PCR Trouble Shooting - new primer, wrong dilution of primers, wrong band, and many more (Jun/13/2007 )
I have to admit (like many do), I don't know much about PCR but I am required to do one. I had the protocol so I used to do well and had no problem till now. But, for another gene we got a new primer and we had a literature telling us the protocol. So, we started.
But the problem started.
- the one who diluted the primer we got from the company did a slight miscalculation and made wrong dilution. He under diluted two primers and over diluted one. I ran PCR as they were but I got a very light band of around 1000bp while I wanted 750. The negative controls were blank.
- I then corrected the dilution calculating the molar mass but still I got the same picture. Negative controls were blank and the sample showed light band of wrong length.
- I changed annealing temp for a big range 55 to 70 but still didnt get the correct band. either there was no band or very weak band at 1000bp.
So, heres what I wanted to ask
- the protocol runs 40 cycles .. . isn't that too much? I think the amount of product plateaus at around 30 to 35 cycles. What do you say? I read that half-life of Taq is just 30 mins at 94 deg C. So, if too many cycles then it becomes inefficient.
- the length of fragment I want is 750bp. The protocol recommends 90sec of annealing at 55 deg cel. while the extension step at 68 deg cel. is 60 sec. Should the annealing time be so long and isn't extension time short?
40 cycles is too much, you will out of linear range. I don't go above 35, if I have no choice then 37
The annealing time sounds fine, although I usually do 60sec for both annealing and elongation
I think that the anneling is to long. If the product is 500bp or less with an anneling of 30-45 sec will be fine, so for a 750bp the anneling should be around 50-60sec. To much time of anneling will give oppootunity to polimerase to extend to much so thats the possible reason for the 1K product. Check the Tm of primers to make sure that 55 is the correct temp. Because wrong anneling temp will give unspecify products. For extension 72C for 60sec.
Ok, I tried reducing the cycles to 35. Still, didnt work.
I need to screen some transgenic mice and I got the protcol from a paper.
For the Wild type gene, it worked well throughout the whole range 55 to 70 deg C. and with 40 cycles and 35 and also 30 cycles.
But for the Knockout gene (pGK Neo), it does not show band and when it shows it is higher. Wanted 750 but shows 1000 or above.
Tm for forward primer is 85 and reverse primer if 75.6.
PCR Prog (as per the paper which worked for them) - 94 deg for 30sec; 55 deg C for 90sec; 68deg C for 60s for total of 40 cycles.
I have till now tried Annealing temp changed to 58, 64, 68, 70. at 55 and 58 it showed the wrong band while in others there were none. The problem is further complicated by wrong dilution of the primers. I was using 2.5ul (10pmol/ul) but now have to adjust that to 2.47ul and 1.98ul for the two primers. I wonder if not adjusting will make any problem.
So, next I intend to try by changing the annealing and extension time. Reducing the cycles made the bands for the wild type gene sharper.
So, any more suggestions. Please keep them coming.
Every PCR is a little different. In this way, when you get new primers especially, a bit of optimization is required. I know it is hard, but just step back a second and think about it:
PCR is nothing more than an enzyme amplifying a region of DNA. Sure, there is an "optimal" mix of reagents, but generally you need some primers, some template, an enzyme, and some buffers. If you have all those things, then it's possible to make it work. Unless you are adding far, far more of one component than the others, we can make it work (people used to use waterbaths for pete's sake!). If you think in this way, it's a little easier to rule out things like changing the number of cycles, as that isn't really affecting the reaction from taking place (or helping it).
Make sure your primers at the concentrations you calculated truly are equal. I would run some multiple of your reaction amounts on a gel to compare ratios. You should have two bands that are of equal intensity. It *is* important to have equal amounts of each primer. Adjust the amounts until this is the case (on your gel). I wouldn't fret about the amount of primer you add as long as they are equal. Primer dimers may form, but you should still get some product.
If you are amplifying from genomic DNA, you only need 1uL normally in a 50uL reaction, from plasmid DNA, I use 0.5uL. Concentration isn't the most important. It's like cooking.
Then, take a look at the primer design. What is the annealing temp if you only consider the portion designed from template (excluding restriction sites, etc.)? Here is a good calculator to make it easy: http://www.idtdna.com/analyzer/Applications/OligoAnalyzer/.
I assume you have used the Taq, buffer, and dNTPs before (recently) with no problem -- that was the positive control you mentioned, right? If this is the case, now come up with your protocol. If the temps you found from full length vs. initial annealing portion are very different, I would do 4 cycles at the lower temp, then 25 at the higher temp. Too low is okay as far as temperatures, too high is not.
