Silencing inhibits Cre-mediated recombination of the Z/AP and Z/EG reporters in adult cells

Abstract

Background: The Cre-loxP system has been used to enable tissue specific activation, inactivation and mutation of many genes in vivo and has thereby greatly facilitated the genetic dissection of several cellular and developmental processes. In such studies, Cre-reporter strains, which carry a Cre-activated marker gene, are frequently utilized to validate the expression profile of Cre transgenes, to act as a surrogate marker for excision of a second allele, and to irreversibly label cells for lineage tracing experiments.

Principal findings: We have studied three commonly used Cre-reporter strains, Z/AP, Z/EG and R26R-EYFP and have demonstrated that although each reporter can be reliably activated by Cre during early development, exposure to Cre in adult hematopoietic cells results in a much lower frequency of marker-positive cells in the Z/AP or Z/EG strains than in the R26R-EYFP strain. In marker negative cells derived from the Z/AP and Z/EG strains, the transgenic promoter is methylated and Cre-mediated recombination of the locus is inhibited.

Conclusions: These results show that the efficiency of Cre-mediated recombination is not only dependent on the genomic context of a given loxP-flanked sequence, but also on stochastic epigenetic mechanisms underlying transgene variegation. Furthermore, our data highlights the potential shortcomings of utilizing the Z/AP and Z/EG reporters as surrogate markers of excision or in lineage tracing experiments.

Figure 1. Genomic organization of Cre-reporter transgenes.

In the Z/AP strain (A) and Z/EG strain (B), transcription of a β-galactosidase/neomycin phosphotransferase fusion gene (β-geo) is driven by a hybrid CMV/β-actin promoter (pCAGGS) and terminated by polyadenylation sites (white boxes). Cre recombines loxP sites (white triangles) to remove the β-geo gene, activating expression of human placental alkaline phosphatase (hPLAP) or enhanced green fluorescent protein (EGFP) in the Z/AP and Z/EG strains respectively. In the R26R-EYFP strain (C), Cre-mediated recombination of loxP sites removes a phosphoglycerate kinase (PGK) promoter and a neomycin phosphotransferase gene (NeoR), allowing the endogenous ROSA26 genomic locus to drive expression of the downstream enhanced yellow fluorescent protein (EYFP) gene.

Figure 2. Activation of the EGFP reporter is inefficient in the blood of adult TIE2-tTA/Tet-O-Cre-/ZEG mice.

EGFP expression in the blood of a representative triple transgenic mouse, 3, 6 and 9 months following removal of doxycycline from the diet is shown in (A–C) respectively. EGFP expression in the blood of a representative triple transgenic mouse, bred in the absence of doxycycline is shown in (D). In all triple transgenic mice analyzed, activation of the EGFP reporter was not observed in peripheral blood leukocytes (PBL) earlier than 6 months following removal of doxycycline and remained low at nine months post induction (E).

Figure 3. Activation of the Cre-reporter gene is less efficient in Z/AP and Z/EG mice than in R26R-EYFP mice.

All reporter strains were crossed to the general deleter strain, pCX-NLS-Cre, and activation of each reporter gene was assessed by flow cytometry of peripheral blood leukocytes at 12 weeks of age (A). All reporter strains were also crossed to the myeloid-specific LysM-Cre strain, and activation of each reporter gene was assessed by flow cytometry of peripheral blood granulocytes at 12 weeks of age (B). Representative histograms demonstrating reporter expression in peripheral blood granulocytes are shown in (C–E) for LysM-Cre/R26R-EYFP, LysM-Cre/ZAP and LysM-Cre/Z/EG mice respectively (solid lines). Reporter expression in BL/6 mice is shown in dotted lines.

Figure 4. Mechanisms contributing to the inefficient labeling efficiency of Z/AP and Z/EG cells.

Gr-1 positive cells from a LysM-Cre/Z/EG mouse were sorted into reporter negative (−) and reporter positive (+) populations (A). Genomic DNA from these populations was then analyzed by PCR utilizing primers binding within the pCAGGS promoter and EGFP cDNA. This combination of primers (Exc) generates a 240bp fragment in the presence of the recombined locus whereas the distance across the intact locus is too large to facilitate exponential amplification under the conditions used (B). Primers recognizing the IL-2 inducible T-cell kinase (ITK) gene were also utilized as a positive control in reporter negative (−) and reporter positive (+) reactions as well as in a no-template (NT) control. Gr-1 positive cells from a LysM-Cre/Z/AP mouse were also sorted into reporter negative (−) and reporter positive (+) populations (C). Genomic DNA from reporter negative and reporter positive populations was subjected to bisulfite sequencing in order to determine the methylation status of a 200 bp region of the pCAGGS promoter. This region contains 36 CpG dinucleotides, significantly more of which were methylated (filled circles) in clones derived from reporter negative cells than in those derived from reporter positive cells (D,E). The degree of methylation at the pCAGGS and ROSA promoters in both reporter negative (−) and reporter positive (+) populations is shown in (F) and (G) for cells derived from LysM-Cre/ZEG and LysM-Cre/Z/AP mice respectively. The degree of methylation at the ROSA promoter in cells derived from the LysM-Cre/R26R-EYFP strain is shown in (H). Silencing of the Z/AP and Z/EG transgenic loci was demonstrated via quantification of the percentage of peripheral blood leukocytes expressing β-galactosidase by FDG staining (I).