hi,
i created this device which can mechanically compress cells. I did immunostaining of the cells and observe some apoptotic nuclear fragmentation of cells after compression.
I was thinking of the ideal way to statistically quantify apoptosis. What kind of statistical analysis should i do? I was thinking of looking at 200 Dapi stained cells in control and compressed cells and quantifying proportion of cells exhibiting nuclear fragmentation. But what kind of analysis must i do to say that the proportion seen in compressed cell is statistically higher than in control?
i am also looking at caspase 3 western blot. This part is not really an issue.
To measure statistical significance, ideally you would do multiple rounds of compression or non-compression on different sets of cells, on different days (biological replicates) and check the frequency of each biological replicate; you could then use a simple statistical test like Student's t-test to determine a p-value given the magnitude and variance of your non-compressed and compressed samples. This is a very simple metnod for comparing a single treatment, i.e. two samples.
You can extend this analysis if you want to e.g. try multiple different cell lines, or multiple different pressures or the like, but you'd be starting to move into ANOVA (Analysis of Variance) territory there, and you'd have to start compensating for multiple comparison's (I'm not great on this side of things, but I think a Bonferroni's test or similar could be a way to go?).
If you're looking for greater total counts and greater biological specificity than you get with the microscope, you could always lift the cells, fix them and stain with propidium iodide and annexin V-FITC - you can run these on the FACS and cound multiple thousands of cells in a few seconds, as well as gating on the live (PI- AV-), early apoptotic (PI- AV+) and late apoptotic/necrotic (PI+ AV+) populations.
I hope that all helps!
Regards,
Dan
"To measure statistical significance, ideally you would do multiple rounds of compression or non-compression on different sets of cells, on different days (biological replicates) and check the frequency of each biological replicate; you could then use a simple statistical test like Student's t-test to determine a p-value given the magnitude and variance of your non-compressed and compressed samples. This is a very simple metnod for comparing a single treatment, i.e. two samples"
Thanks for your insight! Regarding the above, what kind of observation should i look at? Is looking at nuclear fragmentation by DAPI good enough?
Hi Iwan,
I don't have a huge amount of experience with nuclear fragmentation, so you're probably better placed to assess its efficiency as a method than I am! Ultimately, the statistical approach above works whatever sampling method you use - as long as you get frequency measures from multiple biological replicates, it doesn't matter how you physically determine that frequency!
For apoptosis assays, I usually use the PI/Annexin V-FITC method I mentioned above, and run it on FACS. This allows me to count many more cells than I could by eye down a microscope - I usually count 10,000 for each sample, in about 30 seconds - and that means that I've got a much more robust sample, which is less prone to sampling biases etc.. It also allows me to identify cells that have died by necrosis instead of apoptosis. There are lots of commercial kits around to do this labelling (I won't recommend any particular one publicly, but can send you protocols/methods directly if you'd like to message me), and it is a very simple protocol.
There are of course other methods out there for doing apoptotic labelling - if you're comfortable and confident in your method, go for it!
Cheers,
Dan
Hi Iwan,
As part of my PhD, I looked at apoptosis by staining. Cleaved caspase 3 (CC3) is a good marker of cell apoptosis (antibody from Cell signalling was good).
Two ways of doing it, PI/Annexin V-FITC method via FACS or CC3 staining and imageJ counting.
DAPI staining for nuclei fragmentation isn't enough. As positive control, use Staurosporin, then you could assess CC3 staining with DNA fragmentation.
If you are ambitious, simultaneously with FACS, you could perform a SCGE (CometAssay), to look at DNA fragmentation?
Hope my answer completed the others. Good luck.
Cheers
Stéphane
If even this could be the case, you must ask yourself if this nuclear fragmentation is really apoptosis or something else due to mechanical stress because it takes few hours before observation of nuclear fragmentation during apoptosis. Did you observe this fragmentation during compression or few hours later? How long did you submit cells for compression? One easy way for checking global cell cytotoxicity is to used trypan blue. If apoptosis is engaged, you should also observe cell death.
One question : have you ever check for degree of microtubule acetylation during your mechanical stress ?
Regards,
Daniel
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Cell Biology