubas L, Roche P, Commandre M. Front and back illumination of coated substrates for in-depth localization of absorption. Optical Engineering 1997;36(4):1196-1203.

Kajihara K, Ikuta Y, Oto M, Hirano M, Skuja L, Hosono H. UV-VUV laser induced phenomena in SiO2 glass. Nuclear Instruments and Methods in Physical Research Section B 2004;218:323-331.

Shelby JE. Introduction to glass science and technology. Cambridge: RSC Paperbacks, 1997, 244 pp.

Shriver-Lake LC, Donner B, Edelstein R, Breslin K, Bhatia SK, Liger FS. Antibody immobilization using heterobifunctional crosslinkers. Biosensors and Bioelectronics 1997;12(11):1101-1106.

Bhatia SK, Shriver-Lake L, Prior K, Georger JH, Calvert JM, Bredehorst R, Liger FS. Use of thiol-terminal silanes and heterobifunctional crosslinkers for immobilization of antibodies on silica surfaces. Analytical Biochemistry 1989;178:408.

Melzak KA, Brooks DE. Modification of silica with a covalently attached antigen for use in immunosorbent assays. In: Horbett TA, Brash JL, eds. Proteins at Interfaces II: Fundamentals and Applications Washinton, DC: American Chemical Society, 1995.

Heiney P, Gruneberg K, Fang J, Dulcey C, Shashidhar R. Structure and growth of chromophore-functionalized (3-Aminopropoyl)triethoxysilane self-assembled on silicon. Langmuir 2000;16(6):2651-2657.

Huetz P, Schaaf P, Voegel J-C, Mann EK, Miras B, Ball V. Reactivity of antibodies on antigens adsorbed on solid surfaces. In: Horbett TA, Brash JL, eds. Proteins at Interfaces II: Fundamentals and Applications. Washington, DC: American Chemical Society, 1995.

Liger FS, Georger JH, Bhatia SK, Calvert JM, Shriver-Lake L, Bredehorst R. Immobilization of active agents on substrates with a silane and heterobifunctional crosslinking agent. U.S. Patent No. 5,077,210.

DeRosa RL, Shrader P, Shelby JE. Hydrophilic nature of silicate glass surfaces as a function of exposure condition. Journal of Non-Crystalline Solids 2003;331(1-3):32-40.

Delehanty JB, Liger FS. A microarray immunoassay for simultaneous detection of proteins and bacteria. Analytical Chemistry 2002;74(21):5681-5687.

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other subproject views: All 3 publications 2 publications in selected types All 1 journal articles
Other center views: All 34 publications 8 publications in selected types All 6 journal articles
Type Citation Sub Project Document Sources
Journal Article DeRosa RL, Cardinale JA, Cooper A. Functionalized glass substrate for microarray analysis. Thin Solid Films 2007;515(7-8):4024-4031. R830420 (Final)
R830420C004 (Final)
X832541C001 (Final)
  • Abstract: Science Direct Abstract
  • Supplemental Keywords:

    ELISA, lab on a chip, water contaminants, glass surface modification,, RFA, Scientific Discipline, Water, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Environmental Chemistry, Aquatic Ecosystem, Environmental Monitoring, Recreational Water, pathogens, recreational water monitoring, risk assessment, microorganisms, aquatic environments, water quality criteria, water quality, microarray system

    Relevant Websites: Exit

    Progress and Final Reports:

    Original Abstract
  • 2004 Progress Report

  • Main Center Abstract and Reports:

    R830420    Center for Environmental and Energy Research (CEER)

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R828737C001 Environmental Impact of Fuel Cell Power Generation Systems
    R828737C002 Regional Economic and Material Flows
    R828737C003 Visualizing Growth and Sustainability of Water Resources
    R828737C004 Vibratory Residual Stress Relief and Modifications in Metals to Conserve Resources and Prevent Pollution
    R828737C005 Detecting and Quantifying the Evolution of Hazardous Air Pollutants Produced During High Temperature Manufacturing: A Focus on Batching of Nitrate Containing Glasses
    R828737C006 Sulfate and Nitrate Dynamics in the Canacadea Watershed
    R828737C007 Variations in Subsurface Denitrifying and Sulfate-Reducing Microbial Populations as a Result of Acid Precipitation
    R828737C008 Recycling Glass-Reinforced Thermoset Polymer Composite Materials
    R828737C009 Correlating Clay Mineralogy with Performance: Reducing Manufacturing Waste Through Improved Understanding
    R830420C001 Accelerated Hydrogen Diffusion Through Glass Microspheres: An Enabling Technology for a Hydrogen Economy
    R830420C002 Utilization of Paper Mill Waste in Ceramic Products
    R830420C003 Development of Passive Humidity-Control Materials
    R830420C004 Microarray System for Contaminated Water Analysis
    R830420C005 Material and Environmental Sustainability in Ceramic Processing
    R830420C006 Interaction of Sealing Glasses with Metallic Interconnects in Solid Oxide and Polymer Fuel Cells
    R830420C007 Preparation of Ceramic Glaze Waste for Recycling using Froth Flotation
    R830420C008 Elimination of Lead from Ceramic Glazes by Refractive Index Tailoring
    R830420C010 Nanostructured C6B: A Novel Boron Rich Carbon for H2 Storage