Picking single cells with CellSorter (FACS in a petridish) machine?

[Simone78]

Dear all,
we are interested in picking single cells for RNA-seq and we are looking around for semi- or fully automatic devices capable of doing that with high throughput in a very short time. Looking of the solutions currently available we came across a company commercializing a CellSorter - FACS in a petri dish. Both fluorescent and unlabelled live cells in a Petri dish observed with a microscope can be automatically recognized by computer vision and picked up by a computer-controlled micropipette fixed to the objective lens. The maximum number of cells picked up from a Petri dish was about 1,000 limited by their current sorting speed of 1 cell per second.
The company is based in Budapest (Hungary) but has no distributors in my country (Sweden). I was wondering whether anyone had heard of or worked with this machine. Is it really as easy as they claim?
Thanks a lot!

[Jon_Keats]

Why not just use a classic cell sorter? Good chance you have a FACS sorter around

[Simone78]

I´m not very familiar with FACS but, as far as I know, it´s very difficult or impossible to get a population of cells that is 100% pure in a short time (minutes). Ideally, we want to have 1 cell/well of a 96- or 384-well plate. This CellSorter seems to be able to do so, but I couldn´t find any paper or lab that is currently using it, and I was wondering whether it´s because it´s new or because it doesn´t work as they claim.

[Jon_Keats]

Almost all current generation cell sorts, such as a BD FACS Aria, will sort single cells or multiple cells into 96 well plates. The speed is dependent on you situation but from a culture of healthy cells when you sort viable cells it should take several seconds. Find your local cell sorter this is every day kinda stuff.

[Simone78]

thanks a lot, I´ll take a look at it! I was just concerned about losing too many cells in the sorting process. This could be an issue when dealing with rare (and small!) samples. But the machine looks promising, albeit much more expensive than the CellSorter I was talking about , which costs only 12,000 EUR (too good to be true...)

[454newbie]

Can you post a link to the supplier? sounds interesting!

[NextGenSeq]

FACS sorted cells VERY often yield degraded RNA.

[Jon_Keats]

FACS sorted cells VERY often yield degraded RNA.

Can't say that is my experience, but methods can very dramatically

[Simone78]

Here´s the link of the cell picker I was talking about:

http://www.facsinapetri.com/

[larsseq]

Hi Simone!

Have you tested the "FACS in a Petri" by now?

Would be an interesting solution for me as I want to sort fixed cells (immunofluorescence) based on image properties, so conventional FACS is not an option. However, as you mentioned, the FACS in a Petri thing seems a bit "too good to be true".

Cheers, Lars

[Simone78]

we saw the cell picker at work in their lab in Budapest last spring and it seemed to be working ok (even if they had some issues with the settings while we were there). The problem is that the cell is picked in 5 ul buffer which, for our application, is way too much (we want it in 1 ul). After further optimization, last month they told us they managed to go down to 1 ul and they are soon going to send us some picked cells. We will check the integrity, try to make RNA-seq libraries out of them and then decide whether to buy it or not.
Best,
Simone

[kwaraska]

I just attended a seminar by Fluidigm. They have a plate that will select one cell into each well by "blocking" the used sites.

It takes a specific machine but from there you can stain them, isolate RNA even create NGS RNA-seq libraries.

Check that out-might be easier than a FACS-epecially if you don't have one!

[Simone78]

yes, the C1 single-cell auto prep system from Fluidigm seems to be a great instrument. The problem I see is that you need to buy an expensive machine + their reagents (probably very expensive as well). We finally bought a cell picker from CellSorter (http://facsinapetri.com/), expected delivery in 2 months. I´ll post my results and impressions about that.

[Jaykay]

Hi Simone,
just wondered if you already got some results with the CellSorter? I want to sort some large macrophages for RNA analysis, which would not survive being sorted by a FACS sorter.

[Simone78]

In principle it works, but there is a big "BUT". There are a few bugs in the software. Sometimes we are ready to pick cells but we have to restart the program because it´s not responding. This increases the time the cells are on the Petri dish, which increases stress (thus, decreasing viability) and lead to unwanted changes in the expression profile. This, at least, it´s my interpretation of things. To give you an idea of what I mean, I tell you the conclusion of our latest experiment. We compared single nuclei (GFP labelled) picked with CellSorter or by FACS. We got comparable libraries before seq (similar yield and avg size of library), but they cluster apart from each other...the FACS-sorted nuclei on one side of the plot, the CellSorter ones on the other! Don´t know exactly what that means. Again, it might be an artificial change in the expression profile due to the longer time cells were on the petri dish. Or it might be true (?).
With CellSorter there are also issues related to how strongly the cells are attaching to the dish. They must be loosely attached in order to be able to pick them, but if they attach too much you won´t be able to get them into the capillary. Of course, the longer it takes for the picking the higher the number of cells that are missed. For some cells, even adding Tryple Express (mild tripsinization) doesn´t help.
In conclusion, I am not sure this is the best system but it seems to be better than other micromanipulators and it´s definitely cheaper. I am still skeptic it is worth the money we paid, though...
Best,
Simone