Establishment of a Nipah Virus Disease Model in Hamsters, including a Comparison of Intranasal and Intraperitoneal Routes of Challenge

Nipah virus (NiV) is an emerging pathogen that can cause severe respiratory illness and encephalitis in humans. The main reservoir is fruit bats, distributed across a large geographical area that includes Australia, Southeast Asia, and Africa. Incursion into humans is widely reported through exposure of infected pigs, ingestion of contaminated food, or through contact with an infected person. With no approved treatments or vaccines, NiV poses a threat to human public health and has epidemic potential. To aid with the assessment of emerging interventions being developed, an expansion of preclinical testing capability is required. Given variations in the model parameters observed in different sites during establishment, optimisation of challenge routes and doses is required. Upon evaluating the hamster model, an intranasal route of challenge was compared with intraperitoneal delivery, demonstrating a more rapid dissemination to wider tissues in the latter. A dose effect was observed between those causing respiratory illness and those resulting in neurological disease. The data demonstrate the successful establishment of the hamster model of NiV disease for subsequent use in the evaluation of vaccines and antivirals.


Introduction
Nipah virus (NiV), from the genus Henipavirus in the family Paramyxoviridae, first emerged in Malaysia in 1998 [1]. Subsequent outbreaks since occurred in Bangladesh or India, beginning in 2001, almost on an annual basis [2]. Several species of fruit bat (Pteropodidae family) are thought to be reservoirs for Henipavirus, including those widely distributed across Australia, Southeast Asia, and Africa [3][4][5], thus maintaining a permanent risk of new outbreaks [6].
There are two main strains of NiV, designated NiV-Malaysia (NiV-M) and NiV-Bangladesh (NiV-B). Due to the different characteristics of human epidemiology, with the outbreak in Malaysia and Singapore having a case fatality rate (CFR) of approximately 40% [1], whereas in Bangladesh this is up to 92% in individual outbreaks [7], it is hypothesised that there may be differences in the pathogenicity of these strains. These strains were therefore compared in hamsters, including via the oronasal route [8], and show similar pathogenic traits, and even some delay of disease kinetic with the Bangladesh strain [9]. Therefore, it is likely that the different CFRs are due to other extrinsic factors. In Malaysia, the main exposure was through close contact with pigs [10], whereas in Bangladesh, exposure was reported through consumption of NiV-contaminated date palm sap [7,11] or direct contact from infected patients [12]. study (Study B) on dose confirmation, further dose reduction and pathogenesis, 16 male and 16 female hamsters were used, aged 6-11 weeks. Groups of n = 12 were challenged with 10 5 TCID 50 and n = 4 with 10 4 TCID 50 via the i.n. route. Groups of n = 12 were challenged with 10 3 TCID 50 and n = 4 with 10 2 TCID 50 via the i.p. route. Food and water were available ad libitum, with environment enrichment included in cages.

Virus
NiV (Malaysian strain; GenBank no. AF212302) was kindly provided by the Special Pathogens Branch of the Centers for Disease Control and Prevention, Atlanta, USA. Virus was propagated and titrated on VeroE6 cells (European Collection of Cell Cultures, Salisbury, UK) grown using Dulbecco's minimal essential medium (DMEM; Gibco, Paisley, UK) supplemented with 2% fetal bovine serum (Gibco, Paisley, UK) at 37 • C. All infectious work was performed in a class III biological safety cabinet line in the containment level (CL) 4 laboratory at UKHSA, Porton Down.

Virus Challenge
Virus was diluted in sterile phosphate buffered saline (PBS) (Gibco, Paisley, UK) to achieve the relevant concentration for the challenge dose.
For i.n. delivery, the virus was instilled in a volume of 100 µL per nostril. For i.p. delivery, the virus was injected in a volume of 200 µL. Challenge was given under isoflurane sedation and animals monitored until a full recovery from sedation was observed.

Clinical Observations
Throughout the study, clinical signs were recorded at least twice a day by experienced husbandry and animal welfare staff. At the same time each day (07:00-09:00) animals were also weighed and had temperatures recorded via an implantable ID/temperature chip (idENTICHIP with Bio-Thermal, MSD Animal Health, Milton Keynes, UK). Clinical signs of disease were assigned a score based on the following criteria: 0, healthy; 1, behavioural change, eyes shut; 2, ruffled fur; 3, wasp-waisted, arched back, dehydrated; 5, laboured breathing; 8, ataxia; and 10, immobility, neurological signs, and paralysis. A cumulative score combining all observed signs was then assigned for each animal at that time point.

