Protocol Online logo
Top : New Forum Archives (2009-): : Protein and Proteomics

Immunoprecipitation troubleshooting - (Aug/01/2014 )

I'm doing immunoprecipitation for the first time. For the pilot experiment, my aim was to pull down ABCA1 using protein A/G agarose beads and then probe for phospho forms. I used 2 ug per sample of an antibody that we are familiar with for Westerns and is labeled for IP.

I ran a gel with the IP samples, the lysates that were the source material and the protein A/G agarose beads. After transfer, I stained with Ponceau S and saw the following:


IP samples: no bands

Lysates: some bands, mostly lower than the expected size

Beads: no bands


I started with 250 ug of protein, which I understand may not be enough. Additionally, the protein A/G agarose beads were from 2011 (opened vial, stored at +4). Further, the samples were spun at about 4000 gs instead of the recommended 2000 gs.My question is, which is most likely to be the cause of the apparent failure- insufficient protein to start, old beads, or beads crushed by excess cetrifigal force? Or another issue...Thanks much.  Below is the basic protocol:



  • Prepare samples (frozen tumor lysates): 250 ug total protein in 250 uL lysis buffer with 2x inhibitors
  • Add 2 ug ABCA1 antibody to samples. Incubate at +4 overnight, rotating
  • Next day: Add 40 ul protein A/G agarose beads, incubate at +4 rotating, for 4 hours.
  • Spin at 2000g cold centrifuge 2 minutes**. Save lysates (supernatant) for gel as control.
  • Wash with 500 ul cold PBS 3 x.
  • Add 5-10 ul buffer w/2x inhibitors to beads.  Add 5 ul of 5x reducing sample buffer; mix gently.
  • Boil 10 minutes; place on ice.
  • Spin briefly to precipitate beads.
  • Collect supernatant. Add lysis buffer and loading buffer to beads.
  • Run SDS PAGE gel at 150 volts, 3.5 hr, +4.
  • Wet transfer overnight, 30 v +4.

**Actually g force turned out ot be higher, about 4000, because of the centrifuge used. Some sources say this can damage the beads, but I'm not sure...


High RCF can indeed damage the beads.


The issues here may be multiple. First I would start with making sure that you can detect your protein on a western blot without IP - that way you can be sure that the WB part isn't the issue. If you can't do this, then you can't easily troubleshoot the IP as you don't know if the bands you are seeing are from the IP or not. Non-specific bands are not an indication that your WB is working.


If someone has done this IP before (with the same antibody from the same supplier), make sure the antibody is working properly (i.e. hasn't degraded), then it should work for IP.

Test a range of amounts of AB (if it is monoclonal you often need to add more), and try a bigger volume of lysate, also try higher and lower concentrations of lysate.

Try adding the beads and AB at the same time.

Make sure you wash the beads before you use them, so as to remove the preservative (usually ethanol and NaN3).

Gentle spins for longer often work better than high RCF spins.

You are using a huge volume of beads, try a smaller volume.

Make sure you dribble the wash down the side of the tube rather than pipetting directly onto the beads; the forces from the pipetting can disrupt complexes.


What are the components of the lysis buffer are you using for preparing the samples?


Thanks for your ideas. I am incubating the membrane now and preparing larger samples.


The lysis buffer is Pierce's T-PER, which is a "proprietary detergent in 25mM bicine, 150mM sodium chloride (pH 7.6)."


I'm surprised you thought the beads volume at 40 uL was high - the product sheet (Pierce Protein A/G Agarose #20421) actually recommends 100 ul of the slurry per sample. I'm using this 40 uL volume based on a protocol developed by a former lab member.


Yes, it does say that... but the next lines are:


 "This amount of resin is sufficient to bind 25 - 250μg of antibody. Depending on the amount of antibody needed to immunoprecipitate the desired amount of antigen, scale the amount of resin and suggested wash and elution volumes accordingly"


Note that you are using 2 ug antibody...


The reason I asked about the lysis buffer was, if it contains too much of or too strong a detergent, then it will make IP difficult. If the datasheet for the reagent you are using says that it is OK for IP, then it should be fine.


Yes, good point - thank you. I am trying again and will see what happens.