The "ACID" solution is sufficiently concentrated to damage whatever it touches (eyes, skin, clothing). Wear goggles and handle with care. Rinse immediately should it get on your skin or clothing. It is phosphoric acid, so it is nontoxic when diluted. (Phosphoric acid is a major ingredient in soft drinks.)

The "SUB" (substrate/peroxide) solution contains organic chemicals and peroxide that are skin irritants and suspected carcinogens. Wear goggles, handle with care, and rinse any spills immediately.

Proteins spontaneously bind strongly and nonspecifically to chemically clean plastic (or glass) surfaces. Most ELISA assays depend upon the ability of protein antigen, or antibody, to bind in this manner to the bottoms of plastic wells in 96-well plates. This is called coating the well.

After the coat, or first layer of protein, is bound, additional layers of antibody or antigen are bound. These, however, must bind specifically only to the protein in the first layer. Therefore it is crucial to block the spontaneous nonspecific binding of these additional layers to the plastic well. One method of blocking nonspecific binding utilizes a protein such as bovine serum albumin to coat any plastic left exposed after the initial coating. A less expensive blocking reagent is a non-ionic detergent with the trade name of Tween-20. It coats bare plastic and prevents protein from binding, but does not remove previously bound protein -- and it does not inhibit antibody:antigen reactions nor the enzymatic reactions employed.

One method for ELISA is to coat the well directly with unknown IgG, which is reported with an enzyme-conjugated anti-IgG antibody (or conjugated second antibody, "direct" or "indirect" ELISA respectively). However, the binding of IgG to the plastic well is greatly dependent upon purity. For example, it is inhibited over 80% when mixed with 20 molecules of albumin per molecule of IgG. Since we wish to quantitate IgG in samples of varying purity, direct coating of the well is unsatisfactory.

We shall use a competition ELISA. All wells will be coated with the same amount of rabbit IgG. The unknown samples or known standards ("competitors" for short) will be mixed in advance with a constant amount of horseradish peroxidase-conjugated goat anti-rabbit IgG ("conjugate" for short). This mixture will be applied to the IgG-coated well. In the absence of competitor, a maximum amount of conjugate will bind to the well (giving maximum absorbance). As the concentration of IgG in the competitor increases, more and more of the conjugate is neutralized in solution, leaving less and less to bind to the well. So a low absorbance means a high concentration of IgG in the competitor. The method in outline:

1. Coat the test wells with a constant amount of rabbit IgG, rinse, optionally dry and store at room temperature until needed.
2. In a separate container (not the final test well), mix various dilutions of rabbit IgG "competitor" (standards or unknowns) with a constant amount of "conjugate".
3. Add conjugate-competitor mixture to coated well, incubate, rinse. (Free conjugate binds to the coat.)
4. Add constant amount of substrate to well, incubate (color develops).
5. Add acid to well to stop reaction.
6. Read absorbances of wells.

• COAT (rabbit IgG in PBS azide).
• PBST (phosphate buffered saline/tween; diluent for competitors).
• STD (standard IgG competitor, µg/ml will be announced).
• CONJ (horseradish peroxidase conjugate of goat anti-rabbit IgG or F(ab')₂).
• SUB (substrate for peroxidase, tetramethyl benzidine + peroxide in dimethylformamide).
• ACID (1 M phosphoric).
• Samples from each stage of your work with rabbit immunoglobulin ("unknowns").

Half-maximum (I₅₀). The final absorbance is most sensitive to the concentration of the competitor at the half-maximum inhibition by competitor, I₅₀. Often the maximum inhibition that can be obtained with high concentrations of competitor is less than 100%. Therefore it is important to determine it, since half-max is half-way between (i) the maximum absorbance (no competitor), and (ii) the minimum absorbance at maximum competitor.

Standard vs. unknown. Suppose we determine that pure IgG ("standard") produces I₅₀ at 0.65 mg IgG/ml. Now, if we find the dilution of an unknown that also gives I₅₀, at that dilution, there must be 0.65 mg IgG/ml. Then, the concentration of IgG in the unknown is simply the dilution (e.g. 9,500 fold) times 0.65 mg IgG/ml.

Standard Curves. Standard curves are used to determine the concentration of IgG that gives half-maximal inhibition. Results from a series of dilutions of known IgG standard can be plotted, and the concentration giving half-maximal inhibition (I₅₀) determined by interpolation. I₅₀ will occur at approximately 0.5 µg IgG/ml (concentration before mixing with conjugate). This will vary within a several-fold range from plate to plate, day to day, person to person. Therefore a standard curve must be included on each plate!

Unknowns. The goal with each unknown is to use a series of dilutions giving a curve that crosses I₅₀.

Blanks on 96-well Reader. The program we'll use for reading your plates on the 96-well reader will average wells A1, A2, and A3, and the average will be subtracted from the absorbances of the remaining 93 wells.

To minimize confusion, in plans and graphs, express all IgG concentations and competitor dilutions as they are BEFORE MIXING WITH CONJUGATE.

