Anti-CENPA antibody [3-19] - ChIP Grade (ab13939)

Apart from an aknowledgement of receipt of the below email, I have heard nothing more on the subject. We are a little at a loss as how to proceed, having used freshly prepared chromatin as suggested.

We have finally done a repeat experiment on ab13939 and are still not happy with the results.

We used latently infected (with HSV1) MRC5 cells as per our normal protocol. We used freshly prepared chromatin (ie immediately after cell lysis, pre-clear for 1hr then add antibody) as suggested, and as a comparison, as we wished to validate two other antibodies, used parallel frozen samples.

Each sample corresponded to 1/4 of a 75cm2 flask (approx 1x10^6 cells). The samples were no antibody, 7.5ul of Upstate 06-599, 10ul of antibody ab13939 or 10ul of ab1791 - to the C-terminal of histone H3.

I note that the epitope for ab1791 is VQLARRIRGLEEG. The equivalent portion of CENPA has the sequence IQLARRIRGERA. Thus I would not be surprised to see some cross reactivity between ab1791 and the histone varient CENPA. Since the data sheet states that ab13939 is purified using Protein A affinity chromatography, we saw no problem in using Protein A to pull down antibody/antigen complexes.

Various PCRs were run. ab1791 and 06-599 performed exactly as we would expect on HSV1 ICP0 promoter, GAPDH promoter, beta-actin coding sequence and gamma-globin promoter and, as expected, there was no specific enrichment with ab13939. We found no differences between fresh and frozen chromatin, but the frozen sample had only be frozen for 2 days to check for gross degredations as we had previously not fully validated ab1791. ab1791 gave IP over input values similar to those we have seen before for these regions, and 06-599 only giving a high IP over input for the active promoter of GAPDH as expected. Most no antibody values were in single figures.

However when we performed the hsSat-alpha PCR (primers as per the ab13939 data sheet) we found no specific enrichment with ab13939 - the PCR values were what I would classify as background or just above, and similar to the no antibody values. ab 1791 however gave a strong positive value. I do not think I can give a fold-enrichment as your data sheet does, as my no antibody values are too low to be of numerical significance (enrichment values for ab1791 are 100-1000x which I consider meaningless since the denominator is such a small number). However, we also used an input (no IP, just chromatin that had been removed from the IP samples, and treated as post-IP chromatin) in the PCR. The IP over input values for ab1791 were similar for hsSat-alpha to that seen for the gamma-globin promoter or beta-actin coding region and somewhat higher than for the GAPDH promoter.

There are several interpretations or observations to be made to this experiment:

1) ab1791 cross-reacts with CENPA histone H3 variant.

2) if observation 1 is true, then we are still not getting ab13939 to work, despite parallel IPs with other antibodies using the same chromatin prep working well and as expected and finding histone H3 or a cross-reactive variant present at the hsSat-alpha locus, and having followed your suggestion of using freshly prepared chromatin.

3) if there is no cross-reactivity between ab1791 and the CENPA histone H3 variant, then MRC5 cells do not possess CENPA at their centromeres or have an unusually high frequency of Histone H3 rather than CENPA.

Do you have any further suggestions?

Yours sincerely

ANSWER:

Thank you for getting back to me. I am sorry that you did not receive a response to your previous enquiry.

I appreciate the lengths that you have gone to in order to get this antibody to work, using fresh chromatin. The reduction in signal at the satellite repeat region following ChIP is something our chromatin post doc scientist regularly finds.

To address your questions. Given the sequence similarity I consider it highly likely that our histone H3 antibody ab1791 cross reacts with the histone variant CENPA.

The enrichment that our postdoctoral scientist detects is subtle. I consider it most likely that you have simply not been able to reproduce this results rather that MRC cells having low levels of the histone variant.

I am happy to provide you with credit against your purchase. For me to do so please can you provide me with your original order details.

Thank you for your patience in this matter. I look forward to hearing from you.

> I would appreciate it if you could tell me whether the chromatin that you prepared was fresh. Our in house postdoc often finds that chromatin that is frozen or aliquotted does not efficienty precipitate when using antibodies directed against heterochromatic targets.

Snap frozen in LiqN2 and frozen at -70 - but we routinely do this for all our chromatin and have never had any problems. However, if heterochromatin is different, that may be a problem. Is there any proposed explanation as to why snap freezing is fine normally but not for heterochromatin?

