Anti-KMT1C / G9a antibody (ab31874)

We do not use a MOPS gel or MOPS buffers. We use a Tris-Glycine gel that has a pH of 8.6, which is similar to the Tris-Acetate gel that has a pH of 8.1ish. We also use a Tris-Glycine running buffer, and transfer buffer, so none of our buffers contain MOPS or MES. We are using a similar SDS-PAGE protocol reported by one of the references listed for this particular antibody (see attachment), which .


It is hard to see on the film image, but the 250 kDa protein standard did not fully transfer, however the 150 and the 100 kDa standards did (there was no evidence of the protein standard at 100 and 150 remaining on the gel after transfer and you could clearly see both had transferred onto the membrane, so we would think if the antibody was effective, we would at least get a hint of a band at 140kDa). Another concern I have, in addition to not seeing a band at the appropriate weight, we are seeing very distinct and clear bands at considerably lower molecular weights, not faint, nonspecific bands, these are clear and obvious, did the technicians have any ideas what this might be?

ANSWER:

Thank you for your reply.

With this antibody, we have seen non-specific bands at various molecular weights below 100 kDa. Using the lot that you currently have against U2OS cells (human osteosarcoma cell line) there is a strong band around 45-50 kDa in addition to a band around 170 kDa. In IOUD2 lysate(mouse embryonic stem cell line, selected for Oct4 expression) there is a weaker band around 65 kDa. These bands are at similar molecular weights to those that you see, but they are only in certain samples and they are believed to non-specific and have not been further characterized.

I am sorry that I don't have any further information about these bands. Please keep me updated about the results with the new vial, and let me know if you have any further questions.

Thanks so much. Once we get the new vial on Monday, I will try to run and probe the blots by the end of the week so we have some answers!

Thanks for all your help!

ANSWER:

Thanks for your patience, and I have heard back from our lab about ab31874.

They tested this lot of ab31874 with both a MOPS and a TA gel, and the results from the MOPS gel showed non-specific bands and no bands around the expected molecular weight. So, if you've been using a MOPS gel I would recommend switching to TA. I apologize that this information was not previously on our datasheet and I have just updated the datasheet.

I also looked back through the images you previously sent, and it's hard for me to see on these images, but did thehigh molecular weight standards (>100 kDa) transfer efficiently to the gel? Sometimes larger proteins don't transfer well and the conditions have be optimized a bit.

Did the replacement vial of antibody arrive today? I am looking forward to hearing about the results with this new antibody. Please let me know if you have any questions or if there is anything else you need at this time.

The mouse brain lysates were taken from the dorsal striatum of the mouse brain, and lysed using an established protocol from a collaborating lab (please see attached paper from Neuron, demonstrating G9a expression in dorsal striatal tissue). Tissue was homogenized using a sucrose based lysis buffer (320mM sucrose, 5nM HEPES, 1%SDS, and protease and phosphotase inhibitors. Total protein content exceeded 1mg/ml, as determined by BCA protein assay. 5, 10 or 15ug of total protein were loaded into each well of the gel (we tried a range of total protein to determine the minimal amoun necessary to get G9a expression) and run on a 4-20% Tris/Glycince Gel. Total protein was within the range of loading protein reported in previous papers. Protein was transferred to PVDF membranes. We usually block in BSA, but we also tried using milk, neither blocking solution worked for this G9a antibody.

As far as we know, G9a is a ubiquitously expressed enzyme that is largely responsible for dimethylation of histone 3 lysine 9, so it should be readily detectable in the striatum.

I would be happy to try any tips from your lab about getting this antibody to work, and I would also be happy to try another vial of this antibody and see if perhaps something bizarre happened in shipping (I am grasping at straws now). If you are able to send another vial of the antibody, I will also prepare fresh tissue and run new gels, and try loading even more protein into the gel (20-30ug), and see if we can't get it to work. Should I pack up the two vials we have and return them?

Thanks for your help with all this and let me know what we should do on our end.

ANSWER:

Thanks for getting back to me and for sending the article.

