All tags ELISA An Introduction to ELISA Principles with Troubleshooting Tips Webinar

An Introduction to ELISA Principles with Troubleshooting Tips Webinar

Watch this on-demand presented by our resident ELISA expert

Review Tips and Techniques on how to develop your own ELISA.

Webinar Topics:

  • Why you would use ELISA assays
  • Different types of ELISA assays
  • Key reagents
  • Quantitation using standard curve
  • Optimization and Troubleshooting 

About the Presenter:

Jeremy Kasanov is currently a member of Abcam’s Scientific Support team. He has a PhD in Biology from the University of North Carolina, Chapel Hill and five years postdoctoral experience in Harvard-affiliated labs.

Jeremy’s research focused broadly on signal transduction, where he used a variety of different ELISA assays to assist in the characterization of protein-protein and protein-nucleic acid interactions.

"It's a great introduction for those who are beginning with the technique."

- Attendee

Webinar Transcript:

Hello and welcome to Abcam's webinar on an introduction to ELISA principles and troubleshooting. The principal presenter today will be Jeremy Kasanov, a member of Abcam's scientific support team. Jeremy has a PhD in Biology from the University of North Carolina, Chapel Hill and five years of postdoctoral experience in Harvard-affiliated labs. His research focused broadly on signal transduction. He used a variety of different ELISA assays to assist in the characterization of protein-protein and protein-nucleic acid interactions.​

Joining Jeremy today will be Catherine Tarrade, Abcam's Immunoassays Product Manager. If you have any questions during the presentation, they can be submitted at any time throughout the Q&A panel on the right hand side of your screen. The questions you submit will be answered during the troubleshooting section at the end of the webinar. I will now hand it over to Jeremy who will begin his presentation.

JK:    Thanks for that introduction, Sarah. I'm happy to be able to present this webinar entitled An Introduction to ELISA Principles and Troubleshooting. I'll begin with an introduction to the ELISA application, and give an overview of the most commonly used ELISA formats. Then we'll go on to discuss important reagents and equipment, as well as considerations when deciding whether to develop your own assay, or to purchase a premade kit.

ELISA stands for Enzyme-Linked Immunosorbent Assay, and common uses of the ELISA assay are to quantitate specific analyte from within complex mixtures, and to characterize molecular interactions such as protein-protein and protein-nucleic acid interactions.

There are a number of benefits to using ELISA over other immunoassay formats. A microtiter plate format permits testing of multiple assay conditions simultaneously, and it allows one to quickly optimize testing conditions. One can easily produce quantitative data with multiple replicates, and due to the small well volume, minimal reagents are required. Also, retaining native protein conformation and optimum reaction conditions throughout the assay permit more physiological relevant testing conditions. It's important to note that ELISA does not provide certain information, such as a target's molecular weight or a pattern of distribution within a cell or a tissue. ​

Here is just a brief overview of a standard ELISA. On the left you see a schematic of a traditional 96 well microtiter plate, and on the right we have magnified one well to demonstrate the general steps of the assay. In step one we'll bind the target to the microtiter plate well, and in step two we'll block the well to prevent unintended binding of subsequent assay reagents. In step three we'll incubate the target's specific antibody. Here we show the antibody conjugated to a reporter enzyme. Step four, add enzyme substrate and you can see the well turn blue, representing the colorimetric product one would see in the microtiter plate well. Magnitude of color production is subsequently determined by absorbance readings via a spectrophotometer.

All ELISA assays make use of the same basic principles, here we'll compare five commonly used ELISA formats: the first direct ELISA, which we just discussed, is noted on the left. Moving along, we have the so-called indirect ELISA because the target's specific antibody does not directly produce signal, rather the signal is provided by a secondary antibody bound to that primary antibody. 

The indirect ELISA is more sensitive than the direct ELISA, this is due to the multiple secondary antibodies which will bind to each target's specific antibody, amplifying the signal. 

Sandwich ELISA is so-called because the target is sandwiched between a capture antibody bound to the well, and the detecting antibodies. The sandwiched ELISA assay is considered even more sensitive and specific than an indirect ELISA. The extra degree of specificity is due to the fact that the two different antibodies need to bind to the same target. Sensitivity is enhanced as well, because each capture antibody will bind target molecules, even if the molecule of interest is present there's a very small percentage of the total sample.

A competitive ELISA is commonly used when quantifying small molecules. In a competitive ELISA a labelled competitor molecule, which is otherwise identical to the target, is incubated with the detection antibody and sample. As the ratio of sample molecule to label competitor increases, the signal is reduced because less competitor is able to bind to the detection antibody. The signal on a competitive ELISA actually decreases, as the concentration of the target molecule in your sample increases.

Lastly, the In Cell ELISA is a quantitative immunoctyochemistry-based method to measure protein levels, or post-translational modifications of cultured adherent cells. For your convenience, I have also compared these ELISA assays based on relative sensitivity, specificity, assay time and costs to set it up. More details on these ELISA assays can be found in the protocol section of our website.

A major decision that you'll face after you've decided that ELISA assay is what you need, will be whether to develop the ELISA assay yourself, or to go ahead and purchase a premade kit from a supplier. So I just wanted to walk you through considerations to help you make your decision. One thing to consider would be the expense, if you combine the cost of all the reagents you'll need to buy, to purchase, to put together your own ELISA, it will likely be substantially more than if you were to buy a single kit from a supplier. However, if you imagine using this assay for an extended amount of time, and think this assay will become a staple for your lab, then it's probably worthwhile to develop your own assay, because in the long run you will pay much less than if you were to purchase the equivalent amount of kits from a supplier. 

Time is another important factor. Time is money and it is much more efficient to simply purchase a kit from a supplier, and receive it the next day. This kit will be guaranteed to work right out of the box, will include a precise protocol and most, if not all, necessary reagents. Developing your own assay can take weeks or even months to perfect.

Frequency of use relates to expense. As I mentioned, if you are using this ELISA assay frequently, definitely consider developing an in-house assay. Then, again, your time is critical or if money is plentiful, then you should consider purchasing the off-the-shelf kits.

Technical expertise is another important consideration, if you really do not feel confident in your ability to put together an ELISA assay from scratch, then regardless of expense I would encourage you to search for premade kits. Of course, your ability to develop an in-house ELISA assay will absolutely depend on the availability of the necessary reagents. If even one critical reagent is not readily available, then it would probably be necessary to look at the purchasing of a premade kit. Similarly, the option to purchase a premade kit will depend on the existence of this kit. If the specific ELISA you're interested in is not yet available as a premade kit, then you are limited to developing one in-house.

Let's assume after these considerations, you decided to develop your own ELISA. In the following section we'll discuss developing your own ELISA in more depth. Four important aspects to keep in mind when developing an ELISA assay are to ensure the assay is specific, sensitive, accurate and reproducible.

This slide shows some of the key materials needed for making your own ELISA assay. In brackets I've noted equipment that, while not absolutely required, will certainly make running ELISA assays much more efficient. Single and multichannel pipettes are required, but it is certainly nice to have digital repeat pipettes available for doing any more than one or two plates at a time. You'll need microtiter plates that have a high protein binding capacity, and I definitely recommend sticking to 96-well plates without robotics. It is extremely difficult to work with 384-well plates, for example, without automation. You'll need at least a squirt bottle for washing the plates, although for washing more than a few plates it is really helpful to have an automated plate washer. Plate washers will also speed your assay and increase reproducibility. Of course, you'll need blocking buffer, protein binding buffer and enzyme substrate if you're using an enzyme conjugated secondary antibody. It's important to cover your plates to prevent evaporation during incubations. Plastic wrap can be used, but it may be more convenient to use plastic covers or adhesive plastic seals. Of course, you'll need antibodies that ideally have been tested to function in ELISA. Lastly, you'll need sample and a spectrophotometer to read your data. 

