A PiggyBac-based recessive screening method to identify pluripotency regulators

Abstract

Phenotype driven genetic screens allow unbiased exploration of the genome to discover new biological regulators. Bloom syndrome gene (Blm) deficient embryonic stem (ES) cells provide an opportunity for recessive screening due to frequent loss of heterozygosity. We describe a strategy for isolating regulators of mammalian pluripotency based on conversion to homozygosity of PiggyBac gene trap insertions combined with stringent selection for differentiation resistance. From a screen of 2000 mutants we obtained a disruptive integration in the Tcf3 gene. Homozygous Tcf3 mutants showed impaired differentiation and enhanced self-renewal. This phenotype was reverted in a dosage sensitive manner by excision of one or both copies of the gene trap. These results provide new evidence confirming that Tcf3 is a potent negative regulator of pluripotency and validate a forward screening methodology to identify modulators of pluripotent stem cell biology.

Figure 1. Generation of Rex1 reporter cells.

A. qRT-PCR analysis of Rex1 and Oct4 mRNA during monolayer differentiation in N2B27. B. Strategy to create the Rex1^GIP knock in allele. C. Flow cytometry of a representative Rex1-Egfp profile in undifferentiated NN97-5 cells. E. Flow cytometry of Rex1-Egfp population in NN97-5 cells during monolayer differentiation in N2B27.

Figure 2. piggyBac mutagenesis and monolayer differentiation screen.

A. Binary piggyBac gene trap system composed of gene trap vector, pGG85, and transposase expressing helper plasmid, pCAGG-PBase. B. G418 resistant colonies produced by co-electroporation of 1 µg of pGG85 and 3 µg of helper plasmid. C. Splinkerette PCR amplified genome junction flanking PB insertions indicating the number of PB inserts in each clone. D. Schematic representation of monolayer differentiation screen.

Figure 3. Gene trap mutants from monolayer differentiation screen.

A. Images show typical differentiated and non-differentiated morphologies after 7 days monolayer neural differentiation assay. P1-1, P1-2, P1-4, P1-11, P1-12, P1-19 and P1-20 are clones carrying Tcf3 gene trap mutation. B. Splinkerette-PCR amplified genome junctions flanking PB inserts. Gel images showing the genome junction flanking PB 5′ terminal repeat region (5′TR) and 3′ terminal repeat region (3′TR). Arrows indicate that a 500 bp 3′TR fragment and a 300 bp 5′TR fragment were amplified in multiple clones. Sequencing locates this band to Tcf3 locus.

Figure 4. Tcf3 gene trap mutants.

A. Tcf3 gene trap (Tcf3^trp) and Cre-reverted (Tcf3^rev) alleles. Cre recombination deletes the gene trap cassette to leave a reverted allele retaining the PB terminal repeats. B. RT-PCR analysis of Tcf3 expression in gene trap mutants. Tcf3 mRNA was not detected in clones P1-2, P1-12 and P1-19 but evident in clones P1-1, P1-11 and P1-20. C. Diagram showing generation of het or homozygous reverted cells. D. RT-PCR analysis of Tcf3 gene trap (Ex3-SA) and Tcf3 wild type (Ex3-Ex7) transcripts. CreA12-1 and CreD10-4 are subclones of CreA12 and CreD10. E. qRT-PCR analysis of Tcf3 expression. F. After 9 days monolayer differentiation multiple ES cell colonies formed from Tcf3 homozygote P1-2, but not from parental NN97-5 or revertant CreA12 or CreD10 cells.

Figure 5. Tcf3 deficiency suppresses serum-induced differentiation.

A. Parental NN97-5 cells differentiate after 4 days in serum without LIF while Tcf3 gene trap mutant P1-2 cells remain undifferentiated and retain uniform Oct4 expression in serum. B. Flow cytometry analysis for Rex1-EGFP positive cells during monolayer differentiation in serum. P1-2, Tcf3 gene trap mutant; CreA12, heterozygous Tcf3 Cre-revertant; CreD10, homozygous Tcf3 Cre-revertant. Graph shown is a representative of two independent experiments. C. Tcf3 mutant (P1-2) and the Tcf3 reverted cells were plated at single cell density in serum with or without LIF for colony forming assay. Colonies were stained after 9 days for alkaline phosphatase (AP) activity and colony numbers were quantified manually. Undifferentiated colonies are showing in red in figure and partially differentiated showing in green and differentiated showing in yellow. D. Images show typical AP positive morphologically undifferentiated ES cell colonies generated by P1-2 cells in serum with or without LIF. The experiment has been repeated once.

Figure 6. siRNA knockdown of Tcf3 in Blm wild type cells.

A. qRT-PCR analysis of Tcf3 knockdown in Tcf3 siRNA treated ES cells and control siRNA treated cells. B. Graph shows population of Rex1-EGFP positive cells in Tcf3 siRNA and control siRNA treated cells after 2 days in monolayer differentiation with or without serum. C. qRT-PCR analysis of Rex1 expression in Tcf3 siRNA or control siRNA treated cells in monolayer differentiation with or without serum. D. Images showing a typical AP positive ES cell colony formed in Tcf3 siRNA treated cells after 5 days in serum while only differentiated colonies formed from control siRNA treated cells. Error bar represents standard deviation from three individual plating.

Figure 7. Tcf3 mutation has subtle molecular consequences.

A. Relative gene expression analysis by qRT-PCR in Tcf3 mutant (P1-2) compared to Cre-reverted (CreD10)(Blue column) and wild type (NN97-5)(Red column). B. TOPFlash assay of Tcf-mediated transcriptional activation. None, N2B27 alone; Wnt, Wnt3A; Wnt+LIF, Wnt3A plus LIF. C. qRT-PCR analyses of Wnt target gene expression. S+L, Serum plus LIF. D. Immunoblotting analysis of Nanog and Oct4 protein expression in serum and LIF. E. NN97-5 and P1-2 cells cultured in serum and LIF immunostained for Oct4 and Nanog. Images show typical heterogeneous Nanog protein expression in NN97-5 cells compared to more uniform staining in P1-2 cells. F. Mean nuclear staining intensity of Oct4 and Nanog in individual cell was quantified using Cell profiler software and presented as a scatter plot using Microsoft Excel. 1600 cells were scored for NN97-5 and 2172 cells for P1-2. The experiment has been repeated three times.