Necropsy Procedures
Hamsters were anaesthetised with isoflurane and then given an overdose of sodium pentobarbital at the scheduled end of the study (21 days post-challenge) or upon meeting humane clinical endpoint criteria. A necropsy was performed immediately after confirmation of death. A sample of blood was collected into animal blood RNAprotect tubes (Qiagen, Manchester, UK), and a sample of brain, liver, lung, and spleen into a dry tube. These were stored at −80 • C for viral RNA analysis. The remainder of the brain, liver, lung, and spleen was collected into histology pots containing 10% neutral-buffered formalin for fixation by immersion and further histopathological analysis.

Quantification of Virual Loads by RT-qPCR Preparation
Tissue samples for viral RNA analysis were weighed, resuspended in 1.5 mL PBS, and homogenised through a 400 µm mesh in a Netwell plate (Corning/Fisher Scientific, Loughborough, UK). Then, 200 µL of tissue homogenate or blood was transferred to 600 µL RLT buffer (Qiagen, Manchester, UK), and after at least 10 min, 600 µL 70% isopropanol was added to each sample. Samples were then transferred from the CL-4 laboratory to a CL-3 laboratory where contents were transferred to new tubes for RNA extraction outside of containment. Tissues were further homogenised through a QIAshredder (Qiagen, UK) at 16,000× g for 2 min and RNA was extracted by KingFisher Flex automatic extraction using the BioSprint 96 one-for-all veterinary kit (Indical, Leipzig, Germany) as per the manufacturer's instructions. RNA was eluted in 100 µL AVE buffer (Indical, Leipzig, Germany). Samples were analysed by RT-PCR using the TaqMan Fast Virus  [25]). Quantification of viral load was determined using a 10-fold serial dilution of the NiV N gene in vitro transcript [2.0 × 10 6 to 2.0 × 10 0 copies µL −1 ] (Integrated DNA Technologies, Leuven, Belgium).

Histopathological Analysis
Tissue samples were fixed by immersion in 10% neutral-buffered formalin for at least 3 weeks before being processed routinely into paraffin wax, and 4 µm sections were cut and routinely stained with haematoxylin and eosin (H&E). All slides were digitally scanned using a Hamamatsu S360 digital slide scanner (Hamamatsu Photonics K.K., Shizuoka, Japan) and examined using ndp.view2 software (Hamamatsu Photonics K.K., v2.8.24).
In the lung, the severity of lesions (broncho-interstitial pneumonia) was recorded. The presence of inflammatory infiltrates was recorded in the liver and spleen, together with the presence of lymphoid depletion in the latter. In the brain, the presence of meningitis and perivascular cuffing was also evaluated and scored. For each parameter, a semiquantitative score was given as 0 = within normal limits; 1 = minimal; 2 = mild; 3 = moderate; and 4 = marked/severe. For the spleen, the sum of inflammatory infiltrates and the level of lymphoid depletion was considered the final score, with 8 being the maximum possible score.
In addition, samples were stained using the in situ ybridization RNAscope technique to identify NiV RNA. Briefly, slides were pre-treated with hydrogen peroxide for 10 min (room temperature), target retrieval for 15 min (98-101 • C), and protease plus for 30 min (40 • C) (Advanced Cell Diagnostics, Newark, USA). A NiV-specific probe (Cat No. 439258, Advanced Cell Diagnostics) was incubated with the tissues for 2 h at 40 • C. Amplification of the signal was carried out following the RNAscope protocol using the RNAscope 2.5 HD Detection Kit-Red (Advanced Cell Diagnostics). Likewise, slides were digitally scanned and evaluated with the Nikon NIS-Ar software (Nikon, Praha, Czech Republic) in order to quantify the presence of viral RNA (percentage area positively stained) by digital image analysis.
Histopathology and in situ hybridisation RNAscope technique were carried out in a ISO9001:2015 and GLP-compliant laboratory and evaluation was performed by qualified veterinary pathologists blinded to the study groups.

Statistical Analysis
Statistical analyses were performed using Minitab, version 16.2.2. (Minitab Inc., Chicago, IL, USA). For comparison of survival, nonparametric distribution analysis (right censoring) was undertaken on Kaplan-Meier plots. A non-parametric Mann-Whitney U statistical test was applied to ascertain significance between groups. A significance level ≤ 0.05 was considered statistically significant.