When binding sample or target to the well, it's important to consider the type of molecule. Typically, large proteins or antibodies are bound to a well, and these types of molecules will absorb to the polystyrene well and retain their ability to be detected by the antibody. Other types of molecules, such as heavily glycosylated proteins, carbohydrates, short peptides, lipids and DNA will need to be modified by the addition of a covalently-linked affinity reagent, such as biotin or a carrier protein before being bound to the well to retain immunogenicity. 

Of course, antibodies are a key reagent when developing an ELISA assay. One thing to consider is if the antibody has been previously tested to work in an ELISA format. For example, Abcam will list antibodies as tested via ELISA, or sELISA, or sandwich ELISA. ELISA refers to the use of the antibody as a detection reagent, and sandwich ELISA indicates the antibody has been used as a capture antibody in a sandwich ELISA assay.

Clonality of the antibody and type of immunogen used to produce the antibody, are also important when considering which antibodies to use for your ELISA. Monoclonal antibodies are known for the extremely high specificity, and polyclonal antibodies promote increased sensitivity, because there will be more than one epitope on each target that can be bound by a polyclonal antibody product. 

I think it's useful to briefly review the significance of clonality and type of immunogen used to produce common antibody products available for immunoassays. Two typical types of immunogens are either a short peptide or a large recombinant protein fragment, or a full-length protein. Short peptides exhibit a limited range of potential epitopes that an antibody may react with. Of course, a monoclonal antibody product will only react with a single epitope, no matter how large the immunogen. But even a polyclonal antibody produced using a peptide immunogen will have a very limited range of epitopes. The benefit of using a peptide as an immunogen is that the epitope will be narrowed down to a very small region, even for the polyclonals. This is useful, for example, when you wish to distinguish alternatively spliced protein products, or to detect a specific post-translational modification.

A downside to using a peptide immunogen is that the epitope may not be accessible when the full-length protein is in its native, properly folded state. This is particularly important to keep in mind if the antibody has not yet been tested in immunoassay applications where detection of native proteins is crucial, such as ELISA or immunoprecipitation. 

Now, let's consider antibody products produced using a large recombinant protein, or full-length immunogen. In this case, monoclonals will still only recognize a single epitope and testing for activity to full-length native proteins is still required. On the other hand, polyclonal antibodies raised against large protein fragments, or full-length proteins recognize the many different epitopes, so one doesn't need to be as concerned with a lack of reactivity to folded proteins. Furthermore, a polyclonal product will allow more antibodies to bind the target, leading to an ELISA with increased sensitivity.

To graphically demonstrate what we have just discussed, here you can see that two monoclonals will maximize specificity. However, you must be sure binding of the capture antibody does not block binding of the detection antibody. To maximize sensitivity and specificity, one common sandwich ELISA strategy uses a monoclonal and a polyclonal raised against a full-length protein. The monoclonal lends specificity and the polyclonal enhances sensitivity. To reduce false positives in sandwich ELISA, you need to ensure the secondary, this is the antibody conjugated to the enzyme, or a fluorescent molecule is not directed against the capture antibody. For example, if your capture antibody is a mouse antibody, you should use a secondary antibody that will not recognize mouse immunoglobulins. Typically, in a sandwich ELISA this is ensured using a capture and detection antibody produced in different host species.

Antibody purity is another important factor, especially when considering capture antibodies for sandwich ELISA assays. Unpurified antibody, such as that found in whole antiserum, tissue culture supernatant, or ascites fluid, will contain other proteins in addition to the target-specific antibody. Purified antibody will, of course, only contain the target-specific antibody and will bind more target, increasing sensitivity.

Buffers are another important consideration in developing your own ELISA assay. Let's consider the coating buffer first. The coating buffer refers to the buffer that you dilute your target in that will be bound to the surface of the well. It is important that this buffer does not contain extraneous proteins, and as well it must be compatible with the sample. This means that the sample must retain any key biochemical features, such as solubility, activity and folding. The blocking buffer should give the lowest background, not interfere or be reactive with any of the assay components, and should not possess any enzymatic activity that may lead to a false positive result.

