Hello Kateryna Bilotkach

I'm sorry to hear you are having difficulty with your IP-MS experiment. Let's walk through what you posted above.

You wrote:

"...to get rid of antibodies in the eluate, I crosslinked antibodies to the A/G magnetic beads (Thermo)..."

This is a very sensible thing to do and should mean that your final eluate has few (if any) contaminating antibodies. I would recommend continuing with this approach.

Next, you wrote:

"...when I compared eluate obtained from the beads with crosslinked antibodies to that from beads with pre-adsorbed antibodies (no crosslinking) it looks like I can only elute protein complexes from the beads with pre-adsorbed antibodies..."

This is not good and indicates a problem. Are all the conditions (sample binding, sample washing, sample elution) identical between the 2 bead conditions (bead condition 1 being crosslinked antibodies & bead condition 2 being pre-adsorbed antibodies)? Have you also performed a control to see what eluate you achieve with both bead conditions without loading a sample? The "background" proteins associated with the beads might be the source of the protein signal.

Next, you wrote:

"...my input (cell lysate I used with beads crosslinked with antibodies) seems not to have protein B. Usually protein B is identifiable on WB, although it is known it is a low abundance protein. Protein B is present in cytoplasm and nucleus. The MS samples prep protocol uses special detergent to lyse cells. I do not know how efficient this detergent is to lyse cells nuclei. I wish I could explain absence of protein B in cell lysate by its low quantity – it was not enough to visualise..."

This is not uncommon for low abundance proteins. Which detergents are you using to lyse the cells? I would recommend RIPA and sonication to lyse the cells, then perform the IP. You could also check the quality of the lysis by running a SDS-PAGE gel to see if you are efficiently releasing histones from the nucleus.

Going forward: once you have your IP eluate, run it 1 cm into a SDS-PAGE gel. Cut the gel lane into three equal vertical slices (not horizontal like you would for gel band identification). You can then perform in-gel trypsin digest of each equal band and pool all the peptides after digestion, ready for mass spectrometry analysis.

Good luck.

Dear Peter Allen Faull ,

Thank you very much for you prompt response.

Answering your questions:

Beads prep was carried out simultaneously under the same conditions except that beads with crosslinked Abs were washed more and low pH elution buffer was used to remove unbound antibodies. Then, both beads preps (crosslinked and pre-adsorbed) were stored in "IP Lysis/wash buffer" for ~1h at 4oC.

Reagents used: Pierce Crosslink Magnetic IP/Co-IP Kit, 88805; Pierce™ Protein A/G Magnetic Beads, 88802

Sample binding conditions - were identical (followed the protocol from Pierce™ MS-Compatible Magnetic IP Kit, streptavidin, 90408; Beads were not! streptavidin coated and were Pierce™ Protein A/G Magnetic Beads, 88802).

Cell lysates were different - they were prepared in different tubes, 1h apart. From the same cell culture. The one I used for crosslinked beads was the freshest one. Of course, it is probable that protein B fall off from the complex with protein A specifically in the second lysate, but I guess this is very unlikely.

I did not do control without loading a sample, yet would not IgG control be sufficient to answer this question? There is no protein B band in IgG control for pre-adsorbed beads.

The detergent to lyse the cells for MS - Pierce does not disclose its recipe. It is a very good idea to check this cell lysate for the presence of histones. Thank you! Most probably I just have a cytoplasmic fraction.

Since the protein yield is low and scaling up with 90408 kit will cost us a fortune I was also thinking to switch to a "classical" IP protocol with RIPA lysis buffer. I am trying to avoid in gel eluate purification/digestion (I have heard that this brings keratin contamination and decreases the protein yeild). Ideally I would like to dry the eluate in a vacuum evaporator. This step is recommenced in 90408 kit. The only problem with using RIPA lysis buffer and in house prepared wash/elution buffers is carrying over the detergents that make MS analysis difficult. Do you know if it is possible to wash the beads so that to eliminate NP40, SDS and other detergents used in lysis and washing buffers?

Thank you again for your detailed answers

Dear Markus Hartl ,

thank you for your answer.

Yes, definitely it is possible I have less “working“ antibodies on the beads. I am guessing, what if a macromolecular complexes just do not fit into a binding “cleft “ on the crosslinked antibody.. or the binding surface on crosslinked antibody is significantly reduced so that the complexes fall off (weak binding). Unfortunately, this will be difficult to test for me and this is not the aim of my experiment. Reduction in antibody binding capacity and selectivity would affect the composition of the pull down and potentially is a huge problem here.

I now reprobing the membrane with antibodies, hoping to increase the sensitivity of the blot. Let’s see if I can see anything at all.

thank you very much for your input!

There are many cross-linkers to choose from, and even if they have comparable target reactive groups, their kinetics vary widely. Have you titrated the DSS, monitoring how well it prevents IgG leakage for a given period of incubation? Although many such protocols are "standard" as Markus mentioned, antibodies themselves are not "standard" and so, some variation and case-to-case optimization is required.

The number of different issues you specify are too many to address in a short post, and may not all be connected.

That being said... one should identify the shortest incubation at the lowest concentration that would reduce or prevent IgG leakage, while leaving the paratope potency intact. This can be done relatively straight forward in 1-2 days of quick experiments.