I can confirm that hTERT RPE1 cells are puromycin resistant as PAC is expressed on the plasmid that contains the hTERT gene. We knocked out the PAC using CRISPR and our RPE1cells are now puromycin sensitive and can be killed in 2 days with 2ug/ml puromycin. Prior to this we would have to treat with 5-10ug/ml puromycin for a long time to kill RPE1 cells.

Hi Alice,

puromycin is usually very powerful, it is difficult to be resistant for any eukaryotic cell. It acts already at a concentration of 1 ug/ml, but you can go up to 10 ug/ml. In 2, max 3 days, practically all your non-resistant cells should be dead (so, if you start the selection 24 hrs after transfection, by 48 hrs after transfection your dish should start filling with dead cells and by 72-96 hrs all cells should be dead except the transfected ones... but it might take another few days until you see the survived cell foci).

If this does not happen, my guess is that the puromycin stock you are using is not good (anymore). Being a nucleoside analog, all rules valid for nucleotides apply:

- make your stock solution in a neutral buffer (typically Hepes, pH 7-7.4) (pH sensitivity)

- make small aliquots of your stock solution because freezing/thawing cycles degrade it

- make your stock at a conc. of at least 10 mg/ml, because freezing more concentrated helps preserve the molecule

In addition, some procedural aspects:

- when you thaw the puromycin, let it equilibrate well at rt and mix thoroughly before you add it to the medium, since it tends to precipitate at low temperature (reducing your effective conc.)

- add the puromycin to low density cells... antibiotics penetrate into eukaryotic cells mostly during mitosis, so if your cells don't divide, the antibiotic is much less effective. If your cells need to grow dense, find a compromise. After transfection try to seed them around 30-40% confluency. remember that they will divide once still without puromycin, and you have to leve space for them to divide again when you start adding the antibiotic after 24 hrs...

- in solution, at 37 C and with the acidification of your culture medium due to cell metabolism (when you see it turning yellowish, if you have the phenol red in) puromycin will probably not be that stable. So I would change medium every day, at least at the beginning when your dishes are still dense, and I would add the puromycin fresh  to the medium every time (with G418, being a trisaccaride, you can prepare the medium in advance and keep it for months due to its very high stability, but puromycin is a different molecule!).

Don't know if this will help, anyway, good luck!

I had a similar problem.  The first thing to do is make sure you can grow wt non-transfected single cells from your cell line you are trying to do the CRISPR in.  Don't spend a lot of time doing the CRISPR in this cell line if you can't grow the single cells to begin with.  If you cannot grow single cells you have the option of doing the puro experiment and using clonal rings. Keep in mind it is difficult to really verify you have a single cell to begin with just by using microscopy.  

For the puro resistance part of your question, I would suggest doing a dose response until you kill all of your cells. If you suspect a bad batch of puro try dosing cells that you know are susceptible.  Good luck!

Pietro and Julie make excellent points. My first suspicion would be puromycin quality and you do need to change the medium every day for it to work.

However some additional points:

Some cells really don't grow well without having lot's of company. You can try to help your diluted cells by using a 50:50 mix of conditioned filtered medium from non-transfected cells with fresh medium (filter to avoid transferring in the control cells and make it fresh every time). a higher FBS percentage could also help. Some cells need an extra rich medium to survive "loneliness".

I've experienced similar issues when making HepG2 cell lines and the conditioned media in mix with extra super duper medium did the trick.

The other thing you could try is to transfect your cells in the usual high density then tryspinize and plate out at dilution 24 h later and start your selection 24 h after that. Selection 24 h post-transfection seems slightly early as you need the transfection to be complete (when using lipofection plasmids need cell division to get into the nucleus) as well as expression of your resistance gene. I usually give it 2 days before starting selection.

In regards to Cell sorting: you could try and sort lots of cells into the same well and give them two days rest before diluting. That way you would spread the stress of sorting and low density over a few days. Also, discuss your non-survival issues with the cell sorter operator. Increasing nozzle size and reducing pressure can make a difference in cell survival.

Thank you for your advice!

Also keep in mind that if you are doing standard transfection, CRISPR/Cas9 only needs transient expression to introduce the mutation. So unless you are using a system that introduces the puromycin resistance gene at the mutation site, treatment of puromycin for long term will result in selection of stably transfected cells and resistant cells. Stable expression of CRISPR/Cas9 may have deleterious effects since it could continuously mutate the target site and have accumulation of off-target effects, both of which could lead to confounding effects on the cells.

