There is a distinct possibility that you have selected for a suppressor gene mutation (or epimutation, meaning that you might have an epigenetic modification) in your knockout strain (line). This is very commonly observed in cases where the knockout gene mutation is highly deleterious, such that nearly only survivors that you get have secondary mutations (or epigenetic alterations), usually those that suppress the effects of the deleterious mutation or sometimes (though rarely) those that might even enhance the growth over the wild type construct. How did you get the knockout lines? Are these from heterozygous mice which were made homozygous by breeding? If this is how you did it, then I suspect that the knockout mutation might be semi-dominant, such that the heterozygous embryo in the early stage was under selective pressure for the suppressor mutation(s) to be selected. One way to get around this would be to do extensive back crosses of the heterozygous mice, and cross two such backcrossed hets. Expensive and slow, but important.

There is a distinct possibility that you have selected for a suppressor gene mutation (or epimutation, meaning that you might have an epigenetic modification) in your knockout strain (line). This is very commonly observed in cases where the knockout gene mutation is highly deleterious, such that nearly only survivors that you get have secondary mutations (or epigenetic alterations), usually those that suppress the effects of the deleterious mutation or sometimes (though rarely) those that might even enhance the growth over the wild type construct. How did you get the knockout lines? Are these from heterozygous mice which were made homozygous by breeding? If this is how you did it, then I suspect that the knockout mutation might be semi-dominant, such that the heterozygous embryo in the early stage was under selective pressure for the suppressor mutation(s) to be selected. One way to get around this would be to do extensive back crosses of the heterozygous mice, and cross two such backcrossed hets. Expensive and slow, but important.

thanks very much! my KO MEFs were derived from homozygous embryos from heterozygotes mating. in our system, we have not detected any obvious abnormality in hets. your secondary mutation idea is pretty good.

but i don't get the point of mice backross. why this can get round the possible issue? you mean to dilute or erase the secondary mutation?

thanks a lot

Yes, that's right, to cross out the secondary suppressor mutation. This purification by backcrossing/outcrossing should be routinely done for any knockout mutation if it is thought that the mutation should produce deleterious growth effect when homozygous. However, I would also caution you of a second possibility: off target effects of the knockdown siRNAs. How confident are you that the knockdown RNAs are indeed specific to the mRNA that you are targeting? Might you be targeting a second gene, which is a downstream negative regulator of the pathway that is controlled by the real target, thus actually suppressing the knockdown effect?

I assume for RNAi you are using vehicle controls to rule out toxicity from the reagents?

In addition to the suppressor gene mutation, read up on genes that are both tumor suppressors and oncogenes. There are plenty of examples of gene products that have dual roles as activators and suppresors depending on the context. Maybe start here: http://www.ncbi.nlm.nih.gov/pubmed/22510570

The cell lines may be expressing proteins that interact with your gene/protein that are not present in the MEFs. Look for known pathways with your gene and see which ones are active in the various cell lines versus the MEFs. Also, the immortalized cell lines are not normal and are "programmed" for growth. It'll be easier to slow them down than to speed them up.

If it's a transcription factor, it may be a strong activator competing for binding sites and when you knock it down a weaker factor can bind. It may recruit chromatin remodeling proteins like HATs or HDACs and have a more complex genome-wide role which may be hard to pin down.

Good luck!

Another possibility for transcription factors is a different availability of promoters in different cell lines. While the factor is capable of regulating gene a and b in one cell line, it can only regulate gene b in another because of promoter inactivation (methylation) of gene a.

I do not have a straight answer, but the difference could be due to the fact that MEFS are primary cells (ie not immortalized nor transformed) to the contrary to 3T3L1, C2H10T1/2 or C2C12 that are growing in the dish for many years (almost 40 years for C2C12, if I remember correctly...).

My suggestion is to rule out the off-target effect. I would try antisense DNA assuming RNAi and antisense DNA could have different off-target effects. Its less expensive in a way.

I do not know if MEFs are heteroduplex proteins of super families. For example, AP-1 is Fos-Jun, Fos-Fos or Jun-Jun duplex. Knocking one mRNA done may not remove AP-1 at all. In fact compensatory loop on negative feedback may actually elevates alternative AP-1 expression. Can this occur in your system?

to Animesh, i overexpressed in KO gene can rescue the MEF phenotype, does this mean the phenotype is caused by primary gene deletion? or, if any, reversible secondary mutation.

to Jason: i have control vector. control shrna from clonetch or GFP rnai. btw, it's a pretty novel gene, but has no TF domain, i know little about this gene.

to Philip: i also think about off-target effect, but this RNAi sequence is used for long time in my lab, it's useful to knock down the target gene. i will try to synthesize additional oligos to minimize the possibility.

thank you so much, all of you!:)

There is a lot to consider here, but it seems very unlikely that there are secondary mutations or epigenetic modifications since your MEFs are pooled from homozygous embryos. Are these homozygotes embryonic lethal? Also, is this truly an effect on rates of proliferation or are there differences in cell death to account for cell numbers? Do the MEFs keep growing? Normally they will become senescent after 5-6 passages. Are your knockout MEFS immortalized? As is often the case, the most unusual phenomenon may turn out to be the most interesting and informative. Good Luck

the homozygous mutants are dead around birth.

the proliferation is difinitely increased, no doubt about this.

not immortalized, primary culture from pooled embryos. i haven't done any transformation assays. so i am not sure whether this KO MEFs are easyier to be transformed or not, it's could be an interesting question whether this gene is involved in cancer.

thank you