Generation of mouse mutant lines for phenotyping (WP4)

Aims and Overview

The generation of a large number of mutant lines for phenotype analysis raises a considerable number of scientific and logistical issues. Through EUCOMM and EUMODIC a large number of mutant homozygotes will be generated. EUMODIC will focus on the phenotype assessment of adult viable homozygous mutant mice in the null configuration. A number of mutant lines will be lethal as homozygotes, however we do not propose to analyse these further beyond cataloguing their status as developmental lethal. Adult viable homozygotes will advance to primary phenotyping, and then a proportion of mice with interesting phenotypes will enter in-depth secondary phenotyping. We will develop a tracking database to track the generation of mice and their entry into both primary and secondary phenotyping pipelines. We initially discuss here the various issues surrounding the generation of mutant mice and then provide a detailed work plan for how this will be implemented.

Genetic Background

The EUCOMM mouse mutant resource will provide the lines for phenotype analysis. Initially, it is expected that this resource will be generated in 129 ES cells. However, early in EUCOMM's operations it is anticipated that production of mutant lines will move to BL/6 as more robust BL/6 ES cell lines become available. Irrespective of the genetic background of the ES cells, mice will be generated by crossing chimaeras to the same mouse strain, thus maintaining mutations on a pure inbred background for subsequent phenotyping. This is particularly important with respect to our plans for generating cohorts of mutants for phenotyping and the assessment of wild-type control mice (see below). If EUCOMM switches from the 129 background to BL/6 EUMODIC will switch to generating mice on the BL/6 background.  We will consider whether it will be scientifically interesting to generate some of the same lines that have already been phenotyped on the 129 background and repeat the phenotyping on the BL/6 background.  This will be balanced by the fact that it will reduce the total number of gene knock-outs addressed.

Generation of appropriately sized cohorts of mutant lines for phenotyping

The need to generate substantial cohorts of mice from any mutant line in order to be confident of detecting disease phenotypes requires the development of innovative approaches to both mouse generation and analysis. We have tackled two significant issues with regard to mouse generation and analysis:

1.     The cohort size requirements for primary phenotype analysis

2.     The analysis of control wild-type mice

Cohort Size:

As is discussed in detail below (WP1), primary phenotyping protocols generally require a minimum of ten mice to detect a disease phenotype taking into account the critical variances that reflect the human disease condition. For this reason, for the bulk of EMPReSSslim tests we aim to phenotype at least ten mice. Moreover, it is clearly important to assess the phenotypes in both male and females. EMPReSSslim has been refined to two pipelines, each requiring ten males and ten females, representing a total of 40 age-matched mutant mice that will need to be generated for each line assessed. To develop such a cohort by standard breeding approaches, involving heterozygous or homozygous matings, would require very substantial time and resources. We propose instead that homozygous mice will be generated by IVF that allows the rapid production of large numbers of age-matched mutant mice and offers very significant economies of scale.

Control mice:

Mouse phenotyping of transgenics and knock-outs to date has by and large employed wild-type sibs abstracted from matings that generate the mutant homozygotes (generally intercrosses of heterozygous mutants). However, we have ruled out the intercross approach as being too unwieldy and too costly for generating sufficient age-matched mice. We therefore propose to adopt a solution that is a feature of large-scale ENU mutagenesis phenotype screens, whereby control data is abstracted from a running base-line of observations on wild-type mice. Our approach also takes advantage of the uniform genetic background of the mutant mice generated (see above) and enables a direct comparison of mutants with a control population of mice generated from the same inbred strain. Thus, each primary phenotyping centre will enter into EMPReSSslim a cohort of ten males and ten females every two months, generating a running baseline of wild-type data to be compared with the ongoing datasets generated from the mutant cohorts. We recognise that this entails a new, but innovative, approach to mouse mutant phenotyping but one that reflects our experience in large-scale ENU programmes where control running base-line data with low variances was routinely produced and effectively utilised. We will adopt this approach but at the same time continuously document control variances within and between phenotyping centres, assessing the standardisation of the EMPReSSslim tests that we use and ensuring the comparability of phenome data acquired. In addition, we will also validate this approach for two mutant lines in the earliest stages of the programme. We will compare phenotype data produced from IVF cohorts and control baselines with data gathered from mutant and wild-type sibs generated through heterozygote intercrosses

Archiving and Dissemination

Each primary phenotyping centre will be involved in generating the relevant mutant lines that are to be entered into their primary phenotyping pipeline. However, these mutant lines also need to be disseminated to secondary phenotyping centres. The use of IVF to generate mutant homozygotes offers additional economies of scale here, since the IVF can be used to also collect embryos to establish a frozen archive of mutant homozygotes. This frozen archive at each of the 4 primary phenotyping centres will be the source of embryos for delivering and rederiving mice at secondary phenotyping centres. In addition, all of the mouse lines will be archived in EMMA and made available to the wider biomedical research community.  Lines that are homozygous lethal will be archived as heterzygotes.