Paper-based ELISA.

This paper describes enzyme-linked immunosorbent assays (ELISA) performed in a 96-microzone plate fabricated in paper (paper-based ELISA, or P-ELISA). ELISA is widely used in biochemical analyses; these assays are typically carried out in microtiter plates or small vials. ELISA combines the specificity of antibodies with high-turnover catalysis by enzymes to provide specificity and sensitivity. We have recently described a 96-microzone paper plate— fabricated by patterning hydrophobic polymer in hydrophilic paper—as a platform for biochemical analysis. Although microfluidic paper-based analytical devices (mPADs) were designed primarily to provide analytical capability at low cost in developing countries, we expect that they will also be useful in applications such as point-of-care clinical analysis, military and humanitarian aid field operations, and others where high throughput, low volumes of sample, low cost, and robustness are important. These devices have so far been prototyped using analyses of simple analytes: glucose, total protein, and certain enzymes. P-ELISA combines the sensitivity and specificity of ELISAwith the convenience, low cost and ease-of-use of paper-based platforms; P-ELISA (at it current state of development) is faster and less expensive than conventional ELISA, but somewhat less sensitive. Porous membranes, including nitrocellulose and filter paper, have been used for decades in dot-immunobinding assays (DIA). Though DIAs are the simplest form of immunoassays on paper, they typically require one piece of nitrocellulose for each assay; the pieces of nitrocellulose have to be processed individually in Petri dishes, and the assays take several hours to complete. Quantitative DIAs have been reported, but DIAs are typically qualitative, and provide only “yes/no” results. Conventional ELISA, usually performed in 96-well plates (fabricated by injection molding in plastic), is quantitative and well-suited for highthroughput assays, but each assay requires large volumes (ca. 20–200 mL) of analyte and reagents, the time required for incubation and blocking steps are long ( 1 h per step, because the reagents must diffuse to the surface of the wells), and the results are usually quantified using a plate reader, typically a $20000 instrument. Paper microzone plates for ELISA can have the same layout as plastic 96-well plates, but each test zone requires only about 3 mL of sample, and the results can be measured using a desktop scanner, typically a $100 instrument. In addition, an entire P-ELISA can be completed in less than one hour. The ease of fabrication of paper microzone plates also opens opportunities for a wide range of non-standard formats, and customized connections to carry reagents between zones. To evaluate the feasibility of P-ELISA, and the potential advantages and disadvantages of P-ELISA and 96-well-plate-based ELISA, we adapted a standard procedure to our format and then demonstrated an indirect P-ELISA using rabbit IgG as a model analyte. We also established that P-ELISA can be used to detect and quantify antibodies to the HIV-1 envelope antigen gp41 in human serum using an anti-human IgG antibody conjugated to alkaline phosphatase (ALP) to produce a colorimetric readout. We used a 96-microzone paper plate with an array (12 8) of circular test zones for running multiple P-ELISAs in parallel (Figure 1A); the Supporting Information describes the details. The array was designed to have the same layout and dimensions as a standard plastic 96-well plate, so that it would be compatible with existing microanalytical infrastructure (eightor twelve-channel pipettes and plate readers). Each test zone was 5 mm in diameter and required 3 mL of solution to fill (e.g., to wet completely with fluid); this design was a good compromise between convenience and conservation of reagents, as it reduced the amount of reagents and sample required for the assay but ensured accurate distribution of fluids when using a manual pipette. We also examined smaller test zones, with the smallest test zone requiring 0.5 mL of solution to fill (e.g., to wet completely). This size is similar to that required in a 384-well plate format. The top and bottom faces of the test zones in papermicrozone plates are open to atmosphere. The advantage of this configuration is that the zones can be washed by adding a washing buffer to the top of the zone while pressing the bottom of the zone against a piece of blotting paper. The washing buffer goes through the test zone vertically and into [*] Dr. C.-M. Cheng, Dr. A. W. Martinez, Dr. J. Gong, Dr. C. R. Mace, Prof. S. T. Phillips, Prof. E. Carrilho, K. A. Mirica, Prof. G. M. Whitesides Department of Chemistry and Chemical Biology Harvard University Cambridge, MA 02138 (USA) E-mail: [email protected] Homepage: http://gmwgroup.harvard.edu