Suggestions for optimizing a multiplex PCR? - Why do my bands keep disappearing in the positive control? (Nov/11/2009 )
Hi All,
I am trying to use molecular methods to study bat diets, by amplifying insect DNA from bat feces. I am trying to set up a multiplex rxn to screen for consumption of three particular insects. I have designed species-specific primers for these three insects (a beetle, the primers produce a 495-bp product; a moth, the primers produce a 320-bp product; and a katydid, the primers produce a 250-bp product).
My primer pairs all work fine on their own (although the primers for the katydid are suboptimal and want to dimerize a bit; at this point, I haven't found a way around it while still keeping the specificity to the target). All the primers have Tm's between 55 and 60 degrees, GC content 40-60%. And, as I mentioned, each primer pair on its own works well to amplify DNA from target species and not from various other insects that I have screened.
Anyway, I would like to combine these into a multiplex rxn. I got a 100 rxn Multiplex kit from Qiagen, which uses their HotStart Taq in a master mix, and an optional mystery product called "Q solution" which does seem to reduce the production of non-specific products in my testing.
I started out by testing this by combining 50 uL from extractions of each of my three target species to make a positive control mix, and started testing out the multiplex with an equimolar mix of my 6 primers. I ran a few rxns with varying amounts of this template, at 57 C as suggested by Qiagen when one or more of the primers has a Tm of less than 60. (No gradient thermal cycler in my lab, boo hoo.) Template amount didn't make much difference, but what was weird was that I got a beetle band at 495 and a katydid band 250, but no moth band at 350.
I figured maybe I used the wrong moth extraction (i.e., accidentally grabbed an extraction from another moth I had extracted while testing specificity of my moth primer set) so I made sure I got the "right" moth, and added another 50ul of this to my master extraction positive control. I set up another couple of reactions with this template, one straight up, and one with a little extra of the moth primer (this is the one that wants to dimerize much more than the others). Spaced and ran them at 60C instead of 57C.
So now, when I ran out the product on the gel (from both reactions), the 320-bp moth product is there. The 495-bp beetle product is there. But now the 250-bp katydid product is missing... and it was there before!!!
Any suggestions on which next steps to take in optimizing this reaction? There are so many possible variables to adjust (template amount, primer concentration, annealing temp and time, yadda yadda) that I'm not sure where to begin. Or, since there are only three species I'm screening, should I just shrug my shoulders and run the reactions separately for each primer pair? Any suggestions appreciated, I'm a first-timer. I've uploaded the Qiagen multiplex handbook, which I have been following thus far.
Unless you are doing hundreds or thousands of samples, I would do separate PCRs in low volume. Almost certainly the answer is to redesign the primers, rather than to optimize conditions.
You'll have more than enough problems dealing with DNA extraction from feces and and with all of the other DNA present in that mess.
phage434 on Nov 11 2009, 02:28 PM said:
You'll have more than enough problems dealing with DNA extraction from feces and and with all of the other DNA present in that mess.
Thanks -- I think that you are quite right.
I used Multiplx to check if my primers (and the amplificated DNA) are compatible and which of them can be used together in a multiplex PCR...it's not bad and the only software I found so far...
Anyway for three primer pairs it should be manageable, but don't use too much time and effort on it, here I agree with phage434.
There are so many other obstacles in such an approach (lots of PCR inhibitors; degraded DNA, mispriming, etc) that I'd focus first on the topic that the primers work well under real conditions i.e. on the more or less degraded DNA-mix out of the faeces.... also have a look on closely related insect species that might be prey of the bats and how long the DNA is good for PCR, as a missing band for a species might come from too degraded DNA or because the species was not feed on at that time....
Here is a paper about multiplex PCR optimisation...also a good source for information:
hobglobin on Nov 12 2009, 01:01 PM said:
Anyway for three primer pairs it should be manageable, but don't use too much time and effort on it, here I agree with phage434.
There are so many other obstacles in such an approach (lots of PCR inhibitors; degraded DNA, mispriming, etc) that I'd focus first on the topic that the primers work well under real conditions i.e. on the more or less degraded DNA-mix out of the faeces.... also have a look on closely related insect species that might be prey of the bats and how long the DNA is good for PCR, as a missing band for a species might come from too degraded DNA or because the species was not feed on at that time....
Here is a paper about multiplex PCR optimisation...also a good source for information:
Thanks for the link to Muliplx -- it looks really helpful! And for your other suggestions. I had tested my primers only on extractions from insects, not from extractions from bat feces (because I have no way to know which bats might have eaten species X). I'm getting some wonky bands for a primer that I thought was "good" so it may be back to the drawing board, or oligo calculator or whatever.