Basics questions

[HugoTharinger]

Dear CyTOFicionados,

I have two general questions about the technology :
1- All atoms are ionized with the same efficiency in the plasma, right ? so how to explain the difference in signal intensity between isotpes ?
2- The yield of cells that are actually analyzed is quite low (compare to regular flow), when are we losing the cells/signals ?

Thank you,

Hugo

[mleipold]

Hi Hugo,

1. Yes, all atoms are ionized with the same efficiency in the plasma. The difference in signal intensity between, say, La139 and Tm169 is due to the mass filtering of the quadrupole mass selector. The quadrupole "kicks out" masses not in the desired mass window; it does this both above and below the mass window. With its current ~AW103-203 mass window (for a CyTOFv1), it does the best job retaining the ions near the center of the mass window. This is why there's the peak of greater "brightness" near the middle (well, shifted slightly to the higher end of the middle, as Tm169 is generally the brightest). You can imagine that in throwing out AW102, it throws out a lot of AW103. Same on the high end: to throw out basically all AW204, some AW203 gets thrown out too.

The mass window is tunable (it's based on the trigger delay), so if you shifted the mass window to 120-220, the "brightest" region would also shift higher (from Tm169 to maybe Lu175). If you shifted it down, the peak would shift down.

CyTOFv2 has a broader mass window (AW78-212, according to Scott Tanner at CYTO2013), so its optimum brightness may be at a slightly different position than a regular CyTOFv1


2. The cell transmission efficiency is primarily due to the fluidics and the spray chamber. The cells stick to the sample loop tubing, the valve, the nebulizer tubing, the nebulizer, and also build up in the spray chamber, ball joint, and the injector. If everything is clean and optimized (coaxial stream/spray from the nebulizer, properly adjusted makeup gas and nebulizer gas flow rates, etc), your cell transmission efficiency should be in the 20-30% range. However, if your nebulizer is partially clogged and is therefore spraying off at an angle, this will clearly lower your cell transmission efficiency.

I would like to make a related point: if the nebulizer or nebulizer tubing are partially clogged, your tuning solution signal will usually be lower. This is also a spraying efficiency issue. However, the tuning solution is *free metal* in solution: if half your tuning solution winds up on the walls of the spray chamber, your ion count/signal will be low.

However, if you then start running cells, the *cell-associated* signal will still give you the correct signal intensity for the cell.....you'll just have fewer cells making it into the machine. This is an important distinction: you won't be losing half your ions/signal associated with a cell (and therefore have an artificially "dim" cell), just losing half your cells.


Mike

[HugoTharinger]

Hi Mike,

Thank you very much for your crystal-clear answers.

Hugo

[antonio]

Hi Mike,
we tried modify the trigger delay and this does not change the brightest region of our CyTOF.
The trigger delay act on the TOF part of the instrument and the brightness as you clearly explained depends on the quadrupole.
I will add depend also by the passage of the ions throw the cones.

Did you test modifing the trigger delay on your machine?

Antonio

[mleipold]

The trigger delay sets the start of the mass window.

It has to be moved a fair amount to shift the mass window. I don't know the numbers off the top of my head, but I think you need to move it at least a few hundred units to get significant shifts.

If you want to move the window, I suggest contacting DVS and getting their input/advice.

[antonio]

At the best of my knowledge the trigger delay unit, let's say nanosecond to be precise, indicate at which TOF the instrument start to register data. It has nothing to do with the ion transmission.

In our lab we tested the following trigger delay: 8400, 8000 and 9000 with no differences in the relative brightness among metals. Consider that 30 nanosecond in the TOF correspond approximately to 1 atomic mass unit at the moment in which the Tb is heating the detector.

Did you test different trigger delays in your machine?

[mleipold]

We haven't changed our trigger delay much.

One way to test how much trigger delay change is necessary to move the TOF window would be to stain cells with both the Rh intercalator (103, bottom of the standard CyTOFv1 mass window) and cisplatin (190, 192, 194, 195, 196, 198).

If your trigger delay would move an isotope out of the acquired mass window, the Analytes tab should give you a red mark for that isotope/mass, and not allow you to save the analytes panel. I know this happened with our machine, where Rh103 and Pd102 wouldn't "save" until we shifted the trigger delay.

I'm not entirely sure it will allow you to measure both Rh103 and Pt198 in the same panel, but I haven't tried it.

[SDTanner]

Sorry it has taken me so long to discover this forum.
Mike's answer is close, though there is a more complete answer for technology-junkie's (if you're not one of those, don't bother with the following):

The questions are basically about ionization efficiency and mass bias.
To a first order approximation, all atoms having an ionization potential below about 9eV (all the lanthanides, and many other elements with notable exceptions) are 100% ionized in the plasma. (for tech junkies: the degree of ionization is determined by the Saha equation assuming local thermodynamic equilibrium - I can send a periodic table with degrees of ionization (or can this forum accept attachments?).
There are elements of mass bias (mass-dependent transmission efficiency) during and after extraction of the ions from the plasma. There is a small mass-dependent bias (actually energy-dependent, since the ions are extracted from the plasma at constant velocity, which provides mass-dependent energies) during the supersonic expansion itself, then a significant (but noticeable really only for masses below about 100 amu) space-charge-dependent mass bias - which the CyTOF largely controls through it's unique extraction ion optics. The dominant source of mass bias occurs in the high-pass rf ion optic. Except for special cases (low ion energy or high pressure), transmission of ions through an rf device is periodic: that is, there is convergence and divergence as the ions experience the rf forces. This periodic focusing is mass- and energy-dependent. We typically tune the CyTOF to provide a maximum focus for ions of mass about 159-169, and there is a reduced transmission efficiency at higher and lower masses. In fact, if you were able to observe the entire mass spectrum you would likely observe another maximum at very high mass (except that there are no atoms at that high mass) and some periodicity at lower mass. There is additional mass bias against transmission of lower mass ions because of the manner in which the dominant argon and argide ions are blocked from the TOF region (where they would cause considerable distortion, resolution reduction and rapid decay of the detector). The periodicity is most easily adjusted by changing the magnitude of the applied rf in the transfer optic (not accessible except in service mode because this must be done with thorough knowledge of the implications): it can also be affected by the rf frequency and ion kinetic energy (not readily adjusted in the CyTOF). There can be some mass bias in the TOF acceleration, reflection and collection regions - but the CyTOF ion optics are designed to minimize this. There is essentially no mass bias induced by the trigger delay.

I can provide literature references for those who really want to know more about the supersonic expansion, space charge and rf-dependent focusing.

[mleipold]

Hi Scott,

Thanks for the more detailed response.

Yes, the forum can accept attachments. Beneath the box where you make your response are two tabs: Options (default), and Upload attachment.

I'm not sure what the size limit is, but people have been posting PDFs of FlowJo screens without problems.


Mike

[HugoTharinger]

Thank you all for the responses,
This forum is coming in handy!!

Hugo