Expression fold is too high (qPCR)

4 posts / 0 new
Last post
tigerlily
tigerlily's picture
Expression fold is too high (qPCR)

I am doing qPCR for osteogenic, adipogenic and chondrogenic genes using human MSC samples. I am using beta-actin as my reference gene and undifferentiated MSCs as my calibrator sample. When I normalize the data to the reference gene and calibrator sample, I get very high expression folds, sometimes up to 50 (using Pfaffl's method).

I am unable to figure out what is wrong as the expression of these genes should not be that high compared to the calibrator. There does not seem to be anything wrong with my qPCR as each melting curve shows only one peak, my NTCs show negative amplification (or very late Ct values), and the Ct values of my samples are between 15-30.

Before doing qPCR, I did RT-PCR and ran the samples on agarose gel. The band sizes were as expected.

What could be the possible causes for the abnormal increase in expression?

Ivan Delgado
Ivan Delgado's picture
Hi tigerlily,

Hi tigerlily,

There are a number of things that could be causing this. Here are some thoughts: 

1. Did you optimize both of these assays? If so, what efficiencies did you get? If the efficiencies of your assays are off by as little as 1 (99.0 efficient versus 98.0) you could get variations like this one. 

2. Did you test more than one reference gene? While actin is an acceptable reference gene, the fact is that reference genes vary greatly and may show expression variations in your cell lines. My recommendation is to test at least two additional reference genes to see if you get the same variation you are getting. My guess is that you will not.

Good luck

tigerlily
tigerlily's picture
 Hi Ivan!

 Hi Ivan!

Thanks for your reply! I am attempting qPCR for the first time so help is much appreciated!

I did calibration curves for each gene. The efficiencies did differ (e.g. 97% vs 93%) but I thought that this would be alright with Pfaffl's method of calculation. Do I have to make sure that the efficiencies are the same? 

I tested the assay with GAPDH and obtained similar results. 

Perhaps my RNA samples were contaminated? I did not do DNase digestion when I isolated my RNA samples. I did not see any gDNA bands but could the abnormal increase be caused by the presence of pseudogenes?

Ivan Delgado
Ivan Delgado's picture
Hi tigerlily,

Hi tigerlily,

I have heard of Pfaffl's method but never used it. While I am sure it works well, as with any calculation if you are not doing it right you may be getting all sorts of wrong answers. 

My recommendation is to try and get both assays to work well instead of trying to get data from questionable assays. I've never used any qPCR assay with an efficiency lower than 99.8% so I cannot even imagine how skewed the results could be with an efficiency as low as 93%. Think about it: with a 93% efficiency every PCR cycle only applifies 93 out of every 100 DNA molecules!

Ultimately I do not think you can say that an assay is working with an efficiency of 93%. Its simply too inefficient. But that is just me.