You will do a series of experiments. Each experiment is done in one 96-well plate. For each plate, you must first list the questions you want the plate to answer. Then you plan the plate in detail, and execute it. After you have the results, you may find that you need to do the plate again, either because your technique was not good enough, or because the design of the plate wasn't good enough. Once your experiment has answered the original question(s), you can complete the design details for a new plate that addresses additional questions. Your goal is to do enough plates to determine the concentrations of IgG in your unknown samples. The number of plates will vary from person to person depending on design and technique.

The first step is to plan your first plate. The ultimate question for the first plate is What concentration of IgG gives half-maximal inhibition? In order to answer this question, you'll have to answer some key questions about the method itself by the use of controls. Your first plate should have only controls plus a standard curve, all in quadruplicate. After you have succeeded in getting satisfactory results you can begin doing plates with unknowns. (What does "satisfactory" mean here?)

Controls. Think about what controls you need and what information each control gives you. In this type of experiment, a "control" is a test that establishes how much certain component(s) of the reaction contribute to the absorbance, in the absence of other component(s). Therefore each control involves omitting some component(s) of the reaction. Decide which of the possible controls are essential and which are optional.

Assay parameters. In the detailed procedure below, most of the assay parameters have already been worked out for you. But you still need to make decisions such as:

• Exactly which controls actually to do.
• How many replicates for each combination: singlicate, duplicate, triplicate, quadruplicate, quintuplicate?
• For the standard and for each unknown, you'll do a serial dilution. What is the highest concentration of each that you need to test? In other words, what dilution do you use at the start of the serial dilution?
• How many dilution steps do you need in each serial dilution?
• What size dilution step in the serial dilutions: 2-fold, 3-fold, 5-fold, or 10-fold steps?
• What kinds of containers to do your dilutions in: test tubes or wells?
• How to arrange all the tests in each plate (the "plate layout").

Check the design of each of your plates with an instructor before executing it. Every plate must include enough of a standard curve to confirm the I₅₀.

1. Wear goggles at all times (acid droplets might inadvertantly come sailing through the air from someone else ...).

   Coating wells.

2. Put only the wells to be coated with IgG in the well holder. Push the wells down firmly (so they won't flip out during rinses).
3. Add 50 µl "COAT" to each well. Tap the plate sideways against the bench to make sure the entire bottom of each well is covered.
4. Incubate 5 minutes. (The time here is not critical because at the concentration used, by 5 minutes, the well bottom is saturated.)
5. Rinse. Flip out the well contents into sink. Fill the wells with tap water, and flip out. Repeat 4 times for a total of 5 rinses. (It is OK if the coated well bottoms dry, but for reproducibility, it is best if all wells are uniformly dried or not. If you need to hold freshly-coated wells for later use on the same day, keep them filled with tap water and flip them out just before use. If you are going to store the coated wells between classes, store them dry.)

Premixing conjugate with competitors.

6. Make your serial dilutions of the competitors in test tubes.
7. Mix each dilution of each competitor with an equal volume of conjugate. Conjugate is expensive. For each mixture, don't make more than twice the volume that you plan to consume in the replicate test wells.
8. Premixtures of competitors with conjugate should be allowed to react at room temperature for at least 20 minutes. Longer is OK (even hours), but be aware that inhibition of absorbance will be slightly greater with longer times. (What is happening during this time?)

   Binding conjugate to the coat.

9. Put 50 µl of premixed competitor + conjugate per well according to your plan. Change pipet tips for each series. Within each series, if you start with the highest concentration of competitor and move towards lower concentrations of competitor, you can use one tip for the whole series.
10. Incubate at room temperature for 15 minutes. (Timing should be accurate here since conjugate binding is not saturated under these conditions.)
11. Rinse 5 times with tap water.

    Enzymatic reaction.

12. Add 50 µl "SUB" (TMB/P substrate) per well.
13. Incubate 15 minutes, mixing several times during the incubation by tapping the plate sideways against the bench. Time here should be controlled carefully.
14. Add 50 µl "ACID" to each well. (Squirt into well from above to avoid contaminating the tip with color, and to allow you to use one tip for all wells.)
15. Read absorbances at 450 nm.
16. When you are sure you have your data, flip well contents into sink, rinse several times, and push the wells out of the well holder into the wastebasket. Save the well holder!

Data analysis.

17. Plot your data on semilog graph paper: concentration of known (µg IgG/ml) or fold-dilution of unknown on the abcissa (bottom axis), and A₄₅₀ on the ordinate (side axis).
18. Interpolate to find the concentration or fold-dilution giving I₅₀.
19. Calculate the IgG concentrations in your unknowns.
20. You now have total protein (from A280) and IgG (from ELISA). For each unknown, calculate the percentage of total protein that is IgG.

    Lab notebook.

21. There is no formal report for your ELISA work. Your lab notebook will be handed in and graded for the semester's work. For ELISA, you should have a clear summary list of questions and answers, with brief comments and interpretation. Each answer should cite a particular plate number in the series of plates you did.
22. You and your lab partner will be analyzing the same samples. Include a table comparing results, and briefly discuss any signficant discrepancies.
23. Before handing in your lab notebook, add a table of contents with page numbers so, for example, we can find your ELISA summary easily, then find the section for each cited plate easily. Review the Policies guide handed out at the beginning of the semester (and on the web) for other lab notebook requirements.