Have you checked if your RPE1 cells carry Tert gene used for immortalisation (this could have already been selected using Puromycin)?

I just experienced the same with RPE1... And the batch of puromycin works fine on other cells. Did you solve the problem? And if yes, at which concentration?

RPE1 have been immortalized collowing hygromycin B selection, no puromycin, so I don't think this is the problem...

The hTERT plasmid used for immortalisation of RPE1 cells contains genes for both Hygromycin and Puromycin resistance (http://www.atcc.org/~/media/PDFs/Culture%20Guides/hTERT_Cell_Culture_Guide.ashx). PAC gene on the plasmid is coding for Puromycin N-Acetyltransferase.

Thanks Rilla! then it makes sense!

I thought RPE1 were immortalised using hTERT carried by a Hygromycin resistance... I managed to "select" hTERT-RPE1 using 20 µg/ml puromycin. I know that's a lot but I found it the litterature. hTERT-RPE1 cells are extremely sensitive to G418 and Blasticidin unfortunately those antibiotics don't act quickly

see my answer to the same question from 24 June. G418 selection works very well; it takes a couple of weeks to select in 0.6mg/ml.

Hi there,

I found the RPE-1 cells were made by this plasmid: pGRN145 hTERT. It only has Hygromycin resistance, but no puromycin. This is the map of the plasmid.

Have you fix the puromycin selection problem?

Thanks!

I just noticed "Resistance to puromycin is conferred by the pac gene encoding a puromycin N-acetyl-transferase (PAC)"

That's why PRE-1 shows puromycin resistance.

........

I've used puromycin at a higher concentration of 7ug/mL  in RPE-1 cells. This is a much higher concentration than I would use in other cells. 

Hi RPE1 users,

I have developed a practical method for gene knockout in RPE1 cells.

I have deposited plasmids required in this method to Addgene. (https://www.addgene.org/browse/article/25223/)

If you are interested, please try it.

Article Practical method for targeted disruption of cilia-related ge...

Hi Alice,

What do you mean by optimised FACS protocol which has proved successful ?

Could you please upload some details?

Thank you!

I can confirm that hTERT RPE1 cells are puromycin resistant as PAC is expressed on the plasmid that contains the hTERT gene. We knocked out the PAC using CRISPR and our RPE1cells are now puromycin sensitive and can be killed in 2 days with 2ug/ml puromycin. Prior to this we would have to treat with 5-10ug/ml puromycin for a long time to kill RPE1 cells.

Pietro and Jessica made excellent complementary points.

I would also like to confirm that hTERT RPE cells have a high tolerance for puromycin. I did kill curves with a new puromycin batch, and needed 10-12 ug/mL puromycin to kill 100% of hTERT RPE after 5 days, when the same batch killed 100% of LNCaP (prostate cancer cells) at 2 ug/mL after 3-4 days.

Could you kindly share your optimized GFP protocol with FACS? I have had problems with GFP+FACS and single-cell survival for RPE1 cells as well, although my other cell types survive fine.

So what I did to overcome the problem of single RPE1 cells not surviving after FACS was the following:

• I bulk sort the brightest 10-15% GFP+ CRISPRed RPE1 cells, 2000 per well in a 24-well plate.
• Important here to use 50% spent medium as recovery medium (meaning 50% sterile filtered used medium from a confluent culture and 50% fresh)
• I let cells recover for a few days, to double up.
• Trypsinize the cells, and plate in 10cm dishes at very low density after good resuspension (to avoid cell doublets), to allow colony formation.
• When clones have grown enough (after 2 weeks or so), I pick them by mechanically scraping off with a 20ul pipette and transfer clones in 96-well plates.
• Again feed clones, up when they are ripe for duplicating in a copy plate.
• Use the copy plate for DirectPCR lysis and proceed to genotyping.
• I have successfully created 6 different RPE1 KO lines this way.

Depending on gRNA efficincy, I had to screen from 20 to 100 clones to get a homozygous deletion.

I hope this helps,

Best,

Chronis

Hi Andrew J Holland,

KO the PuroR gene sounds like a great approach. I will like to do that too. Can I ask for the sequence of your gRNA?

Thanks.

Best,

Arne