2, Immunotherapy Center, Cancer Center and Department of Medicine, Medical College of Georgia, Augusta, GA, USA
The International Mouse Knockout Consortium aims to generate a knockout mouse for every single gene on a C57BL/6 background. Our ability to generate such mice is hampered by the poor economics of producing blastocysts to achieve germline transmission of C57BL/6 embryonic stem (ES) cells. We demonstrate superior utility of (C3H × BALB/c)F1 blastocysts compared with BALB/c blastocysts, with blastocyst numbers and germline transmission from subsequent chimeras at a rate 2- to 3-fold higher than that produced with BALB/c blastocysts.
The completion of the C57BL/6 mouse genome and the advent of the U.S. National Institutes of Health (NIH) Knockout Mouse Project (www.komp.org) and the International Mouse Knockout Consortium (1) will greatly accelerate efforts to generate a knockout for every mouse gene, and will represent an enormous resource for the research community to study and discover cures for human disease (2). Substantial effort will undoubtedly be devoted to the important task of improving the efficiency of generating knockout mice on a pure C57BL/6 background. Although we are committed to generating knockout mice on a pure C57BL/6 background and have generated our own C57BL/6 embryonic stem (ES) cells (3), an inherent difficulty perceived is the relatively low efficiency of production of appropriate blastocysts to generate chimeric mice and achieve subsequent germline transmission. For example, due to the limited housing capacity in our facility, it is not possible for us to breed sufficient numbers of C57BL/6-albino mice as blastocyst donors, and the relatively high cost of purchasing such mice means that regular importation is not economically viable. Also, the rate of blastocyst production from C57BL/6-albino mice (data not shown) was consistent with that reported by others (4) and was not significantly higher than that achieved with BALB/c mice. That is, we have routinely purchased BALB/cAnNCr mice from the U.S. NCI-Frederick Animal Production Program (web.ncifcrf.gov/research/animal_production_program) but our ability to generate blastocysts did not meet our needs. While superovulated BALB/cAnNCr females (see Supplementary Material available online at www.BioTechniques.com) were plugged at a rate of 69% ± 15% standard deviation (SD; n = 38 experiments with at least 20 mice per experiment), only 20% ± 11% of those had actually ovulated, and yielded only 0.51 ± 0.31 useful blastocysts per donor (i.e., total number of females superovulated). The poor super-ovulation of BALB/cAnNCr mice was consistent with expectations of some but not all BALB/c strains (5).
In an attempt to improve efficiency, we tested other potential sources for blastocyst production and their capacity to generate chimeras with our recently described C57BL/6 ES cell line, LK1 (3). This was done at a time that overlapped with the above experiments using BALB/cAnNCr mice, but only one strain of mouse was used per daily session. We began with C3H/HeNCr (MTV-) females as a source of blastocysts because we routinely use such females in combination with C57BL/6 males as a source of oocytes for generating transgenic mice. We used the same BALB/cAnNCr males as studs as above in 20 daily experiments. C3H/HeNCr (MTV-) females were plugged at a rate of 70% ± 14% SD (with at least 20 mice per experiment) but also 42% ± 15% of these ovulated and yielded 1.38 ± 0.48 useful blastocysts per donor (i.e., total number of females). When these data were collated into a larger data set of 70 experiments (with about 40 females per experiment in the additional 50 experiments), C3H/HeNCr females were plugged at a rate of 68% ± 11% and yielded 1.21 ± 0.55 useful blastocysts per female (see also Supplementary Figure S1). Thus, the rate of C3H × BALB/c blastocyst production was 2- to 3-fold higher than that of BALB/c blastocysts.
We also collected data from 23 experiments using C3H/HeNCr rather than BALB/cAnNCr stud males. This approach appeared to be slightly less consistent, with females plugging at a rate of 58% ± 16% and yielding 0.98 ± 0.71 blastocysts per female.
A higher rate of blastocyst production, of course, does not necessarily result in increased ES cell germline transmission productivity. In order to test the latter, C3H × BALB/c blastocysts produced above were used to generate chimeras with targeted homologous recombinant ES cell clones produced in our facility (see Supplementary Material). That is, a number of different targeted ES cell clones were derived using our LK1 ES cells in service to our institutional investigators. The data discussed below were collated from 38 different ES cell clones derived from a total of 17 different targeting events. While nonchimeric C3H × BALB/c mice have agouti coat color, chimeras from LK1 homologous recombinant clones appeared to have a discrete patch of black coat, in particular on or near the head ((Figure 1)A). Based on this criterion, 155 male and 33 female chimeras were produced, representing a chimera rate among the total progeny of 30% ± 14% SD. As a whole, this represents a male chimera rate of 82% of all chimeras. In terms of the individual ES cell clones (i.e., the percentage of male chimeras among all chimeras for each individual ES cell clone, and these numbers then used to calculate average ± SD among all ES cell clones), however, it was 87% ± 18% SD (n = 38 different ES cell clones). This data is consistent with if not even more impressive than that generated by others with other ES cells, as well as that recently published of LK1 ES cells (3), using other blastocysts as discussed elsewhere (3). Furthermore, these male chimeras represented 7.8% ± 4.7% (n = 38 different ES cell clones) of the total blastocysts transferred to pseudopregnant foster mothers that successfully produced litters.