Figure A
The Sandwich EIA technique allows for the detection of a specific antigen in a fluid sample. For example, to detect hCG (a pregnancy hormone) in urine.
For this method the antigen samples are added to wells containing a specific antibody. If the specific antigen is present in the sample, it will bind to the antibody coating. The wells are then rinsed, removing any unbound antigens out of the wells. Another antibody, identical to the first (except that it is conjugated to an enzyme) is now added to the wells. The enzyme-linked antibody binds to any exposed antigenic sites, which must be polyvalent (have multiple binding sites so that a sandwich is able to form. The wells are rinsed again and a chromogenic substrate is added to the wells. This is what allows the colour to develop. The enzyme attached to the antibody catalyses the colour reaction. If no antigens are present, only the antibody coating will be attached to the well surface. Therefore if no antigens bind, the second antibody has no place to attach and is also washed out of the wells. Without the enzymes, there will be no colour reaction.
A competitive EIA is used to determine the concentration of an antigen.
The specific antigens are adsorbed to a well surface. A sample to be tested for the same antigen is pre-mixed with the primary antibody. This step allows the formation of an Antibody-Antigen complex in the sample fluid. If any antibodies are used up in this way, they are no longer available to bind to the antigens in the well. The sample mixture is now added to the wells. Only antibodies that have not already been bound up by antigen in the sample can bind to the well surface. The number of antibodies left to bind to the well is therefore inversely proportional to the amount of antigen in the sample. In other words, the more antigen that was present in the sample, the fewer antibodies are left to bind to the well. The wells are then rinsed, and an enzyme-linked secondary antibody is added to the wells. The number of secondary antibodies is proportional to the number of primary antibodies bound. A chromogenic substrate is added. The intensity of the colour developed in a given amount of time is proportional to the number of antibodies. The colour intensity is compared to a standard curve to determine antigen concentration.
Other methods that are used as part of ELISA technique are the direct method, the capture method and the Elispot assay.
A method to improve these techniques is by treatment of polyclonal Ig using a serum column. Here, a CNBr activated gel is used and incubation occurs at a much lower temperature at only 4oC.
Procedures similar to that of ELISA that can be used are:
- Radioimmunoassay (RIA), a method similar to that of ELISA except a radio-labelled ligand is used.
- Immunodiffusion for quatitation of antigens
- Immunoelectrophoresis for separation and detection of blood proteins
- Immunofluorescence and flow cytometry
- Radioallerosorbent test (RAST)
- Western Blotting
Conclusion
Figure B shows a scatter graph of how the optical density varies with the twelve different dilutions used. It shows that serum A had variations in optical density with different dilutions, whereas serum B did not. Many of the results for serum B were negative and therefore converted into zero, to plot the graph, as a negative reading indicates no binding between the antibody and antigen. The line of best fit plotted for serum A1 and A2 shows that as optical density increases so does the concentration of the serum (becoming more dilute). The optical density for serum A2 is greater for the more concentrated solutions.
Figure C represents the same data as figure A, however the scale of the x-axis is different. Here, instead of plotting the actual dilutions, I have coded each dilution
1-12, so that each point is equally spaced from each other. By doing it this way the graph goes in order of the dilutions in the wells, where 1 represents the most concentrated i.e. 1/64μl and 12 is the most dilute, i.e. 1/131072μl.
Figure C is generally a better way to display the results as it is easier to read off data and goes in the correct direction of concentration, and the points are not all un-proportionally placed like they are in B. The end-point can be determined using figure C, whereas in figure B this cannot be done as it does not show where the values become negative or cross the x-axis. However, with figure C, it is not possible to determine the optical density of a dilution other than that of those that are plotted as it is not in a scale that is proportional to the dilutions.
The reason for this trend is due to the mechanism of the process. Where the concentration of the serum is high, there is a high optical density. The presence of antibodies can be detected with the aid of the conjugate, anti-rabbit IgG. The antibodies are conjugated to an enzyme, in this case phosphatase. The substrate, p-nitrophenyl detects the amount of bound conjugate by the degree of colour produced. Therefore, the more concentrated the serum, the more antibodies are present, so there will be a greater number of antibody-antigen binding, which will be detected by the substrate, so there will be greater colouration resulting in a higher optical density. Figure D shows a diagrammatic representation of this process.
