I'm repeating the experiment tonight with a few tweaks. I will post the results here tomorrow.

Short version:

I re-did the experiment with 1 mL of cell solution in a flat bottom incubation 24 well plate. Both serum+ and serum free samples lost about the same amount of cells and there was no clumping in either sample this time. However, I need to use a non-adherent incubation vessel because both samples lost way too many cells. Feel free to skip to my conclusions at the end if you want, but any advice you have on the details is greatly appreciated :)

The Long Version:

The blood was from a healthy volunteer under a technique development protocol for control samples. It was collected in a BD Vacutainer CPT tube with Sodium Citrate (Ref 362761) on 6/8/16 and processed within 2 hours following our standard protocol for freezing PBMCs in liquid nitrogen. The freezing solution is FBS with 10% DMSO and a cell concentration of ~10^7 cells/mL.

On 6/13 I removed 5 tubes (I was preparing for a different experiment, not expecting the cell loss) which were placed in ice. Two at a time, I quickly thawed the cells by rubbing them between my hand. With a small amount of ice left I slowly added 1 mL of PBS with 2% FBS and 1mM EDTA preheated to 37 degrees. The solution was transferred to a 15mL conical tube containing 8mL of the PBS. I divided the five cryotubes between two conical tubes. They were spun at 500g for 10 min at RT and resuspended in 1 mL AIM V (serum free).

I used a Countless II FL cell counter from Life Technologies for my counts. I'm used to doing manual counts, but this machine seems to be consistent as long as there's no dust on the slide. Plus, it gives the percentage of live and dead cells. The count after the thaw was 1.33 x 10^7 total cells with 94% live. I resuspended in 13.3 mL AIM V for a final concentration of 10^6 cells/mL.

I transferred 980 uL plus 20uL FBS to a single well in a 24 well flat bottom polysterene incubation plate. For the cells I expected to use, I transferred the rest to a 50 mL flat bottom incubation flask. I put them on a shaker (to attempt to prevent cell adhesion) in 37 degree incubator with 5% CO2.

The next day, after about 13.5 hours in the incubator, I removed the cell solutions from the incubator, poured the serum free sample into a 15mL conical tube and transferred some of each to separate tubes for counting. AFTER I removed some for counting, I checked the flask for cell adhesion and there were some cells on the bottom of the flask, so I used a sterile cell scraper to remove them, washed the flask in ~2mL of AIM V and transferred it to the serum free tube. This is one thing that happened between the first and second count that could have effected the out come, although I didn't think of that at the time, not expecting the cell loss.

The cell concentration out of the incubator was 1.02 x 10^6 cells/mL, 94% alive for the serum free sample and 9.38 x 10^5 cells/mL, 80% alive for the serum+ sample. The high death rate in the serum+ sample may just random pipetting error in the count. Again, I am new to the cell counter. Plus, since there was no apparent cell loss in the serum free sample, I wasn't concerned with the accuracy of the serum+ count. At this point, I assumed I was done with it.

I spun the serum free sample at 500g for 10 min at RT and resuspended in 1mL AIM V. For my second count, I kept getting really low counts, but was seeing what I thought was dust on the slide. So I repeated the count several times before getting frustrated and doing a manual count. What I saw was clusters of maybe 15-25 cells that looked like bunches of grapes. They were not necessarily stained with trypan blue, so they may have been alive at this point. I'm not sure. My final total cell count was 1.53 x 10^6 leftover from the approximate 1.2 x 10^7 cells that went into the incubator (87% loss).

At this point, I went back to my serum+ sample, which had been setting covered in the biohood for about 30 minutes, transferred it to a 1.5 mL tube, spun at 500g for 10 min at RT and resuspended in 1 mL serum free AIM V. I did a manual count and did not see clusters, but the count was 7.06 x 10^5 which is a 24.7% loss from the pre-wash count (9.38 x 10^5). I expected the wash to remove some dead cells.

At this point, I couldn't exactly say that it was the absence of serum that caused the post-wash clumping. I thought there might be something in my handling of the serum free sample that was the culprit. On to experiment two...

On 6/15 I thawed one cryotube from the same blood draw as before and thawed it in the same manner, washed with PBS 2% FBS 1mM EDTA, and resuspended in AIM V. I used the cell counter once again and counted 2.88 x 10^6 total cells, 96% live. I adjusted the concentration to 10^6 cells/mL and transferred 1 mL to a single well in a 24 well incubation plate (same as before), and 980uL plus 20uL FBS to another well. I put them on a shaker in the same incubator as before.

The next day, after ~15.75 hours, I removed the cell solutions to 1.5 mL tubes. I pipetted up and down to try to knock off adherent cell, but still saw some on the bottom of the wells after removal. I did not attempt to scrape or wash them off. The count for serum free sample was 5.44 x 10^5 total cells, 80% live (45.6 % loss). The count for serum+ sample was 6.73 x 10^5 total cells, 81% live (32.7% loss). I spun at 500g for 10 min at RT and resuspended in AIM V. The final count for serum free sample was 4.04 x 10^5 total cells, 81% live (25.7% loss from pre-wash). The count for the serum+ sample was 5.0 x 10^5 total cells, 87% live (also 25.7% loss from pre wash).

Conclusions

• The serum doesn't seem to make a difference when the samples are treated the same. 
• The clumping may have been a fluke, or may be due to the large total sample volume. I don't know why it would cause clumping at that step and not before though. So maybe something happened that escaped my attention.
• The adherence to the bottom is not eliminated by shaking as I had hoped. Nor does FBS seem to make a difference. This is troublesome because monocytes are known to adhere and they are known to express my genes interest in response to IFNy. I don't want to lose any of these cells after rest so I'll have to manage this in some way.

Thank you to Dr. Sylvia Janetzki who sent me advice and her protocol for thawing and resting cells, which helped me spot problems and areas for improvement in my methods.

Bart

For ELISPOT and flow cytometry I used Benzonase Nuclease after thawing. 50U/mL DNAse (Benzonase® nuclease, Novagen) as suggested by Smith et al. 2001 in "Development and Validation of a Gamma Interferon ELISPOT Assay for Quantitation of Cellular Immune Responses to Varicella-Zoster Virus". They recommend the use of Benzonase to prevent cell clumping when samples are thawed.