I would very much like to start using transgenic mice that posses promoters that drive the expression of flurophores for in-vivo imaging. What I didn't fully appreciate at the time that I ordered the first set of animals from Jackson was the difference between a targeted mutation and a transgene. The animals that we ordered are heterozygotes carrying a transgene. When I asked our transgenic animal resource manager to initiate the breeding 16 animals she told me this was going to be expensive and take quite a while. Her logic, which sounds totally right to me, was that creating and using homozygous animals was a bad idea because the random relocation of transgenes could have an unpredictable effect on the phenotype of the animal. Her suggestion was to use the offspring of a HET x WT cross. The problem with this, however, is that there is only 1/4 chance of producing a HET animal (best case). Further, as I only want males I was likely only to get 1 animal per litter. Therefore, I would need to breed at least 16 litters if I wanted sixteen animals (i.e. I need to generate 128 animals for a yield of 16).
I was wondering whether this was a common issue - and whether there are any alternative solutions?
Rohan,
The transgene does not move randomly in your animals, it remains in the original insertion point. That insertion point is random, and determined at the time of insertion in the zygote that was injected. It will be different for every line made with a given transgene, but for one line it will not change. Therefore, if the gene inserted near the centromere of chromosome 5, for example, all animals descended from that origian lanimal will have it near the centromere of shromosome 5. It will be inherited loike any o=ther gene, as it is just part of a chromosome now. So, half the offspring will be transgenic, breed transgenic, and you can mate them every few days to another female. For 16 animals, mating one fertile, healthy male to 5 females would give you that with two weeks of mating and three weeks gestation. Then you need them to get to whatever age you need for your study. Bottom line, it is nowhere as expensive or difficult as was originally suggested, and you do need to be clear with the colony manager that it is a heterozygous transgene you are woring with, not a knock-out that needs to be made homozygous.
This is more or less normal with transgenics. One point in the calculation is not right. The het will transmit either the wild-type or the transgene allele. Not both can come from the wt. If you obtained the mice from JAX, they most likely provide a genotyping protocol to distinguish between WT, het, and +/+. If this is the case you could cross the het and remove afterwards all wt and transgene +/+. For sure I never use homozygous as you would create kind of knockout in the transgene insertion site (which mostly I don't characterise exactly.)
Ed is right: A HET x WT breeding will give you 50% HET offspring since either the TG or the wildtype allele from the HET founder animal will be passed on to the offspring. In general, breeding TGs homozygously is not a good idea since this might lead to unwanted phenotypes. Anyhow, I would highly recommend phenotyping your HET reporter line for any transgene-related abnormalities in your experimental setup!
Apologies. Yes I totally agree that at HET x WT will give me 50% HET offspring. Thanks for the pickup!
Based on an average litter size of 8, than I should get 4 HETs 2M/2F. Therefore, I would need 8 Litters to generate 16 animals! It makes sense. It was just a bit of shock, especially when I saw the costings! Is it true that a targeted conditional mutation is better than a transgene approach because the design prevents the gene from randomly moving around the genome?
Again more or less right for the calculations. Litter size of 8 would be correct for C57/Bl6. For instance for NMRI backcrosses I get much bigger litters. Targeted insertion has for sure the benefit that you know where it is. We have a couple of things inserted in the Rosa26 locus. Has the big disadvantage that the expression is weak.
Your animal resource manager is correct. If the transgenic mouse from JAX is a random CRE line (hemizygous) then you will need additional breeding to have targeted mice with your gene deleted and cross these to have homozygous mice for the targeted gene, although you should keep a line of your mice as heterozygotes by crossing to wt.
Rohan,
The transgene does not move randomly in your animals, it remains in the original insertion point. That insertion point is random, and determined at the time of insertion in the zygote that was injected. It will be different for every line made with a given transgene, but for one line it will not change. Therefore, if the gene inserted near the centromere of chromosome 5, for example, all animals descended from that origian lanimal will have it near the centromere of shromosome 5. It will be inherited loike any o=ther gene, as it is just part of a chromosome now. So, half the offspring will be transgenic, breed transgenic, and you can mate them every few days to another female. For 16 animals, mating one fertile, healthy male to 5 females would give you that with two weeks of mating and three weeks gestation. Then you need them to get to whatever age you need for your study. Bottom line, it is nowhere as expensive or difficult as was originally suggested, and you do need to be clear with the colony manager that it is a heterozygous transgene you are woring with, not a knock-out that needs to be made homozygous.