As can be seen from my results, only serum A showed positive results indicating a colour change, therefore antibody-antigen binding. This shows that it was serum A that was specific for the antigen present. Serum B was not the complementary antibody for the antigen given therefore did not have a positive optical density. However, there were some positive optical density readings, especially in serum B2. These results may be a false positive. This is because there could be antibodies present in the serum that although may not bind perfectly to the antigen, but may bind close to perfect, indicating a positive result. Also some of the results found for serum A were zero (negative), which shows that not all the antigens were bound to.
The readings from the controls show no change in colour, however, that of serum A with antigen did show a slight optical density. Those controls without antigen will obviously show no positive results as the antibodies added will not bind to anything, therefore will just be washed away. Those that were in the wells containing antigen, should also not present any positive results as those with wash buffer and conjugate have no antibodies present, hence no binding, therefore no change in colour. Those that did contain the serum, did not have any conjugate, so the substrate cannot detect anything. The reason for a slight optical density for serum A with antigen may be due to experimental errors, which are listed below.
Figure E and F (in appendix) both show the graphs for the results where the negative results have not been converted to zero. As figure F shows compared to figure C the end-point value is different, therefore all negative values must be changed to zero before plotting the graph.
Possible experimental errors that could have occurred are:
- Poor precision – not washing the wells thoroughly enough
Insufficient aspiration of wells – wells should be dry after aspiration
- Insufficient mixing of reagents
- Unequal volumes added to wells
- Pipetting errors
- Insufficient volume of colour reagent added to wells
- Improper mixing of colour reagents
- Fluctuating temperatures around the work surface affecting incubating temperature.
- Insufficient sealing of plate
- Horseradish peroxidase is about ten times more sensitive than alkaline phosphatase, and is used with safe colour development solution, hence it can be used to detect very small levels of antibody. Because of its high sensitivity, the colour produced is more variable, which may be due to the presence of particles from the air, inadequate rinsing or temperature variability.
Questions
1 – Dilutions are coded 1-12 rather than the actual numerical values i.e. 1/64 to 1/131072 because the range of dilution values are too wide spread. Plotting the actual values would present clustering at the beginning of the graph and as you move along the x-axis the spread increases. Also the end-point cannot be determined if the dilutions are plotted as the trend is positive rather than negative because the dilutions are decreasing rather than increasing.
2 – using the data from figure 1, the end-point for serum A can be calculated by at what dilution the trend line crosses the x-axis as this is the point at which there is the lowest positive value. The end point of a sample may be defined in relation to the highest dilution. For figure 1 the end point is 1.5259 × 10-5.
For my experimental data, I can only determine an end-point for serum A2 as A1 does not cross the x-axis. As it crosses at the tenth dilution, the concentration is 3.0518 × 10-5.
3 - the optical density has plateaued at the high concentration of serum A due to there being a fairly equal optical density at these concentrations. There is greater colour intensity here, as there will be more formation of antibody – antigen complexes. As it becomes more dilute, the optical density begins to decrease.
This may be due to the fact that the difference in dilution at the start is a lot less than as the dilutions proceed.
Figure 3 uses the same data as figure 2 but adds trend lines. I think that this is a better way to present the results as it shows both the actual results as well as the general trend so different interpretations can be made, such as the rate at which the optical density decreases (by calculating gradient).
4 – the errors that could have been introduced as a result of undertaking serial dilutions directly into the coated plate are that air bubbles may have formed in the pipette resulting in inaccurate measurements for proceeding dilutions.
The tips should be changed for each proceeding dilution rather than for each serum as this could lead to contamination. Also, the serial dilutions could have been prepared in test tubes prior to the wells to enable proper mixing of the serum with conjugate, substrate etc. The first incubation, was left without being covered which would also cause error.
*The conclusion describes some aspects of the above answers in more detail.
References
Application of immunological methods 3rd edition (1978)
D.M. Weir MD
Blackwell Scientific Publications
Immunobiology – The Immune System in Health and Disease 4th edition (1999)
Charles A Janeway et al
Published by current Biology publications
Immunology 4th edition (1996)
Roitt et al
Published by Mosby
Instant notes in Immunology (2000)
P. M. Lydyard et al
BIOS Scientific Publishers Ltd.
www.immunochemistry.com/2000services/2100immunoassaydevelopment/ 2123competitive-inhibition.htm