Behavioral and functional analysis of mouse phenotype:

SHIRPA, a proposed protocol for comprehensive phenotype assessment.

The material in the following pages gives full details of the SHIRPA protocol and contains material published in
following paper  (reproduced by permission):
  

Behavioral and functional analysis of mouse phenotype:
SHIRPA, a proposed protocol for comprehensive phenotype assessment.

Rogers DC, Fisher EM, Brown SD, Peters J, Hunter AJ, Martin JE

Mamm Genome(1997) 8(10), 711-713

Introduction to the SHIRPA protocol

SmithKline Beecham Pharmaceuticals
Harwell, MRC Mouse Genome Centre and Mammalian Genetics Unit
Imperial College School of Medicine at St Mary's
Royal London Hospital, St Bartholomew's and the Royal London School of Medicine
Phenotype
Assessment

For a more detailed description of each stage please select:  1^ry screen   2^ry screen   3^ry screen  (see also fig1. below)

Transgenic and gene targeting techniques using mice are increasingly common tools in the study of human gene function. Spontaneous and induced mouse mutations are also a critical resource in the characterisation of previously uncloned genes of biological importance, and the analysis of phenotypic differences between strains has become important, for example, in dissecting the effect of genetic background. A mouse model of human disease or gene function is, however, of limited value unless properly characterised and accurate phenotype assessment is the jewel in the crown of genetic manipulation. The vast majority of behavioural data in experimental animals have been obtained from rats, but increasingly structural and functional analysis is focused on the mouse. Phenotype analysis of mice has tended to be qualitative rather than quantitative in nature, with for example, reports of grooming, eating and handling response (Chang et al. 1993). Some groups have applied more sophisticated tests of locomotor and cognitive function to mice. For example, animals in which a portion of the mouse genome corresponding to human chromosome 21 was trisomic, have demonstrated learning deficits (Reeves et al. 1995), and gene targeted animals in which the amyloid precursor protein gene or the Huntington gene have been disrupted have also been demonstrated to display clear behavioural and functional deficits (Muller et al. 1994; Nasir et al. 1995). Such studies have usually been carried out in isolation, and these protocols make the assumption, often tacit, that other phenotypic variables are normal.

One key to accurate phenotype analysis in all species is to ensure that the initial stages of assessment are always performed impartially and that features are not missed, or findings anticipated or biased by making premature assumptions about end results. Behavioural, functional and morphological characterisation methods should ideally be directly comparable between different groups to build a useful body of data by which one can reproducibly assess the effect of a genetic manipulation or strain variation, the usefulness of a model and the effectiveness of a therapeutic intervention. A large number of mouse strains are in common use. There are considerable behavioural differences between strains, for example, in the performance of learning and memory tasks (Paylor et al. 1994) and response to acoustic startle (Paylor and Crawley in press). Obtaining appropriate control data is therefore critically important. In behavioural testing, a number of factors including nutritional state, circadian rhythm, oestrus cycle time, sex and age related differences may affect the outcome and must be taken into account in the design of experiments (Gower and Laberty 1993). An appreciation of the range of occasional findings in normal mice is also necessary at both behavioural and structural levels. To date, there is no agreed screening and testing protocol designed to identify and characterise disorders arising from abnormalities in the mouse genome. We propose that SHIRPA, a systematic, objective protocol for phenotype analysis can provide the framework to meet this need.

SHIRPA (Fig. 1) is a three stage protocol designed as a series of individual tests which in themselves provide quantitative data about an individual performance. Such test-specific performance is directly comparable between animals, over time and between groups. In addition, the tests also provide the opportunity to define abnormalities or variation in the mouse, with each test providing information about the pattern of function of a particular system, for example the brain and neuromuscular system. Collation and analysis of data from all the tests can be summed to give a specific profile of function (Table 1). This approach mimics the diagnostic process of general, neurological and psychiatric examination in humans. Measurement of locomotor activity, for example, gives data regarding the integrated function of cortical arousal, cerebral locomotor control, and neuromuscular function. This test can be taken alone as a coarse functional indicator, but when it forms part of a screening panel including tests of motor strength and neural control, such as grip strength and limb tone, it helps to define specific deficits. Thus it may become apparent that a poor performance in locomotor activity tests reflects a muscular weakness in the absence of central nervous system dysfunction. The functional tests are performed in a sequence that is easy to follow whilst disturbing the animal as little as possible.

SHIRPA utilises standardised protocols for behavioural and functional assessment which provide a sensitive measure for quantifying phenotype expression in the mouse. These paradigms can be refined to test the function of specific neural pathways, which will in turn, contribute to a greater understanding of neurological and affective disorders.

Table 1