It is indeed difficult to test by PCR if the transgene is in heterozygot or homozygot state. Since you usually don't know the exact integration site (by the flanking genome sequence), you also wont find a PCR based assay that tells you the presence of the "wt allele". It is easy to test for the presence of the transgene, of course, but to distinguish between homo- and heterozygote you have to find a way to tell if only one parental chromosome or both carry the transgene. Using an assay for DNA copy numbers is hampered by the usual multiplicity of tg copies at the integration site. You could do a FISH hybridisation, to see if you got a single signal in each cell or two (i.e. homozygosity).
Much easier, however, is the so called Testcrossing approach: You simply apply the statistical laws of Mendelian segregation. For this, you mate the tg animal (unknown het or hom tg) with a wt animal. The offspring of this outbred you than test for the presence/absence of the tg by PCR. If you find a family of which 100% (say 10 out of 10 at least) of the offspring are tg+ , you know almost for sure that the parent was homozygote for the transgene. In contrast, if the offspring are half/half for the tg, the parent was heterozygote.
Once you identified one family with a homozygote tg mother and another one with a homozygote tg father, you use those two and can breed a stable homozygote line (without the need of further genotyping). It might be that in order to minimize the risk of genetic drift or inbreeding suppression, you want to do this in three homozygout families in parallele, and establish a cousine x cousine mating scheme.
Michael
JAX will tell you if the animals you purchased are hets or not and provide a PCR genotyping protocol. You can keep your line heterozygous by mating hets to WT and genotype the offsprings. Half should be het and half should be WT. If all the progeny of the cross between the founder you purchased and the WT is a het, you got a homozygous animal. If you need only 16 het males, you can do harem breeding by mating a het male with several WT females at a time. You may want to plug the females and remove the pregnant animals to make sure you do not have too many pups in one cage. One male can serve at least 4 females at a time, you need maybe 8 litters to make sure you have enough het males.
Agnes
Dear Ed, Kay-Dietrich, Zissis, Jan, Michael, Michael, and Agnes,
thank you so much for your responses. This has given me an enormous amount of clarity with respect to the optimal breeding approaches for these animals. I'm sure that others in my position will also find this thread enormously useful!!
Kind regards,
Rohan.
Hi, I just read quickly the previous ones, so sorry if I'm repeating what somebody already said.
I hit approximately the same problem with a cre-lox line, so here are my advices:
-if someone knows how to do it, you could try to locate your transgene and clarify things up (droso people do this all the time, might be easy if you know some of them). But it could take time if you're on your own.
-given the time and the cost, for my part I chose to generate homozygote animals (you need at least males, and ideally establish a homozygote line), but use them only for reproduction: for ex males homo X female WT. This way, you need only a few fertile homo males that you shift from cage to cage with 2-3 females in each. 100% of the offspring will be het: much better.
-it seems to me that using homozygous animals for reproduction is not as "dangerous" (i.e. bias-generating) as if you used them for your study, especially if it's a fluorescent line used for imaging.
-BUT ideally, to avoid any doubt and criticism, I would advise you to generate a few animals just to control: in your case you could use homo X WT breeding but verify the most striking phenotypes on animals coming from het X WT breeding, for example.
-lastly, to generate your homo line you'll need to discriminate between het and homo animals. If you have no info on the trangene insertion, you can still use qRT-PCR to see if you have 1x or 2x copies of the transgene in your mice. And the perfect control of course is that you should obtain 100% trangenic offspring.
-remark: that's why people prefer to use knock-in mice nowadays: this way you know exactly were your trangene is inserted!
good luck, and last piece of advice: careful to balance speed vs ethics (if you're in a hurry, you'd better start a lot of breeding cages to avoid wasting time; but it will likely mean a lot of animals to euthanasize...)
I would just emphasize that in my work I routinely maintain homozygous breeders for a transgenic line when possible for the simplicity (no need to genotype since 100% of offspring will also be homozygous). When producing mice for experiments we simply breed the homozygous mouse with a WT mouse, and again no need to genotype offspring assuming the genotypes are previously confirmed--all of the pups will be useful hemizygous animals.
We initially derive the homozygous mice by test crossing as outlined by Michael, although qPCR works too and may be faster. It is a pain and takes time to establish the homozygous breeders, but if you plan to use a lot of mice of that line over a long time this is the way to go. The only caveat is that it is possible that the homozygous mice are not viable or are somehow less viable if the random insertion site of the transgene disrupted an important gene by bad luck. It is estimated that about 1 in 10 transgenic lines derived from pronuclear injection and random integration results in the disruption of some other gene. This is a good thing to keep in mind when working with transgenic mice since few people (myself included) ever track down the insertion site.
- Badges
- Science topic
- Similar topics
- Medicine
- Disease
- Disease Models
- Animal Models