FAQ  •  Register  •  Login

Help! Unexplained Background

Forum rules
Please be polite and civil. We know that troubleshooting is vexing...
<<

emallen

Participant

Posts: 18

Joined: Thu Oct 23, 2014 6:10 pm

Post Wed Oct 29, 2014 11:02 pm

Help! Unexplained Background

Thank god I found this place! I've spent the last year in near isolation from other CyTOF users, and I have so many questions that I haven't been able to answer. I can't tell you what a relief it is to find people who know what they're doing!

Lately I've been struggling with high background and low signal. Until recently my cells have been staining brightly with very little background, but in the last few experiments the intercalator-associated signal has dropped and background is higher than I've ever seen across all antibody channels. I have completely standardized pelleting and supernatant removal, so I'm relatively sure this isn't caused by inefficient washing steps. So I did the only thing I could think of... I had been using the same bottle of DVS buffers for many months now, so I decided to try making an in-house CSM (PBS+BSA+azide) and comparing that to my old bottle of DVS CSM in a small experiment. This brought my signal back up, but I'm still seeing background in certain channels containing antibody (mainly channels designated for high-expressing markers such as CD45 and CD3). Just to be thorough, I measured the pH (7.5) and tested for microbiological contamination (culturing on LB agar) in my old bottle of DVS CSM. As far as I could tell, nothing seemed out of order.

This is all very puzzling, and has opened the door to many more questions. Any help offered would be so greatly appreciated!

1) In general, what are the most common causes of background/streaking? What about uncommon ones?
2) What could cause a drop in signal that is concurrent with bright background across all antibody channels?
3) What are the most common sources of background that only occur in some channels - in my case, CD45-156Gd, CD11c-162Dy, HLA-DR-174Yb, CD3-175Lu, CD8-176Yb are the worst offenders. All of these have been titrated and have shown very little background in some previous experiments (though not reliably... see #6)
4) If there is Ba contamination, which channels would be affected?
5) If I'm only seeing background in my high-expressing markers, would that indicate that the problem isn't due to improperly titered antibodies (because high-expressing markers can bind up more antibody than low-expressing markers, therefore it is more likely that I would see excess antibody for low-expressors)?
6) I've noticed background seems to fluctuate a great deal between experiments and even between samples in the same experiment. I'm relatively sure this isn't due to technician error because I have spent a considerable amount of time standardizing my technique. What else could account for this?
7) My background seems to increase and decrease with cell concentration (even if resuspended in the same sample volume - for example, if I resuspend 1e6 vs 2e6 cells (stained with same concentrations of antibody) for acquisition in 1 mL of water, background tends to be higher in the 2e6 cell sample. Has anyone else noticed this to be true?
8) Recently I've been resusupending my pellets in residual volume by light vortexing (as indicated in MaxPar protocols). Could this be increasing my background due to more cell debris, etc?
9) Also recently I've also begun staining my cells on ice outside of the biological hood using non-sterile supplies (reagent basins, tips, etc.) to save money. Could this be contributing to metal contamination?
<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Wed Oct 29, 2014 11:12 pm

Re: Help! Unexplained Background

HI Elyse,

Regarding your sudden loss of Ir signal and sudden increase in background across all channels:

The most common reason for this is insufficient cell fixation. If your cells aren't rock-solid-fixed, they'll start to lyse once you do the MIlliQ water washes and resuspensions. One common indicator of this in your plate/tube is that your pellet was nice while still in PBS, but as soon as you do the MilliQ water washes, the pellet gets smaller and smaller.

If your cells are lysing, you just have cell debris, and the nucleus can be disrupted, thereby leaking Ir intercalator.

I don't know what you are using for fixation, but we generally recommend not using any 16% PFA that's more than 1 month old. Make the 2% fixation buffer fresh every time. PFA is cheap, your experiments aren't.


Mike
<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Wed Oct 29, 2014 11:26 pm

Re: Help! Unexplained Background

HI Elyse,

To address some of your other points:

Barium has several isotopes, but the primary one is Ba138. Therefore, its oxide spillover would be in 154.

I would re-titer your antibodies if you're having background problems. You might also centrifuge the stocks and pipet the supernatant to another tube, in case aggregates have formed....sometimes that happens, but it's an antibody-by-antibody basis.

Background and streaking are usually caused by:

1) Cell debris. Either inherent to your sample (low cell viability) or from improper fixation. Do more washes, and use new(er) stocks of fixative. Note: just because a PFA stock is good for standard flow does *not* mean it's good for CyTOF: flow is usually done in PBS, which is osmotically kinder to your cells than MilliQ water is.
2) Antibody titration. Every stock you make in-house, you should check the background in negative populations. Sometimes you wind up with a super-active prep, and have to titer down. This should be less of an issue with DVS antibodies, but....
3) Antibody aggregates. If an antibody aggregates, only some of the surface of the aggregate will form a high-affinity interaction with the antigen. The rest of it can continue to break off and streak with every wash. Centrifuging your stock and transferring over to a fresh tube may help, as well as filtering your working cocktail through 0.1um filter before putting it on your cells.
4) Inefficient washing. Make sure that your cells are resuspended uniformly for each wash (no clumps, etc). Also, for larger cell numbers, you may need to do more washes at each step, as the wash volume to cell ratio will be lower.

I don't use sterile tips for my work, as I'm not culturing the cells, even overnight. Of course, you will want to work with live cells in a hood for biosafety reasons, but once they're fixed, I usually do the rest of the procedure (perm, Ir, etc) on my bench.


