I am unfamiliar with ULK1. Is it a membrane protein? If so, these tend to be very hard to isolate by standard IP protocols.

What is the host organism of your primary antibody? Some will not bond to Protein G, and instead require Protein A. 

Is your primary antibody described by the manufacturer as being suitable for IP? Some just will not work for IP, even though they may work great for Western blotting or immunofluorescence or ELISA.

Depending on the concentration of the antibody, you can also use 50 ul of Dynabeads with that same volume of antibody. However, it is important that you both have enough beads to recover ample protein, yet have antibody be saturating while bonding to Protein G/A, such that there are no antibody-less beads present when adding cell lysate (though nonspecific binding to dynabeads is pretty low). Or, you could try adding the antibody directly to cell lystae, and then incubating with the beads, in which case, you want your protein of interest to be saturating for the antibody, as free antibody will bind Dynabeads more readily than antibody count to protein. 

If you have a good protein-null negative control, I would drastically increase incubation time of the antibody-dynabead complexes with lysate, and then scale back to the shortest time necessary. For some proteins, 1 hour just won't work. So I'd first try overnight, and see if that improves recovery (without considerable background). If it does, then try something like 8 hours, then 4, then 2, etc., to find the shortest time necessary to get good recovery. Overnight may very well be the shortest time required, however. 

Also, I've never used PBST (even at 0.05% Tween-20) to wash beads. I usually just wash with lysis buffer (unless that's what you lyse in). 0.05% is low, but could the Tween-20 be too strong a detergent for washing? I've always used 0.1% NP-40 in my lysis buffers. I believe NP-40 is not as strong as Tween-20, but I could be wrong. 

Good luck!

I am unfamiliar with ULK1. Is it a membrane protein? If so, these tend to be very hard to isolate by standard IP protocols.

What is the host organism of your primary antibody? Some will not bond to Protein G, and instead require Protein A. 

Is your primary antibody described by the manufacturer as being suitable for IP? Some just will not work for IP, even though they may work great for Western blotting or immunofluorescence or ELISA.

Depending on the concentration of the antibody, you can also use 50 ul of Dynabeads with that same volume of antibody. However, it is important that you both have enough beads to recover ample protein, yet have antibody be saturating while bonding to Protein G/A, such that there are no antibody-less beads present when adding cell lysate (though nonspecific binding to dynabeads is pretty low). Or, you could try adding the antibody directly to cell lystae, and then incubating with the beads, in which case, you want your protein of interest to be saturating for the antibody, as free antibody will bind Dynabeads more readily than antibody count to protein. 

If you have a good protein-null negative control, I would drastically increase incubation time of the antibody-dynabead complexes with lysate, and then scale back to the shortest time necessary. For some proteins, 1 hour just won't work. So I'd first try overnight, and see if that improves recovery (without considerable background). If it does, then try something like 8 hours, then 4, then 2, etc., to find the shortest time necessary to get good recovery. Overnight may very well be the shortest time required, however. 

Also, I've never used PBST (even at 0.05% Tween-20) to wash beads. I usually just wash with lysis buffer (unless that's what you lyse in). 0.05% is low, but could the Tween-20 be too strong a detergent for washing? I've always used 0.1% NP-40 in my lysis buffers. I believe NP-40 is not as strong as Tween-20, but I could be wrong. 

Good luck!

Ryan has already provided a lot of really good suggestions. I just have a couple things that you could add to figure out what your problem is.

With regard to the lysis buffer, the detergent you are using matters. I am also unfamiliar with ULK1. If your protein is a membrane protein, I would suggest you try a couple of buffers to find the one that is strong enough to isolate your protein of interest while leaving the nucleus intact. Try a mild RIPA as well as a buffer containing 0.1% NP-40.

Another problem could be solubility. It's a really good idea to meticulously check every step of your protocol so you can better troubleshoot. This is known as book-keeping. Take a small volume from each step and run it on a gel and blot for your protein (ULK1).

Examples of fractions you should collect: Lysis soluble (the fraction you are applying to the beads), Lysis insoluble (the pellet), unbound (after binding, the supernatant before washing), and each wash step. You would determine the volume based upon how much sample you have. I usually take out enough to run 2 gels, to be cautious. It is common to use lysis buffer as your washing buffer as well. Using something else could run the risk of precipitating proteins.

I also agree that 1 hour is probably not long enough. Overnight is a good place to start. You will still see your protein in the unbound, even overnight. I've never used Dynabeads, my IPs were all done using Protein A/G beads from Santa Cruz. In my system, I would bind my protein to the antibody and then bind to the beads. I used overnight protein binding and 1 hour bead binding, all at 4C. I was working with GFP tags and not FLAG tags so I am not sure how relevant that information is.

I think you have a lot of places to start troubleshooting. I would try these suggestions before getting another antibody. Good luck! 

Hi,

a few comments that might be of help.

1) Ab binding to the Dynabeads are usually very fast due to rapid binding kinetic. We have performed several experiments to demonstrating that increasing the time for Bead: Ab incubation is not required. extending the incubation time to 1h or even overnight will not result in much more Ab binding to the beads. When this is said, the binding efficiency will of course be dependent on which bead you have and the Ab subclass, e.g. IgG1 will bind efficiently to Prot G but to less extent to Prot A beads. In this case the M2 clone should bind well to the Dynabeads Protein G. When incubating Dynabeads coated with Ab with the lysate containing the target the incubation time will strongly depend on the primary Abs affinity for the Ag. Several different clones should be tested. the incubation time may be as short as 10 minutes. this will also reduce the background significantly as longer incubation times will also risk increasing the unspecific binding.

2) The elution efficiency will depend on the volume. Increased volume will result in higher elution efficiency. Our in house experiment shows that by increasing the elution volume from 20 µL to 100 µL will increase elution efficiency by 20%. However, our choice of using 20 µL is related to the fact that we would like to add all sample on the gel and the well capacity (BOLT gels) is 40 µL/well. In addition we do get very nice signal after 20 µL elution in our internal system. but this needs to be optimized for each immunoprecipitation. 

3) Amount of Prim Ab. I would recommend titrating this. in general we recommend between 1-10 µg Ab per 50 µL Dynabeads (per IP). but the yield of target will depend on how good your prim Ab is.

kind regards

ketil