A resource of targeted mutant mouse lines for 5,061 genes

Summary

The International Mouse Phenotyping Consortium reports the generation of new mouse mutant strains for over 5,000 genes, including 2,850 novel null, 2,987 novel conditional- ready, and 4,433 novel reporter alleles.

Despite thirty years of mouse targeted mutagenesis, in vivo function of the majority of genes in the mouse genome are still unknown. This reflects the observation that a small number of genes have been the object of intensive study including the development of multiple mouse models, while a significant proportion of the coding genome remains entirely unexplored ¹ The completion of the sequencing of the mouse genome, coupled with the use of mouse embryonic stem (ES) cells for gene targeting to create complex mutant alleles, presented an opportunity to functionally analyze all the protein coding genes of a mammalian species ^2,3 Taking advantage of comprehensive manual annotation of the genome ⁴, the International Knockout Mouse Consortium (IKMC) systematically generated single-gene, reporter-tagged null alleles for protein-coding genes by homologous recombination in mouse ES cells ⁵,⁶ Subsequently, large-scale mouse production and phenotyping programs deployed these unique resources, establishing the feasibility of genome-scale mouse production and phenotyping ^7–9 Building upon these successes, the International Mouse Phenotyping Consortium (IMPC) was established to coordinate a network of programs around the globe, assuring uniformity and reproducibility of these efforts, including standardization of phenotyping protocols and the use of a single inbred mouse strain background, C57BL/6N, with the ultimate goal of generating and phenotyping a single-gene knockout (KO) mouse line for every protein-coding gene in the genome.

Production of KO mice began in concert with the expansion of the ES cell library, but rapidly accelerated after 2011 with the funding of multiple IMPC programs. To date, more than 17,500 individual production attempts (microinjection or aggregation) have resulted in the germline transmission of KO alleles for 5,061 unique genes (Figure 1a; Supplementary Table 1). These lines have been expanded for phenotyping, providing key insights into mammalian biology and disease ^10–15; www.mousephenotype.org). To date, phenotype data for these lines shows that overall 72% of lines display at least one phenotype, revealing extensive pleiotropy (Supplementary Figure 1). This includes the 35.8% of lines that show partial or complete lethality, consistent with our earlier finding ¹¹. The IMPC contribution extends the total number of genes with targeted KO alleles produced by the scientific community from the 8,391 reported and curated by Mouse Genome Informatics (MGI; ¹⁶)(Figure 1b; Supplementary Table 1), to 11,241, or more than half of the genome. Much of the overlap (2,211 genes) reflects specific community requests for the production of novel complex alleles (see below), targeting on an inbred C57BL/6N background, or for mutant mouse lines unavailable through public repositories. The growing use of CRISPR/Cas9 editing to produce null alleles for the IMPC led to the decrease in ES cell-based production beginning in 2015.

Figure 1:

Generation and impact of targeted alleles for 5,061 unique mouse genes. (a) Cumulative production progress, including all attempts (microinjection or aggregation (black), unique ES cell clones injected (red), unique genes attempted (yellow), and unique genes that achieved germline transmission (GLT; blue). For GLT, the date reflects the date of microinjection, and only reports the first instance of transmission for the small number of duplicate mutations produced. (b) Venn representation of unique gene null alleles produced by the IMPC (orange) and by the rest of the scientific community as reported in MGI (“Non- IMPC”; blue). (c) Unique gene conditional-ready alleles produced by the IMPC (orange) and by the rest of the scientific community (blue). (d) Unique gene reporter alleles produced by the IMPC (orange) and by the rest of the scientific community (blue). (e) Cumulative mouse orders of IMPC lines processed by production centres and mouse model Repositories from 2012–2018 (blue line). The cumulative number of ordering investigators, unique alleles ordered, and unique genes ordered are shown in yellow, grey, and orange, respectively.

