4
$\begingroup$

I harvested some lentivirus from 293T cells and want to titre the result. I infected 293T cells on a well plate with 400,000 cells per well which I infected with virus stock, and 1 in 10, 100 and 1000 dilutions (as well as a few uninfected wells). After 72h incubation I trypsinised the cells and used FACS to titre first, but 293T do not express the promoter under which GFP is located in some samples and hence these do not appear on FACS. As an alternative, I extracted the genomic DNA from all samples using a QIAGEN DNeasy kit (also the samples which I was able to titre by FACS already) and then performed qPCR on them.

I used the ABI PRISM 7000 SDS: http://www.cgenetool.com/products/abi_prism_7000.shtml

Included on the PCR plate were:

  • Number standards: 1000, 10k, 100k, 1m copies
  • Samples in triplicate
  • No-template controls as well as untransduced controls

Each of those was doubled, once with primers and probe for WPRE (an element specific for the lentivirus) and once with primers and probe for beta-actin (a housekeeping gene present in all cells).

The results returned by the machine include for each well the cycle at which the threshold fluorescence was crossed - along with the automatically calculated quantity of copies (which the software calculates from the standards I assume).

Since I named the triplicate samples identically, it also automatically calculated the mean quantitiy among each triplicate (practical, eh?). I checked if there were any outliers which should have been excluded from the means but everything was fine.

So now for example I have a result of 700k for the quantity of WPRE in sample A. The quantity of beta-actin in sample A was 6.82 x 106. The transfection was 400,000 cells per well, sample A was transfected with 10uL of viral stock diluted 1:100.

How do I calculate the virus titre per mL from this?

I have been given the formula: WPRE Qty / (b-act Qty*0.5) * cells in well (400k) * dilution factor (100), but am struggling to figure out a) why b-act Qty is divided by 2, b) whether the result would be virus titre per uL or per 10uL (transfection volume) and c) how beta-actin can have almost 107 copy numbers consistently across all samples when there were only supposed to be about 105 cells...

| improve this question | |
$\endgroup$
1
$\begingroup$

I have 2 doubts here.

  1. You obtain your viruses from the media whereas Actb measurements can be performed only in the cell lysates.
  2. Actb is just a reference to normalize for the cell count: higher the cell count higher will be the viral titre because of higher number of infected cells. But that is immaterial if your aim is lentiviral transduction where you are only interested in obtaining titre values and not measuring the efficiency of its formation.

You can simply do qPCR with the media sample and the standards and regress the measured Ct values of the samples to obtain titres.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks for the reply. I see that I left out relevant information so I added that - have a read of the first paragraph again and let me know if that makes more sense now :) $\endgroup$ – Armatus Sep 5 '13 at 13:11
  • $\begingroup$ I think it makes more sense to qPCR the lysates for copies, with all virions removed, as that should give me an idea of how many cells a mL of virus stock can transfect, which I believe is actually the important factor - I don't think I really need to know exactly how many viral particles that mL contains in order to transfect that many cells. Does that make sense? $\endgroup$ – Armatus Sep 5 '13 at 13:13
  • $\begingroup$ I think it is the media that will contain the viruses and that's what you are going to use for transfection post concentrating the virions to whatever concentration you want. So you ought to do the qPCR with the media right? I might be mistaken so please let me know if that is so. $\endgroup$ – WYSIWYG Sep 7 '13 at 9:13
  • $\begingroup$ The initial prep is 293Ts in a medium with the viral plasmid as well as packaging and envelope plasmids; cells are perforated (but not killed) by adding PEI which leads to production of functional virus. That's then harvested, aliquoted out into stock vials and frozen. I was told it's not possible to just directly qPCR the aliquots - I assume the viral proteins are the problem. I used one of the vials to infect cells with decreasing concentrations of it, extracted their DNA after three days and qPCR'd that in order to analyse how many viruses actually integrated into the cellular genome. $\endgroup$ – Armatus Sep 7 '13 at 20:21
0
$\begingroup$

Your method seems overly complicated to me. Surely you have a control virus? I would recommend a constitutively expressed GFP construct to make it compatible with FACS analysis. Use the GFP virus to directly calculate viral titre by FACS, then verify your other viruses by plaque assay.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Yes, one of them was a constitutively GFP expressing control, and plaque assay would have probably been a lot easier but wouldn't have been possible due to various practical issues (ironic really considering how complex qPCR is...) $\endgroup$ – Armatus May 4 '14 at 13:13
  • $\begingroup$ I'm still not clear why you are trying to titre viral pfu using the manufacturing system (HEK293T cells) rather than your intended experimental system? Infectivity is certainly a function of host cell, but only makes sense in your experimental system. I guess my question is, why is a plaque assay system that difficult to use? Perhaps I have not understood some nuance of your particular system. $\endgroup$ – user560 May 4 '14 at 14:26
  • 1
    $\begingroup$ Haha, don't worry, there's nothing like a particular nuance about the system that would have been a problem; the issues were on a much higher level of practicality :) $\endgroup$ – Armatus May 5 '14 at 12:51
0
$\begingroup$

Having gained a lot more experience with how research works, I can actually answer myself now:

a) beta-actin quantity is divided by 2 because it is a human gene and hence present twice in mammalian cells, dividing it by 2 gives an estimate of the number of cells present.

b) The result gives an estimate of how many viral copies will end up in cells from the amount of virus solution used - thus if it was 10 uL of a 1:100 dilution this represents 0.1uL of neat virus solution. So the result's unit in this case would be "viral infections per 0.1uL".

c) The number of beta-actin copies doesn't correspond to the number of cells initially seeded. 400k cells which double once per roughly 24h would make 3.2m cells after 72h (3.2x106) so that's not all too far off the copy number.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.