Purifying monoclonal antibodies (for CUT&RUN or other reasons)

I spent years of my PhD sequencing mouse instead of Drosophila DNA. Now you don't have to!

Fun fact: if you use monoclonal antibodies from the Developmental Studies Hybridoma Bank (DSHB), they're full of mouse nuclei & genomic DNA. This has a nasty consequence for low-input assays such as CUT&RUN, since the mouse DNA will swamp out real signal. This issue is multiplied when using a model organism with a smaller genome size than mouse (like Drosophila), because outcompeting the contaminating DNA becomes almost impossible when using a smaller genome.

This issue isn't present during ChIP-seq because typically, antibodies are first bound to magnetic beads, which are washed & purified away from the dirty supernatant before IP-ing chromatin on the beads. This approach doesn't work for CUT&RUN because it's in-situ, the antibody must be added in solution to the tissue. Hence, we need a different way to purify the antibody before the digest that retains activity while removing contaminating DNA. Of course, there are other reasons you may not want DNA or nuclei in your antibody, this protocol should work for those, too.

Is my antibody contaminated?

Great question. Here are a few easy ways to check:

1. Quantify nucleic acid concentration of your raw antibody using a Qubit.
   • I like the high-sensitivity dsDNA Qubit kit for this since it is very specific for DNA
2. DAPI stain a small volume of antibody & mount on a hemocytometer
   • If your sample is anything like mine, it will look like the night sky, full of in-tact and blebbing mouse nuclei

You can repeat these steps after purification, and if it worked you should read 0ng/uL of DNA, and see 0 nuclei. Yes it's that clean.

How do I fix it?

After trying a lot of things that didn't work, Kami Ahmad suggested I try using magentic proteinG beads to purify the antibody away from the nuclei/DNA crap. It worked great.

The protocol works as follows:

1. Bind the antibody to magnetic beads
2. Wash away the supernatant contaminated with nuclei & DNA
3. Elute the antibody from the beads using an acidic pH (2.5 for proteinG beads, ~3.0-3.3 for proteinA beads)
4. Quickly neutralizing the acidic elution buffer to prevent denaturing the antibody

Honestly the hardest part is making the buffers, and they're really easy to make.

Some thoughts

I let the antibody incubate for 40 minutes on beads, even though the thermo manual for dynabeads says you can do it in 10. This was mostly because I had an older set of beads and I was so tired of things not working by this point in grad school that I didn't mind waiting an extra 1/2 hour.

I do 2 rounds of elution to maximize recovery of antibody & to cut down on the time the antibody sits in elution buffer. I always pooled the 2 eluates and never tested whether the second one was high yield.

Another great test after purification is to perform immunofluorescence using the purified and unpurified antibodies in parallel. If the purification worked correctly, you should observe similar intensity staining using pre- vs post-purification antibody. To correctly account for the difference in antibody volumes, treat the purified antibody as a dilution of the stock. For exmple, if you input 12uL unpurified antibody, after purification you are left with 120uL of purified antibody. That's a 1:10 dilution. So if in a normal immunofluorescence experiment you would use a 1:100 dilution of stock antibody, you would use a 1:10 dilution of the purified antibody. Use this same approach when determining the concentration of purified antibody to use for CUT&RUN.

I use a Tris-Glycine elution strategy, but there are other gentle antibody elution buffers which may be better for certain antibodies. If you use these, you also have to use an ion exchange column (like a Zeba column to swap the buffers because they're full of chaotropic agents (I think, I dunno what's in them, although I tried to find out... see references). This protocol is so easy to do, I'd try the Tris-Glycine approach first, then if the antibody loses activity in immunofluorescence, you can try a different buffer. But first make sure you're not letting the elution sit too long! Try shortening the elution time and see if that helps first.

Good luck!

  Protocol download here

References:

• NEB Antibody Purification w/ pAG beads

• CSH Phosphate Buffer Recipe

• protein G manual

  • BioRad protein G purification manual (This one is GREAT)

• Pierce Gentle ab elution buffer

• Sigma gentle pAG buffers

• Thought: using phosphate-free system could be better because of CUT&RUN in addition to being better for abs
  • Note: This wasn't an issue for me, but could be in some cases.

• DSHB Product Forms

  • Antibodies are delivered in PBS
  • concentrate is supernatant run through 100kda size exclusion filter

• Paper on MNase inhibitors/biochemistry

  • dNdp's inhibit MNase activity (dA/Tp in particular)
  • ie di-phosphate nucleotides w/ 3' P.

• G biosci gentle elution buffer

  • use Tris/HEPES/MOPS ph 7.2-7.4 buffer w/ 150mM salt (no phosphates)
  • Neutralization Buffer: 1M Tris, pH8.0

• Sadly, I dug up an amazing set of 2 comprehensive proteinA and proteinG antibody purification manuals from thermofisher (or maybe pierce) that were of extreme help that I can no longer dig up. They were like 40 pages each with detailed troubleshooting & biochemistry sections. If anyone finds something like this, please reach out. I asked thermo and they couldn't find them either. Maybe I'm crazy.