Mike
<<

emallen

Participant

Posts: 18

Joined: Thu Oct 23, 2014 6:10 pm

Post Thu Oct 30, 2014 12:00 am

Re: Help! Unexplained Background

All great suggestions, thanks! How fast/long should I spin my antibodies to be assured that aggregates have spun down sufficiently?
<<

emallen

Participant

Posts: 18

Joined: Thu Oct 23, 2014 6:10 pm

Post Thu Oct 30, 2014 12:12 am

Re: Help! Unexplained Background

As far as fixative, we use sealed ampules of 16% PFA bought from the manufacturer (definitely older than 1 month) to make a 1.5% working solution, which we aliquot and store at -20C. That's my in-house fixative. I'm hoping that a sealed ampule of 16% PFA can go longer than a month?

Otherwise I use the DVS Fix/Perm Buffer which has been sitting in a cold room for several months but only recently opened.

I actually compared these two fixation methods, and it doesn't look like there is any difference in the amount of background. I'm fixing either for 30 min at RT or O/N at 4C.

Finally, do you have any product recommendations for 0.1um filters? Ideally I'd like to find some that waste minimal fluid.

Thanks so much!
Elyse
<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Thu Oct 30, 2014 3:11 pm

Re: Help! Unexplained Background

Hi Elyse,

Unoperned ampules of 16% PFA are stable at room temp for any reasonable amount of time. However, once you open one, you should use it within a month. I transfer the contents to a 15mL conical tube and keep it, undiluted, at 4C. I then make the 2% PFA in PBS fresh every time I need it. I date the stock tube, and throw out after a month.

To spin dwn your antibodies to remove aggregates, I would max out a microcentrifuge (usually around 12K rfg) for maybe 10min. For filtering the cocktail before use, I use Millipore Durapore PVDF 0.1um filters, UFC30VV00. The filters will retain 20-50uL; since I stain in 50uL, that typically means I make an extra test's worth of cocktail.


Mike
<<

emallen

Participant

Posts: 18

Joined: Thu Oct 23, 2014 6:10 pm

Post Fri Oct 31, 2014 12:08 am

Re: Help! Unexplained Background

Great thank you! I really appreciate all the help. I'm so sorry for all the questions, I've been saving them up for a long time.

I've prepared some raw acquisition screencaps for you to look at, if you're willing. I included three experiments... 1 normal acquisition, 1 where I saw decreased antibody signal and increased background in all channels, and 1 where I switched out the CSM and/or fixative to see if that would improve signal/decrease background (four screencaps total). From this, I've deduced a few things...

1) I am relatively sure this is not a titration issue, as I've used all of these antibodies (same lot) previously at the same concentration with no problems (see first experiment).
3) In my second experiment, I'm seeing decreased antibody signal + increased background. If the background is caused by cell debris, what would account for the decreased antibody signal on intact cells? If antibody aggregates were to blame, wouldn't background only appear in some of the channels?
2) If you'll look at my third experiment, using new In-House CSM with the DVS Fix (screencap 1) solved some of the background problems and improved antibody signal, but I still see background in only some channels. What gives?

Any thoughts?

Finally, how do you fix? These last two (faulty) experiments were fixed for 30 min at RT, when I normallly fix O/N at 4C. Could that be making a difference? If you have a protocol I could see, I could compare mine to yours.

Again, thanks so much!
Attachments
Background Troubleshooing.pdf
(369.42 KiB) Downloaded 453 times
<<

mleipold

Guru

Posts: 5792

Joined: Fri Nov 01, 2013 5:30 pm

Location: Stanford HIMC, CA, USA

Post Mon Nov 03, 2014 10:04 pm

Re: Help! Unexplained Background

Looking at your PDF:

The problematic experiment (lower, page 1), I would guess is cell lysis/debris. Every channel is affected, in terms of having a high background. Even of the cells that might be high Ir+Ir+, you could check for marker combinations that don't make sense (CD3+ CD20+, for example). But just in general, I don't really see a nice, clearly defined cell (Ir+ Ir+) that's also nicely and clearly positive for other markers. You do have one Ir+Ir+ event about 1/3 down, that's maybe CD45+ CD33+ CD14+ TLR2+ CD38+, but I would have expected a brighter CD45 signal (based on abundance of CD45, as well as how bright that metal is). Also, if it's a monocyte, I would have expected it to be brightly HLADR+, which it doesn't appear to be.

Regarding In-House CSM + DVS fix, I would say that you could use another wash, and/or dilute your sample more: you seem to have an average of 2-3 cell events/screen, which is a bit fast....we usually aim for 1-2 cell events/screen. If you dilute your sample, background will go down, while true cell-associated signal will stay high.


We usually fix 2% PFA in PBS 4C O/N.


Mike
<<

emallen

Participant

Posts: 18

Joined: Thu Oct 23, 2014 6:10 pm

Post Wed Nov 05, 2014 6:38 am

Re: Help! Unexplained Background

Thanks Mike, I appreciate the input. I'm going to do a massive troubleshooting experiment this week, which should give me some answers. I'll post my results in case people are interested.
<<

vtosevski

Contributor

Posts: 44

Joined: Wed Nov 20, 2013 12:50 pm

Location: Zurich, Switzerland

Post Thu Nov 06, 2014 8:58 pm

Re: Help! Unexplained Background

Hi all,

A lot has been said about potential sources of the background and I just briefly wanted to mention a possible explanation for the reduction in signal - could it also not be the background??

By default, "noise reduction" feature is activated on the analysis tab of the acquisition window. As far as I know, this feature will measure the average intensity in between the transients and deduce this from the intensity of the transient itself. If the background is high, the intensity of the transient will be made lower than what you would expect.

Just a thought...

Vinko

Return to CyTOF troubleshooting

Who is online

Users browsing this forum: No registered users and 8 guests