Mouse Adaptation of Enterovirus 71 increases viral virulence in a NOD-SCID mouse infection model

G. Costello¹, E. J. Growcott², C. S. Osborne². ¹University of Leeds, Leeds, ²Novartis Institutes for BioMedical Research Inc, Emeryville.

Introduction: Development of novel therapies for treating opportunistic human rhinovirus infection (HRV) is hindered by the lack of robust in vivo preclinical models of viral replication (1), (2). Enterovirus 71 (EV71), like HRV, is a virus in the picornaviridae family and here, we characterized an infection model using EV71 as a surrogate for HRV.

Method: Animal studies were approved by the Novartis Emeryville IACUC. Seven day old mixed sex NOD SCID mouse pups (≥5 g) were inoculated via intraperitoneal (i.p.) administration of 20 mL/kg 5x10⁵ pfu/mL EV71 (American Type Culture Collection VR-1775). A single litter was used for each group (n=5-9). The smallest litter was sham infected with 20% glycerol to establish baseline growth and movement. Animals were observed for clinical condition, activity and posture. Humane end points were designated based on clinical observations (described in the results) and/or failure to gain body weight. Animals were euthanized by i.p. administration of 195 mg/kg Euthasol (Virbac Animal Health, Fort Worth, Tx, USA) and a whole body perfusion performed with phosphate buffered saline (PBS), before thigh, spinal cord, brain, lung, heart, spleen, kidney and intestine tissue were collected to measure viral load using quantitative RT-PCR. Passaged strains were generated by harvesting thigh tissue from EV71-infected animals and preparing a 10% (w/v) suspension in PBS by homogenisation and centrifugation, with the supernatant collected used for re-infection (VR-1775-P1). The process was repeated (VR-1775-P2) and the infection kinetics between the strains compared. Statistical analysis was undertaken using a Kruskal Wallis ANOVA followed by a Dunn’s multiple comparison test.

Results: Infection resulted in the development of clinical signs including piloerection and lameness that progressed to bi-lateral hind limb paralysis in all animals. Survival duration was viral strain dependent; all pups infected with VR-1775-P2 were euthanized by 3 days post infection compared to 8 days with VR-1775 infection. Virus was recovered from all tissues analysed, with the highest viral burden in the thigh and spinal cord. The viral titer in tissues harvested from animals infected with VR-1775-P2 was significantly higher than VR-1775 infected animals (Table 1).

Conclusion: ATCC VR-1775 passage increased EV71 virulence. Further characterization of the strains may determine if this is due to point mutations. Strain selection would allow the severity and duration of infection to be modified providing a useful tool for antiviral profiling in this model.

References:

(1) Xatzipsalti M and Papadopoulos NG (2007) 14:33-41.

(2) Bartlett NW et al. (2008) Nat Med 14: 199-204