Quantification of Cre-mediated recombination by a novel strategy reveals a stable extra-chromosomal deletion-circle in mice

Abstract

Background: Inducible conditional knockout animals are widely used to get insight in the function of genes and the pathogenesis of human diseases. These models frequently rely on Cre-mediated recombination of sequences flanked by Lox-P sites. To understand the consequences of gene disruption, it is essential to know the efficiency of the recombination process.

Results: Here, we describe a modification of the multiplex ligation-dependent probe amplification (MLPA), called extension-MLPA (eMLPA), which enables quantification of relatively small differences in DNA that are a consequence of Cre-mediated recombination. eMLPA, here applied on an inducible Pkd1 conditional deletion mouse model, simultaneously measures both the reduction of the floxed allele and the increase of the deletion allele in a single reaction thereby minimizing any type of experimental variation. Interestingly, with this method we were also able to observe the presence of the excised DNA fragment. This extra-chromosomal deletion-circle was detectable up to 5 months after activation of Cre.

Conclusion: eMLPA is a novel strategy which easily can be applied to measure the Cre-mediated recombination efficiency in each experimental case with high accuracy. In addition the fate of the deletion-circle can be followed simultaneously.

Figure 1

The eMLPA strategy. (A) The position of the probes A, B, C, and D in case of the Pkd1^lox allele or the Pkd1^del2-11 allele after hybridization. (B) The gap between probes is filled in using Stoffel-Taq Polymerase. (C) The remaining nick is ligated by ligase, completing the formation of the templates AB, CD, and AD which are amplified simultaneously in a PCR reaction containing a fluorescently labeled forward primer.

Figure 2

The AB, CD and AD peaks obtained after capillary gel electrophoresis. When only the Pkd1^lox allele is present two peaks, AB and CD, can be observed (Upper panel). A single peak, AD, is observed when only the Pkd1^del2-11 allele is present (Middle panel). When the Pkd1^lox and the Pkd1^del2-11 allele are both present, all three peaks can be observed (Lower panel).

Figure 3

Outline for calculating the percentage of deletion allele. The percentage of Pkd1^del2-11 for a sample s1 is used as an example. (A) The peak-ratios (P) for all controls and sample s1 is calculated by dividing the height of the deletion peak with the sum of the lox peaks. (B) A fifty percent reference (Ref.) is calculated by taking the median of the peak-ratios from all controls. (C) The peak-ratio for sample 1 (P_s1) is divided by and thus normalized to the fifty percent reference. The ratio of the Pkd1^del2-11 allele to the Pkd1^lox allele in sample 1 is, R_s1 to 1. (D) Thus, the fraction of the Pkd1^del2-11 allele relative to the sum of the Pkd1^del2-11 and Pkd1^lox alleles is R_s1 divided by R_s1 plus 1. The total percentage of Pkd1^del2-11 in sample 1 (D_s1) can now be calculated by multiplying this fraction with the maximum percentage (M) the Pkd1^del2-11 and Pkd1^lox alleles can have together. M is either 50% when initially one Pkd1^lox and one Pkd1⁺ allele are present or a 100% when both alleles were either Pkd1^lox or Pkd1^del2-11.

Figure 4

Quantification of the percentage of Pkd1^del2-11 allele by eMLPA. To validate the eMLPA strategy, the percentage of Pkd1^del2-11 allele is shown for eight Pkd1^{del2-11, lox} controls with an expected value of 50%, and for a dilution series ranging from a hundred to zero percent of Pkd1^del2-11 allele, obtained by diluting Pkd1^{del2-11, del2-11} with Pkd1^{lox, lox} DNA at different ratios. All controls and samples from the dilution series gave the expected results within a five percent interval. The percentage of Pkd1^del2-11 allele is also measured in adult or neonatal tamoxifen treated Cre; Pkd1^lox,+, Cre; Pkd1^{del2-11, lox} and Cre; Pkd1^{lox, lox} mice. The error bars represent the standard error of the mean. Hybridizations were performed three times and PCR's were performed in triplicate on each hybridization.

Figure 5

Characterization of experimental errors in eMLPA and qPCR. (A) Correlation of eMLPA and qPCR. To verify eMLPA, data obtained from samples analyzed with both qPCR and eMLPA are plotted against each other. (B) Inter-experimental variation in heterozygous control (Pkd1^{del2-11, lox}) mice is presented. Raw data from five different eMLPA or qPCR experiments were combined. The percentage of Pkd1 deletion, which should be exactly 50%, was calculated relative to the median of all measurements. An eMLPA experiment consists of two or three hybridizations on which a single PCR was performed and a qPCR experiment consists of a triplo PCR. Each point shown is an average of these duplo or triplo measurements and the error bars represent the standard error of the mean. (C) The performance of eMLPA at low DNA concentrations. Single hybridizations were carried out on 500 ng, 25 ng or 10 ng of DNA from eight Pkd1^{del2-11, lox} mice, followed by extension, ligation and 29, 33 or 35 cycles of amplification respectively. Pkd1^del2-11 percentages were calculated relative to the median from the 500 ng hybridizations which was used as the 50% reference. From the eight Pkd1^{del2-11, lox} samples in each group; the 500 ng, 25 ng and the 10 ng hybridizations, the median and standard deviations from the Pkd1^del2-11 percentages are shown. Whereas the median from both groups at the lower DNA concentrations is close to the expected 50%, the variation in these groups is higher.

Figure 6

Detection of the deletion-circle as a result of Cre-mediated recombination. (A) An example of the four observed peaks in an adult tamoxifen treated Cre; Pkd1^{del2-11, lox} mouse, three peaks corresponding to the Pkd1^lox and Pkd1^del2-11 alleles plus an additional 144 bp peak. (B) The possible locations of probes B and C on the Pkd1^lox allele and on the deletion-circle. Probes B and C can only form a product when the floxed fragment is excised and circularized by Cre. (C) Hybridization of probes B and C followed by extension, ligation, 36 cycles of amplification and agarose-gel-electrophoresis does not result in a product when DNA from two Pkd1^{del2-11, lox} mice was used, but when DNA from kidneys from two adult tamoxifen treated Cre; Pkd1^{del2-11, lox} mice was used as a template, a distinct product of 144 bp was observed.