Tyler Brown, are these different samples from differing patients/samples etc? There could be an innate difference between people for the levels of the house keeping genes. Also have you any runs prior to treatment and on the qPCR did you rune duplicates and were the duplicates ct values matching? Are you seeing a dose change with varying levels of hormone treatment effecting the housekeeping genes +/- ct values?

Lastly what are you using the housekeeping genes for? Just to confirm cDNA or are you using them in later calculations?

Stephen Reynolds this is with an in vitro cell culture, which should rule out the innate differences I would think? The Ct values are consistent between duplicates. We are using the housekeeping genes for relative quantification as well as cDNA confirmation. We've had somewhat inconsistent results when treating with prolactin but have seen dose changes in treatment with estradiol.

We double checked our results and made sure our RNA concentrations were consistent and are fairly confident at this point that we are affecting housekeeping genes in our endometrial cells. I think further dose variations could be a good way to confirm what we are suspecting. You asked some very important questions, thank you Stephen Reynolds

Hi Tyler, I would like to add here few remarks or suggestions;

The term housekeeping gene is now obsolete and from old schools only, as there is no such gene which can be ideally named as a HKG. The best alternate is to use the terms reference gene or loading control or endogenous control. Any gene which is more or less showing expression stability in all of your controls and treatment groups samples can be chosen as reference gene candidate.

GAPDH and B-actin are not good for some cancer or immortal cell lines and if you are working on human cell lines so plenty of pseudogenes exist at various loci all across the human genome, which mimic GAPDH and B-actin so you have to be 100% clean from gDNA. Also, GAPDH has more than 8 known splice variants, which all have functional differences as per the cells state.

So for a safer side, just take the products of B-actin and GAPDH from control and treated samples and run on gel to check they have the desired product sizes or not for all samples so to be sure here that nothing else has amplified and this will also let you know any possible genomic DNA contamination in your samples even in trace amounts.

According to your cell lines, you can select some other reference genes through careful literature search like B2M, PUM-1, POLR2A, UBC, Tfr, PPiB, 18S etc.

In many pathologies, the common reference genes basically are acting as regulatory genes so overall not pretty stable in their expressions as per our expectations.

I would also strongly recommend to transcribe RNA input at least 1 ug or above for cDNA synthesis, as the quality of cDNA equally matters side by side its quantity and poor quality cDNA gives variations even for the reference genes. And, as a rule of thumb, if our reference gene(s) ct values are at or above 25 so our cDNA has compromised quality without any doubt. Check your cDNA samples on your selected reference gene(s) as undiluted and with a series of dilutions like 1:5, 1:10, 1:25, 1:50, 1:75 etc and see the ct values in all the dilutions and replicates to decide finally that where you see the most stable expression.

Relative quantification (RQ) is less reliable if you find a ct difference above 10 between your reference gene and GOIs, so you need to select any other reference gene in such case.

If your reference gene(s) are stable in all the technical replicates of your control and treated samples so you are at least sure that no technical inaccuracies exist here from your side. Also, if there are any expression differences among biological replicates of any certain group in your experiment (like all negative controls or positive controls) so just omit that group from the study and alternatively you can increase the number of technical replicates from the other biological replicates within the same group (treated or control) for good data generation to analyse later.

Regards....

A little too late here but your observations were right. Hormone treatment with estradiol can indeed affect GAPDH mRNA levels in endometrial cells.