Enhanced efficiency through nuclear localization signal fusion on phage PhiC31-integrase: activity comparison with Cre and FLPe recombinase in mammalian cells

Abstract

The integrase of the phage PhiC31 recombines an attP site in the phage genome with a chromosomal attB site of its Streptomyces host. We have utilized the integrase-mediated reaction to achieve episomal and genomic deletion of a reporter gene in mammalian cells, and provide the first comparison of its efficiency with other recombinases in a new assay system. This assay demonstrated that the efficiency of PhiC31-integrase is significantly enhanced by the C-terminal, but not the N-terminal, addition of a nuclear localization signal and becomes comparable with that of the widely used Cre/loxP system. Furthermore, we found that the improved FLP recombinase, FLPe, exhibits only 10% recombination activity on chromosomal targets as compared with Cre, whereas the Anabaena derived XisA recombinase is essentially inactive in mammalian cells. These results provide the first demonstration that a nuclear localisation signal and its position within a recombinase can be important for its efficiency in mammalian cells and establish the improved PhiC31-integrase as a new tool for genome engineering.

Figure 1

Comparison of Cre recombination activity with other site-specific recombinases. (A) Schematic diagram of the dual reporter plasmid facilitating direct comparison of Cre and RecX activities. ePr, eukaryotic promoter; pac, puromycin resistance gene; recX, recognition site for second recombinase (RecX). (B) CHO cells were transiently co-transfected with 50 ng of luciferase standard, 50 ng of the respective dual reporter plasmid and either between 0.5 and 100 ng of pCMV-NLS-Cre or the corresponding amount of RecX expression plasmid (pCMV-NLS-FLPe, pCMV-XisA, pCMV-NLS-XisA, pCMV-φC31, pCMV-NLS-φC31 or pCMV-φC31-NLS). Cells were harvested 48 h after transfection and β-gal and luciferase activities were measured. For each recombinase, at least two independent experiments (with quadruplicate values each) were carried out. After normalization to luciferase activity, the β-gal activity was calculated as percentage of the Cre activity, whereby only datapoints that were within the linear range of the assay were taken into account. The average and the standard deviations (error bars) of these values are displayed.

Figure 2

Quantitative analysis of φC31-Int activity on stably integrated substrate. (A) An NIH 3T3 cell clone carrying a single copy integration of the Cre/φC31-Int reporter (clone 3) was transfected with the indicated amounts of pCMV-NLS-Cre or φC31-Int expression vectors (pCMV-φC31, pCMV-NLS-φC31 or pCMV-φC31-NLS). Forty-eight hours after transfection the cells were harvested and β-gal activity was measured. Results show the averages from two experiments with duplicate values. Error bars represent standard deviations. (B) NIH 3T3 clone 3 was transfected with either 50 ng of pCMV-NLS-Cre expression plasmid or 50 ng of pCMV-φC31-NLS expression plasmid. Cells were harvested 4, 8, 12, 24, 30, 48 or 72 h after transfection and β-gal activities were measured. (C) X-Gal staining of mock-, Cre- or φC31-NLS-transfected NIH 3T3 cells. Representative areas are shown.

Figure 3

Quantitative analysis of FLPe activity on stably integrated substrate. An NIH 3T3 cell clone carrying a single copy integration of the Cre/FLP reporter was transfected with the indicated amounts of pCMV-NLS-Cre or pCMV-NLS-FLPe. Forty-eight hours after transfection the cells were harvested and β-gal activity was measured. Results show the averages from two experiments with duplicate values. Error bars represent standard deviations.