You need to be carefull with using serum when doing ELISA's of conditioned media. The ELISA will pick up not only the analyte produced by your cells but also any analyte that was present in the serum. Good to have a control of using just media+serum without any cells so you know the baseline amount you introduced to the culture in the media.

To your second point about serum starvation. This is the most common technique to look at secreted proteins in conditioned media. While it eliminates interference from serum, you need to do a preliminary experiments to make sure your cells are viable for the conditioned period (24 or 48 hours).

A recent study in Nature Biotech introduced a new technology to look at secreted proteins over the background of serum. In addition to developing a new technique, they also showed that serum starved cells have a distinct secretome compared to full serum cells. Reference below:

1. Eichelbaum, K., Winter, M., Diaz, M. B., Herzig, S. & Krijgsveld, J. Selective enrichment of newly synthesized proteins for quantitative secretome analysis. Nat Biotechnol 30, 984–990 (2012).

Hi, simply spike samples with different amount of serum to evaluate the effect on ELISA.

Regards, H.-W.

Is the ELISA being done in culture medium? I would have thought that ELISA assays would require washing with serum-free buffer followed by other procedures that don't involve serum. The presence of a lot of extraneous protein in any immuno-assay is likely to compromise the assay.

You shouldn't need to starve the cells of serum before the assay. Simply dumping the media and using 'normal' ELISA buffers should be enough. If you are worried, then do a wash after dumping and before the assay. Though I would do a comparison of a pre-washed plate vs. not washed plate.

No effect in MOST cases, follow H-W V''s advice though. Serum WILL affect ELISA in case peptides are involved (as interacting partners or inhibitors). Cheers

a great question I have had myself. i am looking forward to any help!

thank you,

If your test runs indicate that serum causes a problem, try adding 30-50% RIPA buffer to all standards, samples, and controls.

I think you don't need to worry. As commented above, serum "could" affect the ELISA but is not probable. Yo can use the ELISA standards in the same serum media of your cultures, and compare with the ELISA handbook conditions (if the ELISA is commercial). If the ELISA has been developed "in-house" it will depend on the antibodies. You only need to test them

Why don't you do side by side dilution of your unknown protein directly in serum, compared to dilution on PBST. I did this recently to titrate a mouse antibody concentration in human serum or PBST dilutions, and since our secondary was an anti mouse antibody there was no difference between the two serial dilutions. We bought a secondary antibody that DO NOT CROSS REACT with other species (Jackson has several).

You need to be carefull with using serum when doing ELISA's of conditioned media. The ELISA will pick up not only the analyte produced by your cells but also any analyte that was present in the serum. Good to have a control of using just media+serum without any cells so you know the baseline amount you introduced to the culture in the media.

To your second point about serum starvation. This is the most common technique to look at secreted proteins in conditioned media. While it eliminates interference from serum, you need to do a preliminary experiments to make sure your cells are viable for the conditioned period (24 or 48 hours).

A recent study in Nature Biotech introduced a new technology to look at secreted proteins over the background of serum. In addition to developing a new technique, they also showed that serum starved cells have a distinct secretome compared to full serum cells. Reference below:

1. Eichelbaum, K., Winter, M., Diaz, M. B., Herzig, S. & Krijgsveld, J. Selective enrichment of newly synthesized proteins for quantitative secretome analysis. Nat Biotechnol 30, 984–990 (2012).

Dear Cinzia Calzarossa,

It dependents to the type of blocker is being used in your Elisa System and the type of serum you use for your cell culture. If both of them originated from bovine I do not think there is any problem in your experiment if your serum supplied from human source maybe it could show cross reaction with the blocker( due to presence of anti BSA Ab or RF) . if this is the case you should pre-absorb your samples with assay diluent containing PBS+25% serum. By any way you could prepare the solution of (culture media + 25% serum) as a control to check any false positivity in your system.

Good luck

Amrollah Mostafazadeh

Hi Cinzia; Are you using a commercial kit to measure the TGFb1 and 2 in your conditioned media? If possible try a standard curve using your complete media as the diluent. It is possible that serum might not affect high concentrations of your protein of interest but might affect lower concentrations. Compare this curve to your regular standard curve and see if there is any difference. Also you can test an aliquot of your complete media as a background control. Best, Barb

The effects of serum in sample on the ELISA includes:

1. deacrease the sensitivity as a block factor, especially when the protein concentation is higher than your blocking buffer. Adding some detergent may reduce such effects or trying to reduce the concentration of nonspecific protein may help.

2. deacrease the sensitivity by increasing background due to the serum may contain natural antibody against your detection antibody. Try to add some nonspecific immunoglobulin from the same strain animal as your detection antibody. If you use poly clonal antibody, you'd better also add the same serum to your detection antibody diluent to reduce the possible reaction between your detection antibody and the protein in the serum.

In addition, starving your cells may reduce the production of your analytes, which also reduce the sensitivity of your study.

