Hi,

plainly said: doing what you want to achieve is extraordinarily (aka impossible) difficult using current technologies.

Knock-out mice are currently the best tool to study gene function in mammals. Since there are hypomorphic mice (lower expression levels) available for many genes of interest that might be worth looking into as well. Those of course can theoretically be interbred to generate several hypomorphic or knock-out alleles in one strain. But hypomorphic allels are usually not inducible, so you would probably need inducible gene expression systems in the mouse for multiple genes.

To my knowledge there are basically two well established systems for modulating gene expression, Tet-inducible and estrogen (or Tamoxifen)-inducible. If you want to look at actual expression as a readout you will have to stick with the Tet-system as the ER-system is based on fusion proteins and will modulate protein function rather than expression. You can use both of these systems as transgenes or knock-ins or even for knock-outs (with eg Cre-ER or Tet-Cre mice). In theory expression (or activitiy in case of the ER system) levels can be fine tuned by titering Tet and Cre administration to whatever you want. So in theory you may have 2 genes to play with, but practically I am not aware of anyone ever having published more than 1 manipulated gene in a live animal. Also, depending on the target construct titering expression levels exactly can be next to impossible because there's interindividual variation in the mouse, the systems can be leaky, and plainly some genes are not inducible after putting them in a mouse as a transgene. So you are probably not going to be able to achieve what you want to do in the near future.

But there is hope: Further development, eg of CRISPR/Cas based systems may allow at some point allow us to generate mouse strains combining multiple hypomorphic alleles of multiple genes in the same mouse, potentially driven by both tissue- and differentiation-stage specific promoters. This may come as close to what you're aiming at, but unfortunately this is (to my knowledge) not yet possible.

Hi David, I agree with Christian, you are for the present a bit too ambitious. For deleting a gene in a tissue-specific and time-controlled way, the best system is indeed producing a flox allele (the so called conditional allele) in one mouse and cross it with a second mouse line containing a CreERT recombinase under the control of a tissue-specific promoter: the estrogen-binding domain on the Cre will allow the temporal control, since the Cre will enter the nucleus and recombine the LoxP sites only in the presence of Tamoxifen, which you can provide to the mouse at the time point you need. The Tet-On system is more complex, since it has three components to be both tissue- and time-specific. In fact you will need to express the tTA repressor under a tissue-specific promoter and a Cre under a TRE-regulated promoter as a transgenic insertion. The system is quite leaky and it is believed the Tet-Off system is more efficient. The Tamoxifen system is becoming more and more used, but it is also not completely immune from efficiency problems.

In general the problem of combining many floxed alleles and different Cres is the breeding scheme. You might be lucky and manage to select lines which are homozygous floxed on two or three genes, but when you start crossing with the Cre lines, they'll go in heterozygosity and you have to start again to homozygose them. Furthermore, Cre lines as knock-in to have the best tissue- and time-specificity of expression often cannot be kept in homozygosity, since they knock-out genes and are often lethal. But transgenic Cre lines are difficult to genotype, since the insertion locus is very very seldom known, so to know if you have one or two copies you have to genotype by qPCR. Still you can try to combine tissue- and developmentally-regulated Cres with CreERT and possibly activate 2 or 3 Cres in your mouse.

In conclusion, I don't want to say that certain projects are impossible, but definitely very ambitious, and will have to face high costs for the breeding scheme and require a lot of mouse space, which is in contradiction with the "reduction, refinement, replacement" strategy in today's animal research, in the Western world at least. This is probably why there is not much work combining multiple tissue- and time-specific knockouts, as Christian correctly states.

The Crisp/Caspase system is indeed a great improvement in recombination strategies, but to make it tissue specific will still require some work, since the plasmids delivery will have to occur in a tissue-specific way, and this is still something we are fighting for gene therapy...

I would try the Crispr/Cas system. 

I agree with Matthew's suggestion. Here's a great review in Nature Biotechnology on the Crispr/Cas system. http://www.nature.com/nbt/journal/v32/n4/full/nbt.2842.html

Go CRISPR! 

Hello, most of the answers did not answer your core question: " the question is how many genes are possible to knockdown and not delete them completely but manipulate expression e.g. I want that in mouse at P12 days increased A gene in brain or especially in Hippocampus at 30% than expression of another gene B parallel to A would be increased at 60% and in adulthood would be decreased gene C so one (how many genes are possible to change maximal amount?)."

For conditional (not complete) knockdowns or overexpression the CRISPR/CAs System is not useful. Here conventional KO can be generated (more that one at the same time), or for conditional (here Cre mediated knockout) alleles you would have to insert loxP sites flanking your genes, which is laborious even for one gene. I would go for RNAi (miRNAs), expressed by tissue specific promoters (Pol II), which could theoretically be used at the same time for the knockdown of different genes in different organs. They could also be expressed inducibly/reversibly using the Tet system. However, what you plan, different gene in different organs at different time points that practically impossible, maybe you could combine trangenic expression with use viral injections and this would allow you to study 2-3 genes. But still very difficult.

kai

kai

I agree with Kai. First thing I thought about when I read you want to knockdown genes was miRNA under tissue specific promoters. Combining this with viral injection at specific timepoints could theoretically work, but the amount of controls for this system would have to be huge (tissue specificity, time of miRNA expression, efficiency of knockdown, miRNA stability and so on for every gene). I can't answer how many genes could be studied this way simply because I haven't heard about anyone trying it on such a scale.  

Also, one needs to keep in mind, that 30% knock down with Cre is actually 100% KO in 30% of cells or 50% KO in 60% of cells, most probably, something in between, but that is different from 30% KD in each cell. If that is what you really want, you have to use miRNA.

@ Jana,

Is that true? I always think that 30% knock down means that the gene can only now expresses 70% capacity in every cells.