I have emailed the lab for suggestions and/or more information about this lot, but due to the time difference (our lab is located in the UK), I won't hear back until Monday. I'll go ahead and send a fresh vial of antibody for you to try, and this vial should arrive on Monday as well. The replacement vial is coming free of charge on the order ***, and if this vial does work then I'll go ahead and send another one.

I apologize for the inconvenience with this antibody, and hopefully this replacement will produce the expected results. I'll be in touch with you on Monday, and I look forward to hearing from you.

I have attached a few more images for you from our additional trials of the anti-G9a antibody. We did 2 things, we tried antibody from each of the two vials we had here (see pic3 attachment), and we didn't get any bands in either of those blots - there was a hint of a band around 17kDa, but nothing around 140kDa where G9a should be. Since we were loading only 5ug of total protein in each lane, which is on the lower end, but has worked well for all of our other targets, we ran some additional gels and loaded 5, 10 and 15ug in consecutive wells, and then tried the antibody again (from only one of the vials) but in two different concentrations (see pic4). Again, we did not see even a hint of a band around 140kDa, but we got two strong bands around 45 and 17kDa.

I am not sure how to explain all this. Perhaps we just got two bad vials?

I looked at the other G9a antibodies that you carry, and it appears that the only other one approved for WB is ab40452, and unfortunately it corresponds to a section of the human G9a protein that differs dramatically from the mouse amino acid sequence, so it is unlikely to work. I am wondering if there is another lot of the ab31874 that would could try and see if we can't get it to work. I haven't explored other companies to see if they carry an antibody to G9a, but we have had such great success with 98% of the antibodies we've ordered from abcam, that I would prefer to try to get this to work with your AB.

Let me know if it would be possible to get another lot of ab31874, or if you have any other suggestions.

ANSWER:

Thank you for your reply and for keeping me updated. I'm sorry to hear that neither of these vials produced the expected results.

We only have vials from the same lot that you have previously tried, but I can send another vial from this lot if you would like. I'm going to email the labto ask for some further data about this lot and ask whether there is anything tricky about using this antibody in Western blot. Could you also send some additional information about the particular samples you are using? My notes indicate that the samples are murine brain lysates, but is it possible that the samples are from a region or time point that does not express G9a?

I look forward to hearing from you, and I will let you know what I hear back from the lab. Please let me know if you have any questions or need anything else from me, and I'll be happy to help.

Thanks for your quick response. I have attached a picture of the blot, and as you can see there is one dark band at approximately 45 kDa, and a lighter, but distinct band, at approximately 60kDa - G9a has a MW of 138 kDa (one of its isoforms is reported to have a MW of 179 kDa), but as you can see there is not even a hint of a band at the expected MW.

Our protocol is as follows: we load 5ug of total protein (the reference on your website said 10ug, but we found with the other antibodies we are probing for, 10ug was actually too much, and with half the amount of protein, we should still see at least a hint of a band, and even if we didn't, the nonspecific binding at 45 and 60 kDa is concerning). The tissue is mouse neural tissue lysed with a sucrose/sds/HEPES buffer with protease and phosphotase inhibitors. We block the membranes in BSA for 1hr at RT, and incubate with primary (G9a 1:1000) for 2 hrs at RT on a rocker, followed by 3X 10 min washes in TBST. Incubate in secondary (Abcam ab97064, 1:15000) for 1.5 hr at RT on a rocker, followed by 3X 10 min washes in TBST.

We initially ran the G9a antibody on membranes that had been probed with other antibodies and stripped. To ensure that these bands were not an artifact of previous immunoblotting, we reran a new gel with the same tissue lysate, and probed with G9a as described above (varying the primary concentration as follows, from right to left in the picture primary ab concentrations were: 1:1000, 1:750, 1:1500. Secondary antibody concentration was 1:15000 for all membranes).

I will try the second vial, as we discussed, and hopefully have an answer for you on that by Monday.

ANSWER:

Thanks for sending this information and the Western blot image. I look forward to hearing about the results as soon as they are available.

Please let me know if you have any questions, and have a great weekend!