Reader Comments
Post a new comment on this article
Post Your Discussion Comment
Please follow our guidelines for comments and review our competing interests policy. Comments that do not conform to our guidelines will be promptly removed and the user account disabled. The following must be avoided:
- Remarks that could be interpreted as allegations of misconduct
- Unsupported assertions or statements
- Inflammatory or insulting language
Thank You!
Thank you for taking the time to flag this posting; we review flagged postings on a regular basis.
closeFlaws in this paper
Posted by afstewart on 17 Feb 2007 at 17:07 GMT
The result shown in Figure 1 with FLPe is wrong. A stable selection protocol like the one used here will give nearly 100% recombination with FLPe. Indeed even wtFLP will give nearly 100% recombination in a stable experiment. The authors equate their experiments with experiments in reference 8 to confirm that FLPe gives only 5% recombination with mosaicism. However the experiments of reference 8 were transient expression experiments without selection, not stable expression experiments with selection as performed here. These two types of experiments cannot be compared. Furthermore, the authors show no evidence that codon alteration improved FLP mRNA or protein expression levels. For example, if exisiting FLPe constructs contained cryptic splicing signals, these can be detected by RT-PCR. Minimally, improved protein expression levels are needed to draw the conclusions presented. Rather, the conclusions are based on a fundamentally flawed experiment that has no control. The simplest explanation of the FLPe result of Figure 1 is that the expression construct is damaged. Hence there is basis to conclude that FLPo is better than FLPe. It may be, but these experiments do not permit that conclusion.
Other concerns
MHafner replied to afstewart on 28 Feb 2007 at 12:25 GMT
I appreciate the efforts to generate a codon optimized phiC31 integrase because we were thinking about the same. However I have some concerns about this paper:
I second Francis’ concerns: As with flpe the results on phiC31 integrase published by Ralf Kühn’s group (Andreas et al. Nucleic Acids Res. 30(11):2299-306) were based on transient transfection assays. Stable transfection relies on integration in appropriate loci that allow good expression. Taken the number of analyzed clones into account I admit though that it is unlikely that the different numbers of positive phiC31 and phiC31o clones are due to different insertion sites. However, one issue that wasn’t addressed by Raymond and Soriano are possible differences in copy number (according to the Materials and Methods section they only genotype the clones by PCR).
Another issue I think needs deeper analyses are possible chromosomal aberrations induced by phiC31. Just one generation of pubs and the lack of obvious deleterious effects may be insufficient to make any judgement. Chromosomal rearrangements may stay hidden as long as the mice are backcrossed with wild type mice. Indeed, one should setup crosses of siblings or actually do some cytogentics.
Currently, the phiC31o mouse seems to be not really suitable (at least for our purposes which is phiC31 mediated resistance marker deletion) because according to the picture one rather induces mosaic then ubiquitous deletion. Unfortunately, there are not too many question for which such mice are needed (you may look up how often the balancer Cre strain from the Rajewsky lab has been used). Indeed, for ubiquitous deletion one would rather like to have expression in the zygote or even better limit expression to the zygote to omit putative deleterious effects later in development.
BTW, a good flpe deleter strain is available and we never had problems with flp mediated deletion in ES cells when we expressed the protein from a vector with a puromycin resistance gene (2µg/ml Puro for two days).
Still, I believe, flpo recombinase and phiC31o integrase may be useful in the future.