> Can you also tell me the results from the antibodies that you applied in parallel to ab13939. Did they behave as you expected. Did you observe the expected enrichment using ab8898?

I've not done ab8898 in the parallel experiment I'm afraid as we've finished our dataset with that antibody, but other experiments where we've used ab8898 expecting enrichment, we've found it.

In experiment 2, we used other antibodies which gave the expected results - ab1791 and upstate 06-599 behaved exactly as we'd expect. Experiment 1 was simply to test the levels of enrichment with ab13939 as per the datasheet.

ANSWER:

Thank you for getting back to me.

I have been in correspondence with our in house post doc as as I suspected in her hands, when doing chip with markers against heterochromatin "it is important to have fresh, very good quality chromatin...I get a reduction of the signal and in the end I completely lose it (e.g. chromatin stored at -70oC for 2 month works very well for 8580 (histone H3 tri K4) but I don’t get a signal with any of the heterochromatic markers). This may be due to the fact that heterochromatin may be less soluble and `falls out`".

This is something that she routinely finds and may be attributable to the solubility of heterochromatin following freeze thaw.

I would like to suggest that you apply this antibody to fresh chromatin using parallel eurochromatin and heterochromatin control antibodies e.g. histone H3 tri K4 (ab8580 and Histone H3 (tri methyl K9) as this will demonstrate whether it is the chromatin preparation or the antibody that is causal in the lack of signal that you have been obtaining.

I hope this information helps, please do not hesitate to contact us if you need any more advice or information.

BATCH NUMBER 184400 ORDER NUMBER not know

DESCRIPTION OF THE PROBLEM Direct repeat of experiment shown on datasheet. ChIP using chromatin from MRC5 cells (human - 1x10^6 cells) with 10ug CENPA antibody or no antibody. PCR primers as on datasheet hsSat alpha primers. Relative enrichment with antibody as opposed to no antibody was 1.8 fold. Second experiment using our standard ChIPs layout gave IP over input values around 0.0003 for CENPA and the hsSat alpha PCR primers SAMPLE MRC5 cell line latently infected with virus. DETECTION METHOD Real time PCR using primers specified on datasheet with Sybr green. MgCl2 and SG optimisation done on primer set. Other primers used are all Taqman. POSITIVE AND NEGATIVE CONTROLS USED Expt 1 - antibody v no antibody to replicate datasheet QC Expt 2 - different latently infected cells - viral DNA versus Hu satellite DNA versus Hu globin gene not expected to be associated with CENPA - IP over input values the same for globin as for satellite. ANTIBODY STORAGE CONDITIONS As recommended. Entire batch (only apparently 40ul in tube even after recommended centrifugation even though there should be 50ul) used in 2 simultaneous experiments. SAMPLE PREPARATION 4-5x10^6 cells, formaldehyde fixed for 10 mins, lysed in presence of Roche EDTA-free mini protease inhibitor + PMSF. Lysate split 2.5:2.5:2.5:2.5:1 for 4xChIPs and 1xInput analysis X-CHIP or N-CHIP X-CHIP using standard protocal based on Abcam and Upstate protocols which works successfully with various other Abcam antibodies (eg ab8898, Upstate 06-599) CROSSLINKING Standard 10mins formaldehyde. DNA FRAGMENTATION Sonication - fragments 300-800bp IP STEP O/N - 10x diluted chromatin (250ul into 2.25ml dilution buffer + protease inhibitors) which corresponds to approx 1x10^6 cells, 10ul antibody (probably more than recommended, but we always start high with a new antibody as we can lower the amount in subsequent experiments if we get a strong signal on the appropriate samples), o/n 4degC. 150ul proteinA Sepharose - 3-4 hr 4degC, washes as Upstate and other published protocols, elution in 1%SDS/0.1M Na bicarb, 3x >10min, last elution at 65degC, others at room temp. DECROSSLINKING O/N 65degC 0.2M NaCl, 4 hr 37degC proteinase K at 0.66mg/ml DNA PURIFICATION Phenol, CHCl3, EtOH (with glycogen/linear acrylamide carrier) - EtOH step o/n at -20degC. 70% EtOH wash done. PELLET AFTER PRECIPITATION Large pellets of carrier seen and not lost. Pellets resuspended in 250ul water and 8ul portions used in triplicate real time PCRs. Inputs decrosslinked and purified as IPs and resuspended in 800ul. PRIMERS TESTED ON GENOMIC/INPUT DNA Standard curve done on HEK293 DNA - looked OK although not as good as my Taqman PCR results normally are. Products also run on gel as I don't trust Sybr green - product appears to be approx 200bp - all other PCRs we do are Taqman. Inputs amplified OK with the primers. ie I'm fairly happy that the PCR is working fine. NO-TEMPLATE CONTROL IN PCR Yes, as usual. And all PCRs are run with standard curves - all samples done in triplicate. HOW MANY TIMES HAVE YOU TRIED THE APPLICATION? 2 DO YOU OBTAIN THE SAME RESULTS EVERY TIME? Yes WHAT STEPS HAVE YOU ALTERED? None so far as our standard protocol works with various other antibodies and I have run out of antibody.