Here I have listed a few of the most common ELISA reporter enzymes and their substrates. Alkaline phosphatase and horseradish peroxidase are the most commonly used reporter enzymes, and, as such, there is a variety of substrates developed for these enzymes. The substrates differ in type of signal, sensitivity and cost. The three main types of substrates are chromogenic, which produce a colored substrate, whose absorbance is read with the spectrophotometer, fluorescent substrates and chemiluminescent substrates. As you can see, relative limited detection vary greatly from 100 ng per ml for the alkaline phosphatase substrate pNPP, to 25 fg per ml, the purported sensitivity, of a LumiGLO HRP chemiluminescent substrate. An important consideration here is manner of detection. You need the correct equipment to detect a chromogenic fluorescent or chemiluminescent signals. Sensitivity is another important factor. It's not always better to get the most sensitive substrate, because this could potentially lead to high background, or development of the signal which occurs too quickly.

There are two types of data one can derive from an ELISA experiment: one is qualitative data. Here are two examples of qualitative data. On the left, example one, can represent an experiment where a sample is extracted from cells, treated with a range of different drugs that have increased protein expression in some samples. Or it could represent phage display peptides of different sequences, and their relative affinity to a protein target. Example two might illustrate increasing concentrations of a drug treatment that lead to increasing protein expression levels. 

Quantitative data can also be derived from an ELISA assay. Here's an example of a standard curve derived from one of our IL-6 ELISA kits. I will use this example to demonstrate how one can quantify the amount of target protein present in a sample, based on a comparison to the standard curve. So, for example, let's say an unknown sample gives a reading of 0.7 OD units. You can use your standard curve to estimate a quantity of IL-6 in the unknown sample, or plug the values into the line equation to get more precise data. So based on this example, you can see how easy it is to quantify unknown samples using your standard curve.

Now that we've talked about the important aspects of developing your own ELISA assay, let's talk a little bit about optimization and troubleshooting. Here's an example of where the amount of capture antibody loaded onto the well is optimized to reduce waste. Four different dilutions of the antibody reused, and, as you can see, a 100 full dilution of the antibody gives a higher sensitivity with the least amount of antibody use. Here's another example where reaction buffer component was optimized. Here the intent was to use the maximum amount of sodium chloride that will retain the maximum signal. Based on the data here, we were able to determine the 300 mM salt concentration was ideal.

As you may have realized, there are many variables when setting up an ELISA assay. I feel this makes an ELISA assay one of the more difficult immunoassay applications to troubleshoot. Here are four important troubleshooting points to keep in mind during assay development, and the optimization that will inevitably follow. Matrix effects are thought to be due to interfering substances that inhibit intended antibody-antigen interactions. These interfering substances can be other antibodies, proteins or small molecules, for example. Matrix effects are particularly prone to occur when working with plasma samples. A good way to determine if matrix effects are in play, is to spike positive control antigen into your sample. Matrix effects may be present if the signal of these controls are lower than expected. 

False positive results give the appearance of real signal, this can be due to non-specific or unintended antibody interactions, or a buffer or sample components that artificially enhance the signal. For example, labelling, competitor compounds for a competitive ELISA may reduce its affinity to the detecting antibody, leading to artificially inflated readings. 

False negatives produce artificially low or absent signals. The hook effect is a common example of this phenomenon. The hook effect can occur when target is present in much higher numbers than the detecting antibody. This leads to a situation in which increasing analyte actually leads to an apparent decrease in signal.

Lastly, it is critical to include positive and negative controls on every plate. If pure target protein is available, it can be used as a positive control and to produce a standard curve. 

Here is a specific troubleshooting example which highlights the importance of consistent data, and the utility of replicates in an ELISA assay. Here is an example outlined in gold of the standard curve. The blue arrow in the graph on the left shows the standard curve, and based on the standard curve the concentration of two unknown samples were determined.

