I am not the expert for this specific question - although I once had transgenic mice to analyze (with a very restricted promoter, but expressing in the wrong, originally unwanted region, but unfortunately no phenotype, no paper). I do not see anything obvious being wrong in your approach. Here are just my brainstorming ideas: Did you sequence your final plasmid with the transgene to check for any point mutations in the promoter? Is the orientation of the transgene in your original plasmid known to be relevant?

Hi Robert, thank you for your suggestions.

We have used the transgene construct in stable cell lines and know that when rtTA is expressed, we get induction of the transgene + GFP. We then cut out the cDNA and this was purified and used for embryo injections. Therefore I am pretty certain that the construct is ok (for we also see in the tissues that do express that the transgene levels are up).

The promoter is on a separate mouse line - the JAX ROSA26-rtTA mouse which we have used when crossed with a TRE-H2B-GFP reporter line and we see expression of H2B-GFP in all the expected tissues so there is no issue with the promoter mouse per se. One difference is that the H2B-GFP is a fusion protein (and H2B is v.v. stable) whereas our Tg contains and IRES and separate GFP protein and we have no idea of the stability of either protein. However the mice are on a constant, high dose of dox so they should be continually making new proteins.

Again, the orientation of the Tg should not matter as it is downstream of TRE (Tet responsive element) and therefore should be expressed when rtTA + dox present irrespective of the orientation. That is if I am getting you question right?

Hi Chris

I don't have much specific knowledge in this area, however it seems clear from what you detail that the issue is related to your injected DNA and not the ROSA26-rtTA. I have three thoughts:

- I notice all the tissues you see expression in are highly proliferative. Perhaps high levels of global transcription are important for your transgene - ie perhaps it's under a weak promoter?

- IRES: There may well be a tissue-specificity with the ribosomal-IRES interaction. I'd do as you say and test your transgene by qPCR. Certainly in my systems I use a viral 2A system as this gives a stoichiometric relationship between molecules of Tg and reporter produced, unlike the IRES which in certain contexts can give Tg-only expression or reporter-only expression.

- dox penetration/action in tissues: perhaps this might be a distant possibility (especially if your Tg is expressed at low levels/off a weak promoter) however your H2B-GFP experiment points against this (though this is a very highly expressed model and may not be directly comparable). I mention this as other induction agents (eg Tamoxifen) can get mopped up by hormone responsive tissues (obviously) making Tg induction in those tissues very difficult. I know Dox doesn't have this characteristic, but thought it was worth a mention.

I'll be interested to see what others have to say and will watch this thread with interest.

Best of luck with your experiments Chris

Al

Hi,

Random integration into the DNA is kind of uncertain (I did Recombineering instead). Along with mRNA detection, I will suggest trying to check the DNA methylation status of the inserted transgene in different tissues, due to that can regulate transcription. Methylation status changes at different tissues and it may help you to try to understand what is going on.

Good luck,

Manuel

Hi Chris, Thanks for further clarifying your approach; so my idea of any tissue related regulatory sequences on the transgene containing construct is less likely; I wondered, if there could be a point mutation somewhre, since you asked a few months ago about the possible importance for not using ethidium bromide when cutting out DNA from agarose.

Thanks for the ideas.

The transgene is under the control of the TRE promoter (Tet Responsive element), therefore expression is driven in the presence of dox + rtTA (which should be in all tissues).

With regards to the IRES, we are looking at GFP as a marker of overexpression by imaging the whole mouse/organs for GFP fluorescence as a first step, but we are also performing Western blots on the tissues for our transgene, therefore we know the relative levels of the protein in all tissues in induced and uninduced mice - regardless of whether the GFP is expressed downstream of the IRES.

I have considered the fact that the tissues that are expressing are the ones most in contact with the dox-containing water. My boss doesn't believe this theory and as you mentioned, the H2B-GFP mouse also goes against it (although also as you mentioned its a much better expressor/more stable and is a specific site-located integration @ the col1a locus).

I should also mention I have performed IHC on the various tissues, and even within the tissues that are inducibly overexpressing the transgene by GFP fluoro + WB, we see very patchy/heterogeneous expression of the transgene/GFP.

Methylation status of the Tg is also a good point, and something we could look into.

Indeed, there may be a point mutation in the cDNA that we injected into the mice, however the DNA was exposed to UV for far less time than I normally do with DNA for cloning for cell culture constructs and I have never had a problem with mutations - although maybe I just got unlucky this time!

mRNA expression profiling seemed easy in theory - but I am having difficulty isolating pure RNA/mRNA from stable cells expressing the same construct!

Just another idea concerning the dose or stability of the transgene inducing Doxycyclin. You do see some effects, and not in all cells of these tissues; could there be a critically low dose or old preparation of your Dox stock solution explain all these effects? One reason could also be that both, Calcium or Magnesium in the water or nutrition could form complexes with Dox, leading to bad absorption and low levels in the tissues. Is your Dox containing water, perhaps, supplemented with in this case contraproductive Vitamin/Calcium/Magnesium molecules? Or is the nutrition high in Calcium/Magnesium?

If you won't find the solution - and you think of repeating similar experiments - I suggest sequencing the construct before incection, to make sure the construct is OK and to exclude any strange mutations which can occur during subcloning (not only due to UV).

A propos, UV may also affect doxycyclin...