Cell Composition

[Sebastian Ramisch]

Hi Everybody,

We took a first look at the cell composition of our organoids, and a few questions came up:

1. Following our first flow analysis, we received similar results to those shown in the protocol (using a slightly modified panel), however almost half of our cells could not be categorized into any of the lineages presented in your protocol (Fig. 1E). Do you have any insight into what these cells might be, and could you share the percentage of “unidentifiable” cells from your dataset?

2. Our flow data also identified a cluster of larger cells with higher granularity within our cell population. (Screenshot attached)

Have you encountered something similar or have an idea what this could be? (maybe cells from the primitive hematopoietic wave?)

Thanks for your assistance!

Best,

Sebastian 

[Abdullah Khan]

Hi Sebastian, 

That's great to hear - well done! 

We will go through our data and get back to you re- the fraction of unidentified cells. 

It would help if we could see your panel so we can get to grips with your modifications? 

Do you have any imaging you can share? 

Also remind me what iPSC line do you have? 

Are they CD45- or +? CD71+ or CD235-? 

We have had a quick chat here and likely culprits for your large granular cells would be: 

- Megakaryocytes 

- Macrophages (can check with CD63 and CD168)? 

- Granulocytes of some description (CD117 and FCER might help you work this out too) - these seem less likely as they are usually small. 

I would suggest back gating on that large granular population and seeing what markers do come up? 

Have you done any imaging of your samples at a corresponding flow time point? It's possible you had an inefficient mesodermal diff (either because of a problem with a low O2 incubator or something to that effect). 

If you give us a bit more detail as above we can probably pin point the issue. 

Thanks! 

Abs 

[aude-anais.olijnik]

Hi Sebastian,

Thank you for your message and well done on getting the bone marrow organoids on the way in your lab!

Just to further add to what Abs was saying: please could you share your flow panel and gating strategy to see where these unidentified could come from? In my hands, working with the KOLF2 iPSC line, I have ~15% unidentified cells in the haematopoietic panel and ~3.5% unidentified in the stromal panel (average over 3 timepoints, day 14-18-21).

About the "granular" cells: are you looking at haematopoietic markers or stromal markers? Please can I ask at which timepoint this screenshot has been taken? Do you see a change in the proportion of these "granular cells" over time? 

Thank you!

Aude.

[Karolin Stumpf]

Dear TooT,

also regarding the flow cytometric assessment of BMOs: Do you usually add FcR block or just FBS to the staining mix?

Thanks,

Karo

[aude-anais.olijnik]

Dear Karo,

Thanks for your email. We only add FBS to the staining mix, using a standard FACS buffer recipe (Dulbecco's PBS + 10% FBS, I personally always filter my FBS with 0.45um filter before mixing to remove any micro aggregates).

Thank you!

Aude.

[Karolin Stumpf]

Hi everyone,

regarding the cell composition assessed by flow: how much variability do you see between organoids if, for example, only pooling 2 organoids for dissociation or even doing single organoid flow?

Best

Karo

[Abdullah Khan]

I think you have a copy of the Nat Protocols manuscript right? it should have single organoid and experimental variation quantified in the supplement. 

Z. has been doing single organoid experiments with good results but I am still a bit wary of pooling less than 4-6 per experiment just to be completely sure of inter-organoid variability. 

Abs

[Zora Baumann]

Following up on the cell composition, I was wondering what the overall ratio of CD45+/CD45- cells are in the comBO? I only found relative abundances in the PrePrint.
Of course it likely depends on the hiPSC line, but I was curious about rough ratios.

[Abdullah Khan]

Hi Zora - it is a bit line dependent but 60% CD45+ is usually what we expect, though for some lines it is lower and others higher. With comBo, the CD45 fraction expands a lot more at later time points. What ratios are you getting? 

Abs 

[Cecilia Reyneri]

Hi Abs,

We have been testing the ComBo protocol on Gibco and SCTi003A, checking the cell composition at DIV21 with flow cytometry. Our organoids have most of the cell types showed in the BioRxiv (endothelial cells, MSCs, different types of hematopoietic cells), but the overall amount of CD45+ cells is quite low (around 10-20% of all live cells).

I believe the only difference is in the cytokines that we used, as we are using Peprotech cytokines instead of Stem Cell Technologies. Do you think this could explain the low frequency of hematopoietic cells? Have you ever tried using other manufacturers?

Thanks,

Cecilia

[Zora Baumann]

Hi Abs, thanks for the response. I got 60-70% CD45+ at d34 across an hESC and a hiPSC line :) It's the first time I managed to get a good viability (previously I had lower percentages, but also lower viability). 

Hi Cecilia, I also use Peprotech for all my growth factors (for the comBO protocol). I've never done flow cytometry on d21, but maybe as Abs says the fraction expands a lot after changing to the StemPro media? I for one see hematopoietic cells (erythrocytes, neutrophils, and monocytes, as well as some CD45-) being shed from around d28 onwards :) 

[Abdullah Khan]

Thanks both - i think Zora is right it's the time point, moving into normoxia + stemPro you get a big expansion of the haem fraction. 

Well done both, very pleased you're getting good data! 

Abs