Characterization of NFH-LacZ transgenic mice with the SHIRPA primary screening battery and tests of motor coordination, exploratory activity, and spatial learning

Abstract

NFH-LacZ transgenic mice express a fusion protein between a truncated form of the endogenous neurofilament of heavy molecular weight and the complete E. coli β-galactosidase. NFH-LacZ transgenic mice could be distinguished from controls in the SHIRPA neurological battery by the appearance of action tremor and hindlimb clasping and a lower body weight. Despite normal exploratory activity and spatial learning, NFH-LacZ transgenic mice were deficient in stationary beam, coat-hanger, and rotorod tests of motor coordination. These results are concordant with neuropathological findings in spinal motoneurons and the cerebellum and indicate that despite the absence of paralysis, these transgenic mice may serve as an experimental model of the early stage of amyotrophic lateral sclerosis.

Introduction

The growing number of transgenic and knockout mice has increased our understanding on the behavioral effects of specific proteins (Anagnostopoulos et al., 2001, Surjo and Arndt, 2001). The use of a wide array of measurements should be encouraged, as genetic modifications often result in test-specific deficits. On the other hand, test standardization ensures a higher degree of interlaboratory replicability (Moldin et al., 2001). It is for these reasons that the SHIRPA (SmithKline/Harwell/Imperial College/Royal Hospital/Phenotype Assessment) test battery was presented (Rogers et al., 1997) and developed through large-scale analysis of mutations induced by N-ethyl-N-nitrosourea (ENU). The primary screen consists of quantitative and semi-quantitative evaluation of reflexes and basic sensorimotor functions as well as measurements of body weight and size. Secondary and tertiary screens encompass more detailed analyses of motor coordination, spatial orientation, and pain sensitivity. The SHIRPA protocol has been used in the description of several mouse strains (Rogers et al., 1999), ApoE and coloboma (cm) knockout mice (Hatcher et al., 2001), and the legs at odd angles (Loa) ENU-induced mutation (Rogers et al., 2001).

We have reported on the sensorimotor functions of transgenic mice with a fusion protein between the endogenous neurofilament protein of heavy molecular weight (NFH) and E. coli β-galactosidase (Lalonde et al., 1999, Dubois et al., 2002). The genetic modification causes the accumulation of neurofilaments in the cell body as opposed to their usual localization in axons, reduces axonal diameters in the peripheral and central nervous system, and reduces Purkinje cell number (Eyer and Peterson, 1994, Tu et al., 1997). This neuropathology resembles amyotrophic lateral sclerosis (ALS) (Pioro and Mitsumoto, 1996, Julien, 1997, Munoz et al., 1988). NFH-LacZ mice were impaired in several sensorimotor tests by comparison to normal mice of the B6C3 hybrid strain (Lalonde et al., 1999) as well as C3H and FVB strains (Dubois et al., 2002). The purpose of the present investigation was to determine whether the SHIRPA protocol is sensitive to the neural abnormalities observed in NFH-LacZ mice and whether the same behavioral phenotype can be reproduced on the DBA/2 background during testing in an open-field, an activity chamber, and three motor coordination tasks (stationary beam, coat-hanger, and rotorod).

Contrary to frontotemporal dementia with motor neuron disease causing hypometabolism in cortical association areas (Garraux et al., 1999), ALS patients are not demented. Nevertheless, impaired problem-solving (Strong et al., 1999) and working memory (Abrahams et al., 2000) abilities, as well as emotional lability (Newsom-Davis et al., 1999), giving rise to pathological laughing or crying when associated with pseudobulbar symptoms (McCullagh et al., 1999, Strong et al., 1999), have been described in ALS. It is, therefore, of interest to examine cognitive and emotional aspects of ALS-like murine models.

We determined whether NFH-LacZ mice differ from DBA/2 controls in three tests of exploration as well as spatial learning. Exploration was first evaluated with the spontaneous alternation test in a T-maze. After being forced to enter a maze arm, mice or rats usually choose the opposite arm, thereby exploring novel stimuli (Dember and Fowler, 1958). This test is sensitive to lesions of multiple brain regions, including the cerebellum, as well as induced mutations (Lalonde, 2002). Exploration tensencies were further evaluated in the elevated plus-maze test of anxiety (File, 2001, Pellow et al., 1985), as mice tend to explore safer (enclosed) arms as opposed to more anxiogenic (open) arms, taking into account the normal rodent response of avoiding open spaces (agoraphobia). The elevated plus-maze has been pharmacologically validated in DBA/2J mice for the GABAergic synapse, as diazepam decreased anxiety whereas picrotoxin increased it (Dalvi and Rodgers, 1996). This test is sensitive to several induced mouse mutations (Clement et al., 2002, Holmes, 2001). In the final method of investigating exploration, the emergence test was used (Holson, 1986, Holmes, 2001). This test evaluates the tendency of mice to remain inside a small safe compartment as opposed to a larger more anxiogenic one and is sensitive to induced mutations (Holmes, 2001). Lesions of the medial frontal cortex increased the time taken before rats emerged as well as elevating anxiety levels in several other tests, underlining the convergent validity of this measure (Holson, 1986). Spatial learning was assessed in the Morris maze, in which mice were trained to escape from a pool of water by reaching a platform (Morris et al., 1982). This test is sensitive to several mutations, including NFH-LacZ mice when compared to B6C3 controls (Lalonde et al., 1999).