Finally, how old is your PCR machine? The newer ones go between temperatures very very quickly, leaving little to no time for error on temps. The old ones take their time, so if your annealing temp is 60 for example, your tubes spend some time at other temperatures as it is heating and cooling. If ever I am having trouble with a PCR, I use the oldest machine possible to give myself a little cushion. 
Hope that helps!
Amanda
If the primers worked with the wildtype DNA, maybe you need a pcr enhancer for the KOdna. could try betain.
Thank you for all your suggestions. They were helpful but still I am not able to get the band for KO gene (pgKNeo). But, it's a little early to consider new primer as I have not yet done all that are done with not working PCR. So, I am holding on. Today I was busy with my other work but found time to run one more try and keeping all the conditions same, ran at annealing Temp 53. Again, the Wild genes are wild and bright but the KO is no where seen.
Amanda, you really raised my hope. You have such a convincing writing and it made me feel as if I can do it. You explained everything in such simplicity. I had never understood PCR so well.
I am not sure if I have the primers of equal amount. So, would like to do what you suggested but I am unable to understand how that is done. Please, can U explain. Well, the primer pair for the Wild is working well.
Amanda, you really raised my hope. You have such a convincing writing and it made me feel as if I can do it. You explained everything in such simplicity. I had never understood PCR so well.
I am not sure if I have the primers of equal amount. So, would like to do what you suggested but I am unable to understand how that is done. Please, can U explain. Well, the primer pair for the Wild is working well.
I'm glad it helped (even if no product yet!). Since you have wildtype working well, I'm not sure primer inequalities is truly the problem, but certainly needs to be optimized anyway. Here's what I would do:
Take the ratio of primers you have been using (I forget what you said, but say 2uL of one and 7uL of the other for example's sake). Since it is possible you will not see this amount on a gel, I'd multiply both by 5 or so to get enough to visualize (and keep the ratios). In this way, you'd run 10uL of one, and 35uL of the other (this is a lot -- I'm pretty sure you aren't using this much). Then, you should see bands. These need to be equal. If one is much brighter than the other, play with it a bit until they're equal. Then simply divide by a multiple to get an amount you can actually use in a PCR reaction. Does that make sense?
I would strongly suggest doing the few cycles at the annealing temp of the portion of the primer which does not include restriction sites and added sequences first (from that calculator), then the remainder at the annealing temp of the whole primer. In this way, you should cover multiple variables in one reaction.
Hope that helps!
Amanda
Hi Nabin. Just to muddy the waters a bit more, here's my 5 cents worth (you can send me a cheque if it works )
If your Tms are correct (and there's no reason to suppose they aren't) I'd anneal at 68 and extend at 72. You don't want to go too far below your Tms or you'll start getting non-specific priming.
Is there any chance you could make the forward primer a bit shorter to bring the Tm difference below 5C? If you need the primer as is (say, you have an RE site at the 5'end) ignore those bases in your calculations for Tm, as they'll only come into play later on, when your product is the major component (by which time there won't be anything of significance to misprime from).
Last of all, I'd try the old touchdown PCR, starting the annealing from 75 degrees, go down 1 degree every two cycles, for the first 15 cycles, then use that annealing temp for the rest of the expt (max cycles 35).
Good luck.
Pipette. Cry. Repeat. No, dammit, I will not be defeated by some silly enzymes, sugars, aromatic rings and phosphates!! Instead: Pipette. Conquer. Rejoice.
Cheamps and Swannny,
Thank you. Your suggestions are making me understand many things I had never thought about PCR. I had other programmes in the weekend so could not be concentrated on this though I tried some of them.
The primers didn't give equal bands first and I had them adjusted till I got almost equal looking bands.
The calculator is cool, Cheamps. I was relying on the data sheet of the manufacturer. The forward primer (GALFDJH) was supposed to have Tm at 85deg and reverse primers (NeoU and BGLGAL) were supposed to have them at 75.6 and 74.1 deg C. The annealing temp the paper I referred suggested using 55 deg C as annealing Temp and was wondering why so much difference. But, when I calculated with the calculator I found them to be 68.5, 60, and 55.4 deg cel. Which should I be taking as the true Tm? Why is there so much difference between the one provided by the manufacturer and the one that was calculated?
Another question (may be very basic question) - how do I find the sequence of primer without the restriction site or additional site?
I tried Swanny suggested Touch down PCR but I got it all mixed up and ended in a mess with primer-dimers. I will try fixing and running again. But, may be after a few days as I have a presentation to make today.
Thank you for all your help.