Hello Jeya,

did you removed the the whole gene or did you added fragment inside the gene of interest and inactivated it? If it is the second one the probable reason must be that your primer for qPCR should be in the region before the fragment which is inserted, so the partial mRNA will be picked up by the PCR .(probable answer not sure)

1. did your qpcr target all isoforms (if there is any that is more abundant) or your antibodies target a specific isoforms

2. is the knockdown a hairpin method?

3. change a control scrambled exons and see it makes any difference

it might be that the qpcr detect something else. try design more primers on 3' utr or span exons and check the pcr products whether there is single amplicon or non-specific bands exist.  

I agree with the above concern about the primers hitting a different target. I'd run a quick QC/QA on the primers (serial dilution for linearity/efficiency, no RT controls to show RNA specificity, a no-template control to ensure there's not template contamination somewhere, etc). Primers are quick, cheap and easy to redesign and buy if it does seem like something is wrong with your current set.

I assume you are using shRNA for knockdown. Are you using a polymerase II promoter to drive production of the shRNA? Depending on the degree to which your shRNA is being produced, you may find that you still detect the mRNA in your knockdown cells, but it should be reduced. Seems that it is working considering that you are getting knockdown of protein, but perhaps it can work better.

When you perform the qPCR, if there is no difference in the Ct (or Cq) values between the two samples (Control and KD), then you may consider a redesign of your primers, as others have suggested.

A few considerations for your qPCR:

The qPCR results may be due to the formation of primer dimers, to help avoid this  use a primer concentration at no greater than 125nM. Primer dimer formation will be more evident in the no template control, if you are not getting primer dimers appearing in this control, then you probably don't have an issue with primer dimer formation. If you are, you can consider spiking the negative control with a non-specific template to match the amount of cDNA you are using in your experimental samples. 

We had an issue in our lab, where we were getting unexpected qPCR results. We found that one or some of our reagents were contaminated with DNA (probably plasmids) containing our gene of interest. We used everything fresh and the problem went away. This is of particular concern if your laboratory regularly manipulates DNA (e.g., plasmid) containing your gene of interest. I would designate a set of reagents for only qPCR and perform the qPCR in an enclosed environment.