The circled data point is highlighted, because it is unexpectedly high. One potential solution is to remove this set of data points from the standard curve. In this case, we can do that because we have so many other consistent data points. As you can see, removing the one erroneous data point drastically changes the shape of the standard curve, and also improves the data. Comparing the percent error between the two experiments, you can see that the percent error is reduced dramatically from 56% in the top dataset, to 10%. By tightening up the data you can see the concentration of the unknown samples has also changed. For sample one, the concentration changed from 66 ng per ml to 127 ng per ml after correction. For sample two, the concentration changed from 52 ng per ml to 100 ng per ml after correction. I hope this example highlights the importance of careful data analysis, as well as the utility of the ELISA assay.

Lastly, I just have a few ELISA-related resources that I thought you might find useful. Here is a schematic of the typical 96 well microtiter plate format. A grid pattern like this will really help you map out your ELISA experiments, and it is excellent for record-keeping as well.

It's also important to keep track of every reagent you use in your ELISA assay. It is important to keep track of not only the catalogue numbers, but also the lot numbers. This is an example of one way you might keep track of the reagents used in a capture ELISA assay.

Abcam has an extensive protocol library regarding a large range of immunoassay techniques. To browse these protocols hover your mouse over the scientific support link and select protocols and troubleshooting. Here you'll be able to browse protocols by application, or by research area. For example, we currently list 12 ELISA protocols which can be viewed on our website or printed as PDF files. Now I'd like to introduce my colleague Catherine who will highlight our immunoassay catalogue.

CT:    Thanks, Jeremy. So I would like to take a couple of minutes to introduce the Abcam immunoassay catalogue that includes over 1,000 optimized ELISA kits. Kits that can easily help you track your favorite targets. Abcam is offering a comprehensive range of ELISA, In Cell ELISA kits and dipstick assays. Our conventional ELISA kits cover over 500 targets and are available in a wide range of formats: sandwich, indirect, competitive. Our kits are also available in different detection methods: many colorimetric, and also purinergenic and infrared.

Looking at ELISAs, let me introduce a brand new phospho-ELISA kit called PhosphoTracer. PhosphoTracer ELISA kits enable you to detect protein phosphorylation more sensitively than western blotting. The protocol is unlike any other, as both the analyte and the assay reagents are added to the microplate at the same time. In just over an hour you can measure, for example, phosphorylated Akt, JAK-STAT proteins.

For the prediction of phosphoproteins we also have In Cell ELISA kits that measure the relative amounts of protein in cultured cell lines. They are powerful tools for assisting the effect of various treatments in one experiment. Finally, another interesting assay would be the dipstick ELISA. They use a well-established lateral flow concept whereby capture antibodies are striped onto nitrocellulose membrane, in a Whatman paper pad, and draw the sample through the antibody bands. The sample is added to the well and then the dipstick is inserted. Within minutes the line of each target is reviewed and the protein detector, on typically gold complex, binds with the capture antibodies.

As mentioned earlier, we are offering a wide-range of protein targets for each ELISA format. ELISA kits are mainly detecting cytokines, cytokine receptor, growth factors, cardiovascular proteins, cancer proteins or phosphoproteins. In Cell ELISAs are suitable for signaling pathways: hypoxia, inhibitors and activators of mitochondrial biogenesis. Whereas dipsticks can track Complex I protein, pyruvate proteinase or enzymes involved in fatty acid oxidation. 

Before finishing this section, I wanted to let you know we are currently offering 20% off the least price of our entire ELISA portfolio. At the end of our webinar you will be redirected to our webpage where information on this offer, plus a PDF of the presentation would be available for viewing and download. I will now hand you over to Sarah who will quickly mention some upcoming webinars and events, which you may be interested in attending. Thank you for your attention.

SD:    Thank you very much Jeremy and Catherine. I would just like to quickly take the opportunity to highlight a few additional Abcam events coming up in the next few months. On May 29th, Abcam will be hosting another webinar on post-translational modification. Please visit the displayed link for further presenter details and topic overviews. Additionally, on June 21st please join us for In Cell ELISA webinar. Registration for this webinar is available six weeks prior to the webinar.