Clinical Parameters
To ascertain the susceptibility of golden Syrian hamsters to NiV infection, dose ranging studies were conducted. Study A used doses of 10 6 and 10 5 for i.n. and 10 5 -10 3 for i.p. challenge. Study B then used the previously used lower dose and a one-log-lower dose. Results from the two studies were combined ( Figure 1). All animals challenged via the i.n. route met humane endpoints, whereas one animal survived until the scheduled end of the study when i.p.-challenged with the lowest dose, 10 2 TCID 50 (Figure 1a,b). The clinical course of disease was accompanied by a loss in weight (Figure 1c,d). A drop in temperature was evident in animals i.n.-challenged, whereas temperatures were more stable in those challenged by the i.p. route (Figure 1e,f). Clinical signs were observed from day 3 post-challenge when delivered i.n., but were observed later in those that were i.p.-challenged (Figure 1g,h). lenged (Figure 1g,h).
Animals met humane clinical endpoints based on the severity of the disease. These signs were divided into those associated with the respiratory system (dyspnoea and wasp waisted) and neurological signs (paralysis, ataxia, and neurological). Animals challenged via the i.n. route mainly exhibited respiratory signs, whereas those challenged via the i.p. route at high disease increasingly exhibited neurological signs ( Figure 2).  Animals met humane clinical endpoints based on the severity of the disease. These signs were divided into those associated with the respiratory system (dyspnoea and wasp waisted) and neurological signs (paralysis, ataxia, and neurological). Animals challenged via the i.n. route mainly exhibited respiratory signs, whereas those challenged via the i.p. route at high disease increasingly exhibited neurological signs ( Figure 2).

Direct Comparison of Intranasal and Intraperitoneal Inoculation Routes
During the dose determination studies, challenge doses of 10 5 and 10 4 TCID50 were tested using both i.n. and i.p. routes. When these routes were compared, the survival graphs were similar, with all animals meeting humane clinical endpoints 4-8 days postchallenge ( Figure 3). There were no discernible statistically significant differences in survival kinetics for either the 10 5 or 10 4 challenge doses (p = 0.910 and p = 0.065, respectively; log-rank survival analysis).

Reproducibility of Challenge Dose in Two Independent Studies
Challenge doses of 10 5 and 10 3 TCID50 were tested in Study A and Study B via the i.n. and i.p. routes, respectively. Survival plots show that these doses were consistent with all animals meeting humane clinical endpoints ( Figure 4). The kinetics of the i.n. route were extremely similar, with no statistically significant effect seen (p = 0.462, Mann-Whitney test), whereas there was a delay in some animals meeting humane endpoints in the second study after i.p. challenge, which reached statistical significance (p = 0.010, Mann-Whitney test); however, the mean time to reaching humane endpoints only differed by 1.5 days (6.2 days for study A and 7.7 days for study B).

Direct Comparison of Intranasal and Intraperitoneal Inoculation Routes
During the dose determination studies, challenge doses of 10 5 and 10 4 TCID 50 were tested using both i.n. and i.p. routes. When these routes were compared, the survival graphs were similar, with all animals meeting humane clinical endpoints 4-8 days post-challenge ( Figure 3). There were no discernible statistically significant differences in survival kinetics for either the 10 5 or 10 4 challenge doses (p = 0.910 and p = 0.065, respectively; log-rank survival analysis).

Direct Comparison of Intranasal and Intraperitoneal Inoculation Routes
During the dose determination studies, challenge doses of 10 5 and 10 4 TCID50 were tested using both i.n. and i.p. routes. When these routes were compared, the survival graphs were similar, with all animals meeting humane clinical endpoints 4-8 days postchallenge ( Figure 3). There were no discernible statistically significant differences in survival kinetics for either the 10 5 or 10 4 challenge doses (p = 0.910 and p = 0.065, respectively; log-rank survival analysis).

Reproducibility of Challenge Dose in Two Independent Studies
Challenge doses of 10 5 and 10 3 TCID50 were tested in Study A and Study B via the i.n. and i.p. routes, respectively. Survival plots show that these doses were consistent with all animals meeting humane clinical endpoints ( Figure 4). The kinetics of the i.n. route were extremely similar, with no statistically significant effect seen (p = 0.462, Mann-Whitney test), whereas there was a delay in some animals meeting humane endpoints in the second study after i.p. challenge, which reached statistical significance (p = 0.010, Mann-Whitney test); however, the mean time to reaching humane endpoints only differed by 1.5 days (6.2 days for study A and 7.7 days for study B).

