Beware of your “oocyte specific” Cre line: somatic cell Cre expression in several <i>Zp3-Cre</i> lines and the <i>Gdf9-iCre</i> transgenic line

Several zona pellucida 3 (Zp3)-Cre driver mouse lines are used to enable conditional loss-of-function studies in oocytes. The Knowles Zp3-Cre line, currently maintained live at the Jackson Laboratory, is the most widely published. We recently found that the transgene expressed in the Knowles line contains a truncated metallothionein (Mt1) sequence that is expressed at high levels in oocytes from transgenic mice. This finding led us to search for an alternative Zp3-Cre line that did not express an exogenous Mt1 transcript. We tested a second transgenic Zp3-Cre line and then created our own transgenic Zp3-Cre line, neither of which was oocyte-specific as documented by crossing to the tdTomato reporter line. Similar testing confirmed the exquisite oocyte specificity of the Knowles Zp3-Cre line. An alternative to Zp3-Cre is to use the Gdf9-iCre line, which is reported to be oocyte-specific and expressed beginning at the primordial follicle stage. This line similarly showed high somatic tissue Cre expression. Reasoning that a knock-in approach would ensure oocyte specificity, we used a CRISPR/Cas9 approach to insert Cre into the endogenous Zp3 locus. However, crosses of this knock-in line with tdTomato females revealed high somatic tissue expression. Both the Zp3-Cre knock-in and Gdf9-iCre alleles, when paternally inherited, induced Cre expression by the blastocyst stage, explaining the broad tissue distribution. We conclude that the Knowles Zp3-Cre transgenic line remains the best model for generating oocyte-specific expression, though investigators should be aware of the spurious Mt1 expression from the transgene.