The purpose of this study was to utilize MRI to assess structural brain changes that occur in a Golden Syrian hamster model infected with Nipah virus via intranasal inoculation. This study was completed in the first and only biosafety level 4 facility designed to provide MR imaging capabilities within a maximum containment facility. The identification of non-invasive imaging biomarkers of NiV-related neurologic disease progression could aid in the examination of potential vaccines and therapeutics.Originally identified in 1998 during an outbreak in Malaysia, Nipah virus (NiV) continues to re-emerge with regular outbreaks in Southeast Asia and India.¹ NiV infection induces both encephalitis and respiratory illness in patients, resulting in a case fatality rate of 40-73%.^1-3 Often one of the most severely affected organs is the brain, with patients presenting with fever, headache, dizziness, reduced levels of consciousness and occasionally seizures. To date, no effective vaccines or treatments have been identified. The examination of animal models of NiV infection is of interest for the investigation of disease pathogenesis, correlation to human disease, and development of medical countermeasures. To better understand the acute neurologic features of NiV, a longitudinal neuroimaging study was performed using a golden Syrian hamster model infected via intranasal infection. Four Golden Syrian hamsters were infected with the NiV (Malaysia strain) via the intranasal route. Two animals received 1000 plaque forming units (PFU) and two received 10,000 PFUs. All animals were imaged on a Philips Achieva 3 Tesla MR clinical scanner using a Philips Solenoid Rat Coil (Philips Healthcare, Cleveland, OH, USA). Animals were imaged before infection and at predefined time points after infection (2, 4, 7 and 9 days post exposure). The animals were anesthetized using isoflurane delivered via a nose cone and positioned on the MR table in a supine fashion to reduce motion artifacts associated with breathing. All images were obtained in the coronal plane. A Magnetization Prepared Rapid Gradient Echo (MPRAGE) sequence was performed with in-plane resolution of 0.14 x 0.14 mm², slice thickness of 0.3 mm, a total of 60 slices ,TR/TE =13/6.5 ms, NSA=4, TFE factor = 66, and FOV of 40 mm x 20 mm x 18 mm (~4 min). A T₂-weighted sequence was acquired with in-plane resolution of 0.08 x 0.08 mm², slice thickness of 0.825 mm, a total of 27 slices,TR/TE of 4500/114 ms, NSA=4, TSE factor=15, and FOV of 40 mm x 20 mm (~7min). A fluid attenuated inversion recovery (FLAIR) sequence was performed with in-plane resolution of 0.14 x 0.14 mm², slice thickness of 1.0 mm, a total of 8 slices, TR/TI/TE of 9280/2100/99 ms, NSA=4, TSE factor =15, and FOV of 40 mm x 20 mm (~8.3 min). Positioning of the FLAIR slices focused on the olfactory bulb and frontal cortex. A rest slab (15 mm) was place across the lower jaw of the animal to reduce foldover artifacts.

Nine days after inoculation with NiV, three of the four animals showed focal hyperintensities in the subcortical regions of the olfactory bulbs on both T₂-weighted and FLAIR images (Figure 1 A-D). One out of the three animals had unilateral involvement and two had bilateral involvement. The fourth animal was found to have slightly enlarged ventricles compared to baseline, possibly reflecting hydrocephalus (Figure 1 E-F), which obscured the evaluation of the olfactory bulb abnormalities. Our imaging findings are concordant with previously published serial sacrifice pathology study⁴ of this NiV animal model which showed intracranial extension of the virus from the olfactory epithelium, along the olfactory nerve fascicles through the cribriform plate, and reaching the olfactory bulb. This led to productive NiV infection in the olfactory bulbs within 8 days of infection, inducing edema, inflammation and perivascular infiltrates.

Additionally, this study enabled us to identify the abilities and limitations of imaging a small animal model in a biosafety level 4 imaging facility using a 3T clinical scanner. Due to the accelerated nature of this disease and the instability of the animal’s condition, the development of quick, high quality imaging sequences (4-8 min.) that can identify neuropathology is of great importance. We found this animal model tolerated repeated imaging in a 45-minute time window even as lung disease progressed. In future imaging endeavors, breath-gating to limit motion artifacts and surface coils that could provide more optimal signal-to-noise should be utilized.

As an imaging model of acute NiV infection, the intranasal infected mouse shows FLAIR and T₂ hyper-intensities in the olfactory bulbs consistent with known histopathology and suggestive of encephalitic involvement.