Strain Development and Description of FLExDUX4 mice

FLExDUX4 mice (Figure 1) were created at genOway, S.A. in France using a cre-dependent one-way genetic switch (FLEx) system by knockin to the Rosa26 locus. The FLExDUX4 targeting vector was designed with (from 5' to 3') a loxP site, a frt-flanked neomycin resistance gene, a lox511 site, an inverted human double homeobox 4 (DUX4) gene with the homologous polyadenylation sequence, and another loxP site and lox511 site, both in opposite orientation to the 5' sites. This entire construct was inserted between exons 1 and 2 of the Gt(ROSA)26Sor locus via electroporation of C57BL/6-derived embryonic stem (ES) cells. Correctly targeted ES cells were injected into B6(Cg)-Tyrc-2J/J blastocysts, and resulting chimeric males were bred to C57BL/6J females. Offspring were subsequently bred together to create a homozygous colony of FLExDUX4 mice.

Homozygous mice carrying this DUX4 conditional allele are viable and fertile. Two sets of incompatible outward facing recombination sites (loxP and lox511) flank an inverted human double homeobox 4 (DUX4) sequence, including exons 1-3, both introns, 5' and 3' DUX4 UTRs, the DUX4 polyadenylation signal, and the pLam region downstream (see Genbank format .docx sequence file). The PCR genotyping protocol distinguishes homozygotes from heterozygotes.

The endogenous DUX4 gene expresses at least two short non-pathogenic isoforms (DUX4-s) and a longer cytotoxic isoform (DUX4-fl). This strain contains 4 point mutations in the 5' splicing donor sites for the two prominent DUX4-s mRNAs, abolishing expression of the short isoforms and only generating the pathogenic DUX4-fl mRNA isoform. Because this construct was targeted for insertion into the Gt(ROSA)26Sor locus, DUX4-fl expression is driven by the Rosa26 promoter and determined by which tissue(s) express Cre recombinase. When bred to mice that express Cre recombinase, the resulting offspring will have the loxP or lox511 sites recombined, resulting in the inversion of the human DUX4-fl sequence, ending in a sense orientation (Figure 2).

The FLExDUX4 transgene sequence can be found here in ApE format and text format.

Hemizygous and homozygous mice have a very low but consistently detectable level of DUX4-fl expression in the absence of Cre Recombinase due to an antisense transcript at the Rosa26 locus. These mice exhibit alopecia, soft stool, mild inflammation, and minor muscle weakness. Homozygous are more affected, as are males compared to females. Hemizygous and homozygous male and female mice are very fertile and live normal lifespans.

The FLExDUX4 strain was generated in C57BL/6 from Charles River, however, we have backcrossed mice into the Taconic C57BL/6 background as well. We have not found any differences between the two backgrounds as of yet.

When bred to STOCK Tg(ACTA1-cre/Esr1*)2Kesr/J mice (Stock No. 025750), resulting double transgenic offspring express slightly more DUX4-Fl than the FLExDUX4 hemizygous mice prior to tamoxifen induction (TMX) but are still healthy (Figure 3). One TMX IP injection of 5 mg/kg leads to a useful mildly progressive FSHD model with mosaic expression of DUX4 in the skeletal muscles (Figure 4), while 2 injections of 10mg/kg leads to a very severe model and severe pathology in skeletal muscles (Figure 5).

FLExDUX4 mice have been bred to several Cre expressing lines (Table 1). Breeding FLExDUX4 mice to Sox2-Cre mice {JAX 008454 B6.Cg-Tg(Sox2-cre)1Amc/J}, results in an embryonic lethal. When bred to ACTA1-rtTA, tetO-cre mice {JAX 012433 B6;C3-Tg(ACTA1-rtTA,tetO-cre)102Monk/J}, resulting offspring are embryonic lethal. When bred to ACTA1-Cre {JAX 006149 B6.Cg-Tg(ACTA1-cre)79Jme/J}, resulting offspring are still-born at E19. When bred to UBC-CreERT2 {JAX 008085 B6.Cg-Tg(UBC-cre/ERT2)1Ejb/J}, offspring start showing a severe neurological phenotype at around 5 weeks of age without tamoxifen induction, perhaps due to leaky Cre expression and thus DUX4 expression in the brain, and need to be euthanized.