Reproducibility of Challenge Dose in Two Independent Studies
Challenge doses of 10 5 and 10 3 TCID 50 were tested in Study A and Study B via the i.n. and i.p. routes, respectively. Survival plots show that these doses were consistent with all animals meeting humane clinical endpoints ( Figure 4). The kinetics of the i.n. route were extremely similar, with no statistically significant effect seen (p = 0.462, Mann-Whitney test), whereas there was a delay in some animals meeting humane endpoints in the second study after i.p. challenge, which reached statistical significance (p = 0.010, Mann-Whitney test); however, the mean time to reaching humane endpoints only differed by 1.5 days (6.2 days for study A and 7.7 days for study B).

Viral RNA Levels in Peripheral Blood and Tissues
To ascertain key sites of viral infectivity, a pathogenesis study was conducted where hamsters challenged via the i.n. (10 5 TCID50) or i.p. (10 3 TCID50) routes were culled 2 and 4 days post-inoculation. Presence of viral RNA was assessed in the blood, brain, liver, lung, and spleen. Low levels of viral RNA were detected in the circulation in just 2 animals (Figure 5a). Within the brain, viral RNA was detectable in the majority of animals at day

Viral RNA Levels in Peripheral Blood and Tissues
To ascertain key sites of viral infectivity, a pathogenesis study was conducted where hamsters challenged via the i.n. (10 5 TCID 50 ) or i.p. (10 3 TCID 50 ) routes were culled 2 and 4 days post-inoculation. Presence of viral RNA was assessed in the blood, brain, liver, lung, and spleen. Low levels of viral RNA were detected in the circulation in just 2 animals (Figure 5a). Within the brain, viral RNA was detectable in the majority of animals at day 4 and in all animals that met humane endpoints, irrespective of challenge route (Figure 5b). Viral RNA was observed in the liver at day 2 of all animals challenged via the i.p. route, which is statistically significant compared to the i.n. route (p = 0.0302, Mann-Whitney U test) (Figure 5c). In the lung, viral RNA was detected at all timepoints in all animals, with no statistical differences between challenge routes (Figure 5d). Results from the spleen show viral RNA in the i.p.-challenged animals at day 2, and by day 4-although detectable in the i.n. group-it remained significantly increased in the i.p. cohort (p = 0.0302, Mann-Whitney U test). Moreover, in i.n.-challenged animals, viral RNA was also present in bronchiolar epithelial cells (Figures 7e and 8e,f, insets), while in i.p.-challenged animals, virus was also observed in endothelial cells from blood vessels (Figure 8g,h, insets).