JK: Thanks again, Sarah. During the talk there is a few questions that were submitted, so during the time remaining I'd just like to go over a few of the questions, and answer them as well. One question: I realize the 384-well plates are difficult to work with, however, is it still possible to design an ELISA in a 384-well plate? That's a good question, and, yes, you could definitely use 384-well plates to set up ELISAs in. This is frequently done, as a matter of fact. I only mentioned it would be more difficult, because the wells are so small, it's often difficult to keep track of which wells have been added to which reagents. So if you're going to do 384-well plates for an ELISA in-house, the key thing is to be careful when you're allocating to the wells to keep each addition straight. One thing you can consider doing is to map out each reagent that will go into each well, and check each one off as you've added it well-by-well. This will definitely keep mistakes to a minimum, because if you do make a mistake it's often difficult to go back. Essentially you’ll usually just have to start over because you won't be able to determine what's been added to what well. So definitely it is possible and just take it slow and be careful.

Another question: Can a denatured protein be added to the well? In the talk I really focused on the ELISA assay being useful, because native proteins could be used in their native conformations, and in physiologically relevant conditions. But definitely depending on your needs, you can bind denatured proteins to the well also. You can use whatever means you need to denature your target, and then just add it to the well as you would any other native target.

Another question: Can you use fluorescently-labelled secondary antibodies in an ELISA assay? Well, yes, you certainly can use fluorescently-labelled secondary antibodies as the means to detect protein target in your ELISA. You'll just have to make sure that you have a device that is capable of measuring the fluorescent signal from the secondary antibodies. Just as you would need to have a device that will measure chemiluminescent signal, or a luminescent signal, if you choose to use a substrate like that.

Another question: What are the correct controls to use for an ELISA assay? I think a good negative control for a common type of ELISA assay is to just not include the target protein on the well. So, in this case, what you just do is, for example, load the blocking buffer onto the well and leave out the target protein. So in this way it gives you an idea of how well the blocking buffer works, and in addition you get to see what the background signal will be without the specific target of interest. A typical positive control might be a purified protein target. This would allow you to play a dilution series, for example, of a control protein which you could use for both a standard curve, as well as to ensure that the assay is working as expected.

Another question we received: I have a high background, what can I do to reduce it? There's a number of things that you can do to try to reduce a high background, and get the best signal to noise ratio that you can. I think one of the simplest things is just to work on optimizing the dilutions of the antibodies in your assay. This could be optimizing the capture antibody concentration, if you're using a sandwich ELISA, for example, or optimizing the detection antibody, or the secondary antibody; that is the one with the reporter molecule on it. Using too much reporter molecule can easily increase the background in your assay. 

Another really useful way to lower background, I'd say is to increase a number of washes. I think typically I would recommend people use at least five washes each time after the addition of each reagent. What you can also do is increase the severity of the wash buffer as well, and make it more stringent. 

Another really important way to decrease the background is to actually change your blocking buffer. There's a lot of blocking buffer formulations that you can try, and there's no one right blocking buffer. So you can try standard ones, for example, like 5% milk blocking buffer, you can try 1% BSA-containing buffer, serum-containing buffers and there's other proprietary buffers sold by different companies. You might not know what's in them, but they sell them as being very useful as well, so that's definitely blocking buffer can be a key reagent to change for decreasing background.

Here's another question: I'm not getting any color change after adding the enzyme substrate. The antibodies are conjugated to HRP, what could be the cause of this? There could be a lot of different reasons why you're not getting any signal, and that's, like I mentioned in the talk, it's tricky often at times, to troubleshoot an ELISA assay because there's so many different potential pitfalls, because there's so many different reagents involved. So, for one, it could be an antibody issue, one or more of the antibodies used in the ELISA assay may just not be working as expected. It may not have reactivity to the protein target. It may be that an epitope for that antibody is just not available in the target protein. It's also important when doing sandwich ELISAs, for example, to test the antibodies that you're going to use as a capture antibody, as actually a capture antibody. Because sometimes an antibody may work well as a detecting antibody, but may not work so well as a capture antibody. So often times what people can do is even collect a few different antibodies, and try them in different orientations and in different combinations to determine what gives you the best signal to noise. Sometimes it's difficult to predict, and this is often the best way to go about doing it. It could be a sample issue as well; perhaps the intended target of the antibodies is just not present in detectable levels, based on your ELISA setup.

