Inability to duplicate PCR results - (Sep/24/2009 )

Hi,
I'm pretty experienced with PCR, but recently have had trouble duplicating my results. The PCR assay works very well with DNA extracted from blood, however, when I use DNA extracted from arthropods I have trouble duplicating my results. We are trying to detect an organism that may or may not be present in each of the different sample types. An example of my latest set of assays from arthropods: Assay 1) Amplification in 10 of 60 samples. Assay 2) Amplification of 4 of 60 samples, but none the same as in assay 1. The 2 assays were run one day after each other and to the best of my knowledge everything was the same (protocol, reagents, thermocycler). Some samples have high weight molecular bands, but these too are inconsistent between assays. The positive control (DNA from blood sample) shows up fine in the assays. Any thoughts?

Thanks,

The sequence for the priming site is different/absent in arthropods? There are less inhibitors in the blood DNA?

your target organism is present only in small amounts compared to the host? The reason might be a simple stochastic effect: you are having less DNA per µL used for PCR than the detection limit of your PCR assay.
Like, lets say you can detect one molecule of target DNA. You are using 1µL DNA extract for PCR. But your target DNA is present only one time in 10µL DNA extract. So you will only get one positive result in ten PCR assays even with the same conditions.....

You should test this effect with running lets say 10 paralells of some of your positve samples, look how many are postive and determine the amount of paralells you will have to run for all your 60 samples to get robust data.

Are the bands on the gel looking nice? I guess you are looking for arthropod genes, maybe the primers have a design flaw and don´t actually recognize the arthropod DNA, but bind weakly (and not very reproducably) to human DNA

gebirgsziege on Sep 25 2009, 12:44 AM said:

Thanks for the replies everyone.

Gebirgsziege - I've been thinking along the same lines as you, but wanted to see if anyone else had any ideas. All samples get run twice, anything with a discrepent result a 3rd time. I guess I may have to run more trials as you suggested.

I have to say Gebirgsziege's solution seems unlikely to me, presuming you are using whole arthropods you will have many many copies of any target DNA floating around in the solution. This isn't to say that the target isn't below the detection limits of your PCR (in which case try using a different polymerase e.g. KOD or Amplitaq Gold), but in general a PCR should be capable of picking up any single sequence that is there, usually within 30 cycles.

bob1 on Sep 26 2009, 12:33 AM said:

You are right, if working with artopod genes and whole artorpods the dilution is unlikely.
But when I understood stoffelrt right he wants do detect something like a parasite, which is present not always and when present only in small amounts. so what I was refering to: lets say 100 bacterial cells in the whole artorpode. When working with "systems" like this you must not forget this effect; as too high DNA concentrations will effect your PCR and you have lots of artropod DNA compared to bacterial DNA. So even if you might have enough DNA to detect the bacteria alone (i.e. in water) this might not be sufficient DNA to detect in the heterogenous system artorpod-bacterial DNA.

Maybe you have inhibitors in your DNA? What are you extracting from?
I was having trouble PCRing from individual spidermites - DNA extracted in 20ul, 2ul used for PCR.
ended up having to dilute them in 200ul for PCR to work...

Remember that PCR really is one of those experiments where less can be better. Dilute 1/10 and repeat if you're concerned about inhibitors.