Histopathological Analysis
Histopathological changes were mainly observed in the lung from animals inoculated with i.n. route at day 2 and 4 post-challenge until reaching a humane endpoint (Figure 6a). In these animals, pulmonary lesions consisted of severe multifocal and coalescing areas of broncho-interstitial pneumonia characterised by necrosis of alveolar and bronchiolar epithelium, thickening of the alveolar walls by infiltrating macrophages, and to a lesser extent, lymphocytes and plasma cells (Figure 7a,b and Figure 8a). In most severe cases, neutrophils, cell debris, alveolar macrophages, and mucus plugs were observed within the alveolar, bronchiolar, and bronchial luminae (Figure 8b). In animals inoculated through i.p. route, moderate lesions were observed (Figure 6a), characterised by the presence of multifocal areas of pneumonia, mainly located around blood vessels or airways, infiltrating into the lung parenchyma (Figures 7d and 8c,d). These lesions were mainly composed of macrophages and fewer lymphocytes, plasma cells, and neutrophils (Figures 7d and  8c,d, insets).
Higher presence of viral RNA (RNAScope ISH) was observed in the lung of i.n.challenged animals compared the i.p. route at day 2 and 4 post-challenge (Figure 6e). In i.n.-challenged animals, the viral RNA was located in areas of severe broncho-interstitial pneumonia (Figure 7e,f), whereas in those i.p.-inoculated, the virus was located within the multifocal lesions (Figure 7h). At humane endpoint, a similar viral RNA presence was observed in the lung from different challenged groups, with higher percentages of positively stained areas in those from animals inoculated with 10 2 TCID 50 via i.p (Figure 6e). Moreover, in i.n.-challenged animals, viral RNA was also present in bronchiolar epithelial cells (Figures 7e and 8e,f, insets), while in i.p.-challenged animals, virus was also observed in endothelial cells from blood vessels (Figure 8g,h, insets).      In the liver, no significant lesions were observed at day 2 and 4 post-challenge ( Figure 6b). However, a higher histopathological score was observed in tissues from animals euthanised at humane endpoint, mainly in those challenged with 10 3 and 10 2 TCID 50 through i.p. route, which was correlated with the presence of viral RNA by in situ hybridisation (Figure 6b,f). Viral RNA was located within the liver sinusoids, Kupffer cells, and endothelial cells from hepatic blood vessels (data not shown).
In the brain, more striking lesions were found in i.p. challenge animals at the end of the study, mainly with a 10 2 TCID 50 dose (Figure 6c), the presence of meningitis (infiltration of mononuclear cells within the meninges) being the main observation found. In this group, high expression of viral RNA was observed in the areas of lesions in comparison with the rest of the inoculated groups (Figure 6g and Figure 9). Viral RNA was mainly detected in inflammatory cells within the thickened meninges and endothelial cells from blood vessels (Figure 9b). Additionally, neurons and neuropil within the mid-brain regions showed expression of viral RNA in this group (Figure 9c).
group, high expression of viral RNA was observed in the areas of lesions in comparison with the rest of the inoculated groups (Figures 6g and 9). Viral RNA was mainly detected in inflammatory cells within the thickened meninges and endothelial cells from blood vessels (Figure 9b). Additionally, neurons and neuropil within the mid-brain regions showed expression of viral RNA in this group (Figure 9c).
In the spleen, similar histopathological scores were observed at 2 and 4 days postchallenge and at humane endpoint, with values slightly higher in the animals euthanised at humane endpoint (Figure 6d). Diffuse expression of viral RNA was observed throughout the splenic parenchyma, mainly in the red pulp of these animals (data not shown), and mostly in tissues from i.p. route-inoculated animals at the end of the study ( Figure  6h).

Discussion
In our study, doses resulting in all animals meeting humane clinical endpoints were observed. This conflicts with other reports, where either an LD50 only is reported so it is not possible to know if all animals succumb to severe disease [16], or where even at higher doses some survival is still observed. Using a dose of 10 5 and 10 7 TCID50 via the i.n. route, 62.5 and 33.3% of challenged animals survived, albeit using a recombinant NiV strain [26]. Models where diseases cause all untreated animals to meet humane endpoints are beneficial in the conduction of efficacy studies against severe disease in humans as they exhibit a clinically relevant endpoint of protection [27]. Whilst we ascertained the lowest lethal dose for i.p., we did not with the i.n. route (even whilst using the lowest dose tested of 10 4 TCID50). Of note, where a 10-log dilution series was conducted with an i.p. challenge route, findings align with our results, where survival is seen with doses of 10 2 TCID50 and lower; In the spleen, similar histopathological scores were observed at 2 and 4 days postchallenge and at humane endpoint, with values slightly higher in the animals euthanised at humane endpoint (Figure 6d). Diffuse expression of viral RNA was observed throughout the splenic parenchyma, mainly in the red pulp of these animals (data not shown), and mostly in tissues from i.p. route-inoculated animals at the end of the study (Figure 6h).