You have to balance these factors.

Good luck

Hi Cinizia:

I have performed several ELISA for various cytokines and serum up to 10% FBS did not make any difference in the result/s. As someone (above) suggests that if you are worried then make your standards in the same medium in which you cultured your cells.

Good luck.

Thanks for all suggestions and all points you point out! They are very precious!

James Ankrum: thanks for the paper I'll read it.

About control, I'm agree it is important, if not mandatory to have a control as baseline, and I always do it, keeping medium (without cell) in the incubator for the same time As samples.

Yongxin Zhang: yes I thimk the same, starving cells the results is Destorted and cytokines detected could be up or down expressed without relation with the study

One point I never thouth: I'm using horse serum and I guess EliSa Kit used Bsa or FBS as blocker ( I did not check yet), I'll check the cross reactivity.

Barbara Jacob: usually I use indiluited and 1 or 2 sample diluitions. I never used complete medium as a diluent, I'll keep your suggestion in mind.

To all, thanks for all interesting poits and suggestions pointed out they help me a lot

Cheers

Cinzia

From an Italian friend: always double-check your writing before sending a message.

In any case, I also used serum (fetal bovine 3 to 10%) without noticing any major change in cytokine production (IL6, IL8, TNFa, TGFb etc.).

If you wish to measure a weak cytokine response to any stimulant(s), please culture the cells for 1-2 days in especially formulated "serum-free" medium, apply your stimulus and measure cytokine release at time 0, 3, 6, 9, 12, 24 hrs in a preliminary experiment. Best wishes, Antonio T.

This is an interesting question - and I note you say that the serum added is horse serum. The answers seem sensible that suggest you measure the TGFb1 and 2 in the basal medium (without cells) then compare the result obtained with cells to that, and attribute any difference to TGFb1 and 2 produced by cells. It is unlikely that the presence of serum in the culture medium would make a difference to the ELISA, especially if the ELISA is based on monoclonal antibodies for capture and detection. If polyclonal antibodies are used, there might be some cross-reactivity between any of the ELISA antibodies and some components in the horse serum. If the antibodies are monoclonal then I guess it depends on whether there is any TGFb1 and/or 2 in the horse serum and whether that is "seen" by your capture or indicator antibodies in the ELISA (i.e. cross-reactive with the pruducts of your cells). What species are your cultured cells? Starving the cells of serum might induce changes in activity that could alter what they secrete so seems best to avoid - but why is it necessary to use horse serum, or any serum. There are a number of good serum-free media that your cells might adapt to, and using that would remove the uncertainty introduced by using serum supplemented cultures.

You need to be carefull with using serum when doing ELISA's as ELISA will pick up any analyte that is present in the serum.

Hi,

As others said, you should have a control with FBS+ media-cells or in treatment step use serum free culture medium to avoide any problems with results!

If your serum/conditioned medium should give you too high background, you might try to deplete the specific growth factors from your serum / conditioned medium before you top it on your cells by passing it slowly through a column with immobilized antibodies against the growth factors. May be homemade easily, just couple 0.5..1mg of (pure) antibody to an activated amino-reactive resin (NHS-agarose etc.). Regenerate with glycine buffer as you would do for antibody affinity purification.

Hi,

Test the serum you use for your media with your ELISA. If you have a signal, you will have to dive in ELISA optimization processes...

Good luck!

HI Cinzia,

the last suggestion is the best.

Three opportunities to disturb the ELISA are present:

1. very high protein concentrations (as they are used in your cell culture) are blocking antigen-antibody interactions in your ELISA. To overcome the item, dilutions are required. Test a couple of serial dilutuions to find the optimal point. The problem is that you dilute not only the protein for your cell culture media, you dilute your analyte as well. But you need a at least any dilution anyhow (1:5 ... 1:10), to get the detergent from the ELISA incubation buffer as the major blocking agent into your wells.

2. In the cell culture media can be your analyte as well. If possible you can use a different species for cell culture media and cells. Now you have to make sure that also your antibody (at least the antibody on the solid phase) is able to recognize the antigen from the cell species only and that this antibody didn't react not with the same protein from the species wich is used for cell culture media. In your case it's could be diifucult. There are only minor differences between species. To overcome this problem, check different culture sera for the presence of TGFb1 anfd b2. It's could be helpful to go for wide species differences (include also bird [chicken] sera.

3. There are inhibitors / ligands in your cell culture media. The respective experimental setup is called "recovery testing". Add you antigen in a known concentration into your cell culture media and other serum free media and check them in different dilutions. You have to find your antigen at the same concentrations after including the dilution factor.

Hi Stephan,

It's true the last suggestion is the best!

thanks for your support!

Cinzia

Hi Cinzia,

I do not think the 25% of serum interfere with primary reaction between antigen and antibody in your system because using 1/2 as a dilution factor (50% human serum) in ELISA systems set up for determining the cytokine levels in human serum is common.

Amrollh Mostafazadeh