ANSWER:

Thank you for your enquiry.

I am sorry to hear that you have been having difficulties with this antibody. I have read your technical questionaire and you are performing an approach that I would largely recommend. However, I have a few comments.

I would appreciate it if you could tell me whether the chromatin that you prepared was fresh. Our in house postdoc often finds that chromatin that is frozen or aliquotted does not efficienty precipitate when using antibodies directed against heterochromatic targets.

Can you also tell me the results from the antibodies that you applied in parallel to ab13939. Did they behave as you expected. Did you observe the expected enrichment using ab8898?

I look forward to hearing from you.

Thank you. Do you have any information on suitability of your antibody against CENP-A (ab13939)for use in chromatin immunoprecipitation (ChIP). We recently purchased this from you (lot#142718) and have had no success with it in ChIP despite published evidence (Ando et. Al. 2002) that it should work in this application.

ANSWER:

Thank you for getting back to me.

CENPA antibody [3-19] (ab13939) is not currently ChIP grade. I have been in touch with my colleague in the lab responsible for our ChIP testing program at Abcam. She has actually just finished testing this antibody by ChIP. She mentioned that she observed an enrichment of CenPA on the centromeric repeat (SATa) only. The signal is not very high but the result is reproducible. Furthermore she commented:

"I am not surprised that that customer has problems ChIPing with this antibody. Heterochromatin is difficult to examine by chip as you need good and fresh chromatin. It is important to have a chromatin fragment size down to 500bp as heterochromatin does not sonicate so well and is less soluble (I think you loose it in the last centrifugation if it is bigger). It does also precipitate when the chromatin is stored at -70oC for longer periods…"

I hope that this information helps, please do not hesitate to contact me should you require further assistance.

BATCH NUMBER 169079 ORDER NUMBER 128408

DESCRIPTION OF THE PROBLEM no-specific staining

SAMPLE rat cardiomyocytes human cells (various cell types from heart)

PRIMARY ANTIBODY from Abcam

DETECTION METHOD alexa-conjugated secondary (fluorescent miscroscope)

POSITIVE AND NEGATIVE CONTROLS USED human cells and rat cells

ANTIBODY STORAGE CONDITIONS as recomended

FIXATION OF SAMPLE PFA+methanol PFA

ANTIGEN RETRIEVAL not needed

PERMEABILIZATION STEP 0.2% tritonX in PBS + 4% FBS

BLOCKING CONDITIONS 4% FBS

SECONDARY ANTIBODY used all the time from [another company], it works fine with other primaries done at the same time.

HOW MANY TIMES HAVE YOU TRIED THE APPLICATION? 3 HAVE YOU RUN A "NO PRIMARY" CONTROL? Yes DO YOU OBTAIN THE SAME RESULTS EVERY TIME? Yes

WHAT STEPS HAVE YOU ALTERED? fixation, Ab concentration, Ab lenght incubation

ANSWER:

Thank you for your enquiry. I am sorry to hear that you have been having difficulties with this antibody.

I have read your technical questionnaire and I have a few comments. I would appreciate it if you could provide me a few further details including the dilutions that you have been using and the fixation conditions. You also mentioned that you have varied both during the optimisation. I would be interested in knowing the different fixation and dilution conditions that you have applied. We recommend that this antibody is applied at a concentration of 5 ug/ml.

Could you also detail the nature of the non-specific staining that you have been finding. Was it cytoplasmic, nuclear, distinct foci etc.

I appreciate your comments and look forward to hearing from you.