Discussion
In our study, doses resulting in all animals meeting humane clinical endpoints were observed. This conflicts with other reports, where either an LD 50 only is reported so it is not possible to know if all animals succumb to severe disease [16], or where even at higher doses some survival is still observed. Using a dose of 10 5 and 10 7 TCID 50 via the i.n. route, 62.5 and 33.3% of challenged animals survived, albeit using a recombinant NiV strain [26]. Models where diseases cause all untreated animals to meet humane endpoints are beneficial in the conduction of efficacy studies against severe disease in humans as they exhibit a clinically relevant endpoint of protection [27]. Whilst we ascertained the lowest lethal dose for i.p., we did not with the i.n. route (even whilst using the lowest dose tested of 10 4 TCID 50 ). Of note, where a 10-log dilution series was conducted with an i.p. challenge route, findings align with our results, where survival is seen with doses of 10 2 TCID 50 and lower; but at higher doses, all animals meet endpoint criteria [9]. Following i.p. challenge, the LD 50 was reported as being 68 TCID 50 [9].
We observed that with lower doses, animals were more likely to exhibit neurological signs in line with other reports [16]. However, other studies found no correlation between dose and subsequent respiratory or neurological disease progression [17] or the occurrence of both neurological signs as well as dyspnoea in the final stages of disease [21]. These results are similar to other viral pathogens, such as the influenza virus, where lower challenge doses resulted in differential disease outcomes [28].
We showed no difference in survival between the two different challenge routes, which is in line with other reports [16]. The i.p. route appears more uniform compared to the i.n. route [21], and is the main reported route of challenge for efficacy testing of NiV vaccine candidates [29][30][31][32][33][34][35][36][37] and antivirals [38]. Whilst we did not assess aerosol exposure, it was reported that this route does not change the course of disease compared to i.n. inoculation at similar doses [17]. With the aerosol route, only the highest dose (10 5 pfu) resulted in all animals meeting endpoint criteria, with doses 2 × 10 4 pfu and lower resulting in animals surviving even after 40 days post-challenge [17]. Viral RNA levels measured by RT-PCR showed that in animals challenged via the i.n. route, the virus primarily targeted the lung before then spreading to the spleen and liver, and finally the brain. Using the i.n. route, others also showed efficient virus replication in the respiratory tract [16]. Given the challenge routes used (i.n. and i.p.), the observation of respiratory virus involvement is not surprising. Given the route of exposure via consumption of date palm sap in the outbreak in Bangladesh [7,11], further work could look into virus infectivity and subsequent spread following ingestion. The finding of the virus and histopathological lesions in the spleen align with findings in fatal cases of NiV infection in humans [22]. Our results complement those of Munster et al., where hamsters were i.n.-challenged with the same dose, 10 5 TCID 50 , and serially culled to assess progression from the nasal cavity to the central nervous system (CNS) [19]. Their results demonstrate entry into the CNS by day 4 post-challenge, in line with our observations of viral RNA presence in the brain at this timepoint. For animals meeting humane endpoints, RT-PCR data show little difference between the i.p. and i.n. route, showing that by this time, organ involvement is similar. However, at day 2 post-challenge, there was a clear distinction with viral RNA only being in the lung in i.n.-challenged animals, whereas dissemination to the liver, spleen, and brain was already evident in the i.p.-challenged group.
Interestingly, viral RNA in the blood was rarely detectable, which is a finding corroborated by others [16]. The dissemination of the virus throughout respiratory tissues was hypothesised to be due to physical spread through respiration or via the mucociliary apparatus [20]. Whilst RT-PCR provides evidence that the virus does not disseminate very efficiently via the hematogenous route, it may be that the level of detection was low and that some virus is present, which accumulates in the spleen due to one of its functions being to remove blood-borne pathogens via filtration [39,40]. Spread into the CNS was shown to be possible via olfactory nerves in both pig [41] and hamster [19] models. Importantly, we tested viral RNA from whole blood samples. This is due to its association with circulating cells, rather than being free in serum, due to binding onto leukocytes [42]. After the i.p. challenge of hamsters, NiV was shown to use this method of leukocyte capture and carry, without being productively infected, to transfer live viruses to other permissive cells [42]. Whilst the study was not designed to comprehensively assess the route of virus entry into the brain, the low levels of virus in the blood suggest it is unlikely to be via crossing the blood-brain barrier.
Histopathology results show lower levels of viral RNA detection compared to RT-PCR assay, especially in the liver. Similarly, despite the presence of viral RNA detected in the brain by RT-PCR at day 4 post-challenge in animals challenged via the i.p. route, changes were not observed with the histology score or RNAscope staining levels. This may be due to different limits of the detection of the assays, with histological and RNAscope assessment being limited to a small area of a tissue section, whereas RT-PCR was performed on a larger homogenised sample section. This observation was also highlighted in the analysis of SARS-CoV-2 viral RNA from challenged hamsters [43]. Another plausible explanation is that the RNAscope probe used here targets only the positive-sense RNA observed during virus replication [20].
Given the pandemic potential of NiV infections [44], particularly the well-documented human-to-human transmission events [12,[45][46][47] and predisposition to cause nosocomial infection [48], the expansion of facilities able to model infection will assist in preparedness and strengthen the ability to respond to future outbreaks and assess the effectiveness of intervention strategies. The changing environment and resultant ecological pressures on the fruit bat reservoir altering foraging and behavioural patterns, owing to deforestation and urban development, may also enable NiV and similar Henipavirus infections to continue to emerge and re-emerge across a wide geographical area [49]. Whilst there are non-human primate models for NiV infection [50], the hamster model has advantages with the ease of procurement and reduced housing requirements, thus enabling studies to be fulfilled in numbers able to reach statistical power more efficiently [27].

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.