While the primary goal of the IKMC and IMPC was to generate and phenotype a null allele for every protein-coding gene, the mutant alleles included additional functional features. All alleles included a lacZ reporter cassette to facilitate analysis of gene transcription in situ (Supplementary Figure 2; ^5,6). A large proportion of the alleles have conditional potential, providing future users with a useful tool for detailed, mechanistic analyses (Supplementary Figure 2a). The multifunctional utility of the alleles produced by the IMPC has greatly expanded the repertoire of genetic resources available to the scientific community. Of the 3,674 unique gene, conditional-ready mouse models generated and validated, 2,987 were novel alleles for genes without an existing conditional allele (81.3%). These nearly double the total number of genes with conditional KO alleles produced by the scientific community as a whole (2,987 IMPC conditional alleles added to the 3,295 conditional alleles reported in MGI as mouse lines; Figure 1c). The impact is even more significant for reporter alleles. The IMPC has produced reporter alleles for 5,059 unique genes, of which 4,433 are novel (87.6%), complementing the 2,733 produced by the scientific community (Figure 1d). This has nearly tripled the total number of genes with reporter alleles available to the community as mouse lines.

The generation of mouse lines was underpinned by comprehensive quality control strategies for both ES cell karyotype and targeted allele, which ensured efficient production and integrity of the targeting event (Supplementary Table 1; Supplementary Figures 3, 4). Further quality control (QC) analysis also showed that part of the ES cell collection contained an additional insertion of a wild-type nonagouti (A) gene on chromosome 8, likely introduced with the targeted reversion event in these cell lines (subclone JM8A3¹⁷). However, as the insertion of the wild-type nonagouti gene results in an agouti coat color, this allele can be easily segregated from the mutant allele in most cases (Supplementary Figure 5). High- throughput allele validation of ES cells was performed using either a suite of quantitative and endpoint PCR-based tests or a combination of Southern blot¹⁸ and PCR-based analysis¹⁹, depending on production center (Supplementary Figures 3, 4 and 6; Supplementary Tables 2, 3; Supplementary Note). Despite these efforts, we found that additional quality control (QC) on the mouse lines themselves was required to ensure all IMPC lines resulted from the transmission of the correctly targeted allele (Supplementary Figure 4; ¹⁹). This additional QC at the mouse level identified a small but significant proportion of incorrect alleles that transmitted through the germline of chimera mice derived from clones that had passed initial and secondary validation QC testing in the ES cell. Our experience highlights the importance of careful allele validation before and after mouse production.

As a result of this effort, mouse lines with targeted alleles for more than 5,000 genes on a C57BL/6N genetic background with extensive and documented genetic validation of the targeted locus are now available to the biomedical research community, supporting high standards of reproducibility for future investigations. This resource nearly triples the number of genes with reporter alleles and almost doubles the number of conditional alleles available to the scientific community. When combined with more than 30 years of community effort, the total mutant allele mouse resource covers more than half of the genome. The IMPC resource has shown its usefulness through the continued and robust uptake of mutant mouse lines by investigators around the world. This includes both KO and conditional alleles with mouse lines distributed as live mice and cryopreserved stocks. To date, over 5,000 orders for mutant mice for 3,301 unique genes have been processed and shipped to more than 4,000 investigators around the world (Figure 1e). To date, more than 1,900 publications acknowledge the use of EUCOMM/KOMP alleles (for example ^20–30). This demonstrates the utility of these resources, the cumulative use of which continues to grow over time, and complements the systematic phenotyping efforts of IMPC centers. In the new era of genome editing, this ES cell-derived collection remains of unique value as it offers particularly sophisticated and quality-controlled alleles representing a cornerstone of the collective development of a null allele resource for the complete mammalian genome ².

All data are freely available from the IMPC database hosted at EMBL-EBI via a web portal (mousephenotype.org), ftp (ftp://ftp.ebi.ac.uk/pub/databases/impc) and automatic programmatic interfaces. An archived version of the database will be maintained after cessation of funding (exp. 2021) for an additional 5 years. Information on alleles, together with phenotype summaries, are additionally archived with Mouse Genome Informatics at the Jackson Laboratory via direct data submissions (J: 136110, J:148605, J:157064, J:157065, J:188991, J:211773).