So if this is a problem that you're also having with the positive control, then that's another issue. So if it's a positive control that you've received in a kit that you purchased from a supplier, it's really important to make sure that the dilutions are done as recommended in the protocol, and that the control protein has been stored and diluted as recommended. For example, if you'd accidentally diluted the standard too much, then it will not appear as you expected. In this case, since the antibody is conjugated to HRP, it could be something as simple as the HRP being inactive on that detecting on the secondary antibody. This is an easy thing to test just by adding a µl of the antibody into your HRP substrate; it should turn color immediately.

Another question here: I have purchased an ELISA kit and it appears much less sensitive than advertised, what should I do? Well, definitely if the kit contains a positive control or a standard, I would definitely start the troubleshooting with this reagent. Don't waste any of your valuable sample trying to test the kit any further, if the positive control and a standard doesn't work as intended, then you have an issue and it would probably be advisable to contact the supplier. But, again, you want to make sure that the standard is stored properly and diluted as recommended. If the standard curve works well, then it could be an issue with the sample again, so you'd want to make sure that you're using a buffer that's compatible with the sample type as well.

Here's a question about a reference wavelength, it says: Sometimes our reference wavelength is noted as a control in the protocol, so what is the reference wavelength for? That's a good question. The reference wavelength is often noted as an internal control for consistency of the readings by the spectrophotometer. So, in theory, the idea would be that you'd read the reference wavelength at the same time that you're reading the absorbance wavelength of your substrate, and then you would simply just subtract a reference wavelength value from the absorbance readings of your substrate, in a way, normalizing all the values. Typically, if your spectrophotometer is working well, as they usually do, these values for the reference wavelengths are very small compared to the absorbance readings, and so it really doesn't change the data. So often you'll see that the reference wavelength calculation is labelled as optional in a protocol.

Here's another question regarding microtiter plates: What kind of microtiter plate should I use for my ELISA? So, as I recommended in my talk, I recommend using high-binding plates, as they're called, which would allow the maximum binding capacity of protein to your plate. I only recommend that because it stands to reason that this would maximize the sensitivity of your assay, although there may be reasons to use a low-binding plate as well. These plates also come with different shaped bottoms, you can get round bottom, flat bottom, V-bottom ELISA plates, and you'd have to choose that based on your considerations there for your experiment. I've typically used flat bottom plates as they're most convenient. For another aspect of the ELISA plates that might be important is the color, and typically for, well, definitely for colorimetric substrates or chromogenic substrates where they turn blue or yellow or green, or what have you, you're using a clear polystyrene plate to get the absorbance reading there. However, if you're going to use luminescent or fluorescent signals, then you'd need to use the required opaque, white or black plates to maximize the sensitivity of using a substrate of that kind.

Another question that was asked was: How do I increase the sensitivity of the ELISAs? When developing an ELISA assay you need to increase its sensitivity from a µg to ng per ml levels. So that's a really good question, and a frequent question that we get. So one easy way, if you have the means to read the signal, is to change from a chromogenic substrate to a chemiluminescent or a luminescent reporter system. This change in itself will increase the sensitivity of your assay about tenfold. As discussed earlier in the talk, the use of polyclonal antibodies produced against full-length proteins, for example, will allow more antibodies to bind each target molecule. So then the secondary antibody will have more antibodies to bind to, and this will increase the sensitivity of the assay as compared to using a monoclonal antibody, for example, as you're detecting the antibody. Also, of course, you'd want to maximize the amount of target that you've bound to your plate, and you might need to try a different ELISA method such as the indirect ELISA, or a sandwich ELISA approach to maximize the amount of target bound to the well.

Here's another question: Can I strip antibodies or the associated proteins from the ELISA plate wells? I think the researcher here would like to maybe reuse the capture antibody, for example, and remove anything bound to the capture antibody like you might, say, strip a western blot. I would say - without knowing more, I'd say that this is probably not recommended. The ELISA assay is so convenient to do additional assays with as the wells are quite small, the amount of reagents needed per well is quite miniscule. So I think it's worthwhile to run just extra wells, than to try to strip off protein or antibody from one well and try to use it in another experiment. 

Here's another question: What is the best wash buffer to use for ELISA? I'd say technically the best answer is whatever wash buffer gives you the best signal to noise ratio, gives you the lowest background. But, having said that, a couple of common ones are a Tris buffered saline, or a phosphate buffered saline with small amounts of Tween, like 0.5 or 0.1% Tween. It's important to remember with your wash buffers and your protein binding buffers, and your blocking buffers too is that you need to make sure that your target protein is always happy. So if there's a co-factor that's required for your assay, or for your target proteins to be in its native state, or have the activity that you require, then you'd have to make sure that these co-factors are present in all the buffers that touch the target sample. 

Here's another question: I can detect endogenous protein from my cell extracts, but using even high concentrations of the fusion protein I made is not detected at all, what's going on? This is another - it's a good question, and there could be more than one reason why you can detect endogenous protein, essentially what's probably your control protein, the fusion protein that you made, that's the one that's supposed to work right off. But one thing may be that you should ensure that the epitope's recognized by all the antibodies in your ELISA assay, are all going to be present on your fusion protein. So, for example, if your fusion protein is not a full-length protein, this kind of issue could come up, because the epitope for any one of the antibodies may be missing from your fusion protein; then, of course, you'll get no signal. Another thing to consider about fusion proteins is that commonly mammalian or eukaryotic proteins, when produced in a bacterial system, which they commonly are, may not fold properly. So if proteins are not folded properly, the epitopes that may be exposed on the surface to the antibodies in your native endogenous sample, might not actually be available for binding to those same antibodies from your fusion protein. One other thing that I can think of right now is fusion proteins, for example, a GST fusion protein, and a GST is 26 kDA is rather large, and a fusion protein like this can obscure one or more epitopes on the fusion protein partner. So if that's the case, then that epitope again would be blocked, not allowing the antibody to bind.

It looks like we have time for one more question. Any advice on setting up a standard curve? Well, when I setup a standard curve, the first time anyway, it's really useful to setup a really broad standard curve going from loading protein that will overload the plate, for example if you have a pure protein, and you load at least one microgram into the plate, and then doing twofold dilutions, for example. If you do ten, eleven twofold dilutions, preferably in duplicate, at least, to show consistency of the data, this will give you a really good idea of where a linear range is in your detection system. So finding this linear range then as well, finding the signal range which your spectrophotometer can read accurately, is important for these types of standard curves. So then once you've found your data, you can go ahead and focus future curves just on that linear range that can be read well with your spectrophotometer.

I'm sorry if I didn't get to your questions here, but to ensure that all questions are answered we'll follow-up the email with everyone who has asked a question during the webinar. If you have any further questions, please feel free to contact us at any of our offices worldwide. Please note that we have multilingual customer service and scientific support representatives waiting to help you.

Ladies and gentlemen, that does conclude today's webinar. If you have any questions about ELISA or have a scientific enquiry, our scientific support team will be very happy to help you and can be contacted at If you have any webinar or event-related questions, please contact the Abcam events team at I hope we can welcome you for another webinar in the future. Thank you once again for attending today, and good luck with your research.

The Abcam events team would also like to invite you to join us in Boston on June 25th and 26th for Mitochondrial Trafficking and Function in Neuronal Health and Disease. 
The last event I would like to highlight is Origins of Tissue Stem Cells taking place on June 28th and June 29th in Edinburgh in the UK. The early bird deadline is quickly approaching and will be available until May 28th. 
I would now like to turn the microphone back to our presenter Jeremy, to answer the questions that you have submitted. Any questions not answered during this portion will be answered by our scientific support team following the webinar.