Breeding mice to create new models - How do I get there from here

With the incredible variety of available mutant, knock-out, knock-in, transgenic, Cre-lox, Tet-inducible system,  and other mouse strains, researchers are able to combine multiple mutations and transgenes to generate new mouse models.  As you can imagine, if you are breeding multiple mouse strains together to generate double, triple, or even quadruple and higher multiple mutant/transgenic mice, things can get complicated in a hurry.  Therefore, you need to be able to map out the most efficient breeding scheme(s) to produce the mice you require for your experiments as quickly and in as high a frequency as possible, so that you can minimize the amount of resources (space and costs) you use to produce those mice.

Mendelian genetics and Punnett squares

Although you don’t need a Ph.D. in Genetics to design optimal breeding schemes, it is essential to understand basic Mendelian genetics.  Punnett squares are very useful tools for calculating expected Mendelian ratios from any given mouse cross.  If the thought of drawing Punnett Squares fills you with dread, don’t sweat it; there are online Punnett square calculators that are easy to use.

Determine genotypes

The first and most critically important step in designing ANY breeding scheme is to determine what genotype(s) of mice you need for both your experimental AND your control mice.  Whether you are dealing with a single mutation or transgene, or combining multiple mutations/transgenes, you need to know your end goals prior to starting breeding.  It is like writing a murder mystery; you cannot write the story if you do not know the ending.

Failure to properly determine the genotypes of mice needed is the most common mistake researchers make when planning their breeding schemes.  This leads to a whole host of downstream problems, including using less than optimal schemes that produce a lower percentage of needed mice or not employing crosses that produce required and appropriate controls.

Let’s use the following commonly used Cre-lox scheme as an example:

• Experimental mice:  Homozygous for loxP / Hemizygous for cre transgene
• Control mice:  Homozygous for loxP / Noncarriers (Wild Type) for cre transgene

Working backwards

Once you know the genotypes of the all mice you need for your experiments, the next step is to determine the most efficient end cross that will produce your cohorts of experimental and control mice.  To do this:

1) Look for alleles that are common between the experimental and control mice.  You want to try to fix the common alleles for homozygosity, if possible.

2) Work the cross for each allele individually, then combine.

In the example above, both the experimental and control mice are homozygous for loxP, so loxP is the common allele.  Because we need homozygous loxP for both, we will aim to breed homozygous loxP by homozygous loxP.  For the cre transgene, we need hemizygotes and non-carriers.  If we bred hemizygote x noncarrier, then we expect to produce 50% hemizygote and 50% noncarrier pups. (This is why knowing how to use Punnett squares is helpful!).   Our individual crosses are:

Homozygous loxP   X   Homozygous loxP

Hemizygous Cre      X   Noncarrier

So now, we just combine the crosses for the individual alleles by combining the right side together and the left side together to get:

Homozygous loxP , Hemizygous Cre   X   Homozygous loxP , noncarrier

This cross will then generate both experimental and control mice, in a 1:1 ratio (draw your Punnett square if you doubt it!).  That’s not so hard is it?

Start with parental strains

Once you have the end cross determined, you next have to figure out how to get there from your parental strains.  Of course, you have to start with crossing the parental strains together, but what’s next?  In most cases, you next will want to fix the common alleles for homozygosity, because when you can use homozygotes for breeding, you increase the percentage of useable mice and decrease the overall numbers of mice generated.

Again using the same Cre-lox example: 

1) Breed your homozygous loxP mice (most will be homs) to your cre transgenic strain, which will generate mice that are heterozygous loxP , hemizygous cre.   

2) Cross the heterozygous loxP, hemizygous cre mice back to the homozygous loxP, which will generate some homozygous loxP, hemizygous cre and some homozygous loxP, noncarrier mice (~25% of each), which are the genotypes needed for the end cross!  See our previous blog post on Cre/lox Breeding for Dummies, which illustrates the same cross.

Other factors to consider

There are some final considerations that you should keep in mind as you are working through your crosses. 

1) When combining multiple mutations and transgenes, are they all on different chromosomes?  If any are on the same chromosome, that may limit your breeding scheme options.

2) If any of your mutations or transgenes affect fertility or cause lethality with a specific genotype or  sex, then you may need to adapt the genotypes of your breeders accordingly to compensate.     

3) When working with Cre-lox systems you should also consider: