of EGFPd2 fluorescence in up to 75% of EGFPd2 homozygous cells (Fig. 7E,F, Fig. S8A,B). Interestingly, co-expression of either two or three copies of the same EGFP gRNA cassette from the same AAVS1 locus was sufficient to significantly increase the speed and efficiency of inducible EGFPd2 knockout in all the clonal sublines analyzed (Fig. 7H,I, Fig. S8A). For instance, simultaneous induction of three copies of the same gRNA resulted in a remarkable 95% knockout efficiency following tetracycline treatment (Fig. 7I). Importantly, inducible EGFPd2 knockout hESCs did not show any significant decrease in the proportion of EGFPd2-positive cells nor in their fluorescence after prolonged culture in the absence of tetracycline, even when several gRNA copies were used (Fig. 7G, Fig. S8C,D). This demonstrated that the inducible gRNA expression was tightly controlled. Finally, testing of additional gRNAs against EGFPd2 revealed that the speed and efficiency of the inducible knockout strongly relied on the gRNA. Indeed, an optimal sequence allowed up to 90% knockout after only 2 days of induction (Fig. S8E,F,K). Of note, the most efficient gRNA also resulted in uncontrolled EGFPd2 knockout (Fig. S8G), but this limitation was avoided by simply adding a second TET operon to the inducible H1 promoter
