Nipah Virus Bangladesh Infection Elicits Organ-Specific Innate and Inflammatory Responses in the Marmoset Model

Abstract

The common marmoset (Callithrix jacchus) is increasingly recognized as an ideal nonhuman primate (NHP) at high biocontainment due to its smaller size and relative ease of handling. Here, we evaluated the susceptibility and pathogenesis of Nipah virus Bangladesh strain (NiVB) infection in marmosets at biosafety level 4. Infection via the intranasal and intratracheal route resulted in fatal disease in all 4 infected marmosets. Three developed pulmonary edema and hemorrhage as well as multifocal hemorrhagic lymphadenopathy, while 1 recapitulated neurologic clinical manifestations and cardiomyopathy on gross pathology. Organ-specific innate and inflammatory responses were characterized by RNA sequencing in 6 different tissues from infected and control marmosets. Notably, a unique transcriptome was revealed in the brainstem of the marmoset exhibiting neurological signs. Our results provide a more comprehensive understanding of NiV pathogenesis in an accessible and novel NHP model, closely reflecting clinical disease as observed in NiV patients.

Keywords: Nipah virus Bangladesh; RNA-seq; cardiomyopathy; common marmoset; inflammatory response; neurological disease; nonhuman primate; pathogenesis; respiratory disease.

Figure 1.

Progressive pathogenesis in marmosets after Nipah virus Bangladesh strain (NiV_B) infection. A, Subject No. 300 was positioned in a ventral dorsal position and all images are in a R → L orientation. X-ray images of the chest were taken on days 0, 1, 3, 5, 7, and at euthanasia. Prior to day 7, lungs appeared normal. Progressive lung pathology was detected from day 7 onward, with a decrease in normal lung tissue over the experimental infection course. By day 7, some decreasing opacity was observed along with an appearance of an interstitial pattern. At the time of euthanasia (day 9 postinfection), virtually no normal lung was observed as the lungs develop a nodular alveolar pattern, suggesting the alveoli were filled with fluid. B, Lung of subject No. 425 at time of euthanasia at day 8 postinfection. Acute diffuse pulmonary hemorrhage, with hemorrhage involving all lobes and <10% of normal tissue remaining. C, Liver of subject No. 425. Diffuse massive necrosis of the liver, with necrosis involving all lobes and normal structure visible.

Figure 2.

Histopathological changes in marmoset tissues after Nipah virus Bangladesh strain (NiV_B) infection: (A and C) hematoxylin and eosin staining; (B and D) immunohistochemical staining for NiV nucleoprotein. A, Lung (subject No. 340): mild to moderate infiltration of mononuclear cells (macrophages, monocytes) and neutrophils in perivascular space. B, Lung (subject No. 425): NiV antigen present in endothelial cells and smooth muscle cells of pulmonary vessels (arteriole, venule). C, Heart (subject No. 425): mild myocarditis with necrosis of myocytes and infiltration of mononuclear cells and neutrophils. D, Heart (subject No. 425): NiV antigen present in capillary and arteriole endothelial cells, as well as heart muscle cells. Scale bar for all images = 50 µm.

Figure 3.

Viral titers and viral quantitative reverse transcription polymerase chain reaction (RT-qPCR) data across various tissues. A, At time of euthanasia, tissues were collected. Viral titers were determined by plaque assay in 23 tissues, combined into 10 tissues systems seen here. Values equal to zero were set to 10⁰ for log-scale visualization. Tissue systems with more than 1 tissue type are lower respiratory tract (trachea, left upper lobe, left lower lobe, right upper lobe, and right lower lobe of the lung as well as each of the left and right bronchi), kidney and adrenals (kidney and adrenal gland), gastrointestinal tract (pancreas, jejunum, and colon transversum), central nervous system (brain-frontal, brain-cerebellum, brainstem, and cervical spinal cord), and genitourinary system (urinary bladder and gonads). B, RT-qPCR was performed to determine relative amounts of viral RNA across 23 tissues. Average cycle threshold for uninfected samples is shown (dotted line). Fill color for box and whisker plots denote similar tissue types (eg, all lymph node [LN] samples are adjacent and grey). Both right (R) and left (L) lower and upper lung samples are adjacent and also grey. Box and whisker plot shows median line, box from 25th to 75th percentiles, and whiskers extending to minimum and maximum values.

Figure 4.

RNA sequencing of Nipah virus Bangladesh strain (NiV_B) in infected marmoset tissues. A, Total number of reads aligned to Callithrix jacchus and NiV_B are shown across all tissues and samples. The dotted line is placed 3 standard deviations above the median number of reads aligning to NiV_B across all naive samples, the threshold for inclusion into (B). B, The fraction of total reads aligning to each NiV_B gene across all samples exceeding 100 NiV_B reads, normalized by transcript length. C, The fraction of reads with the indicated number of inserted guanines (Gs) at the RNA editing site in P for all samples with a depth of at least 100 reads at the RNA editing site. Box and whisker plots shows median line, box from 25th to 75th percentiles, and whiskers extending to minimum and maximum values.

Figure 5.

RNA sequencing samples aligned to Callithrix jacchus. A, Principal component analysis (PCA) was performed on all samples showing that organ system is the primary driver of gene expression differences between samples. Samples from similar organ systems (eg, brainstem and spinal cord) also group together. PC1 is not shown as it is primarily driven by expression differences separating male and female gonad samples from the rest of the dataset. B, A volcano plot comparing all infected samples to all uninfected samples. Lines are placed at 2 (high expression in infected samples) and −2 (higher expression in uninfected samples) as well as false discovery rate (FDR) = 0.05. The 293 genes highlighted in yellow are upregulated in each individual tissue comparison of infected versus uninfected, and in the total sample set are at least 2-fold higher expressed in infected versus uninfected with an FDR step-up value of ≤0.05. C, The 293 yellow-highlighted differentially expressed genes from (B) were used to obtain enriched gene ontology (GO) gene sets, and similar terms were clustered in semantic space using REVIGO [33]. Log₁₀(FDR) of the gene set is indicated by the color and log₁₀ of the number of terms in the gene set is indicated by the size of the circle. D, Differential expression analysis across all samples was used to identify the top 25 genes most overexpressed in infected versus uninfected samples—ranked by FDR step-up value—and hierarchical clustering was performed to obtain a heatmap. *An orthologue or “like” protein via the National Center for Biotechnology Information (NCBI).

Figure 6.

Organ-specific responses seen in RNA sequencing samples aligned to Callithrix jacchus. A, Differential expression analysis across all lung tissue samples was used to identify the top 25 genes most overexpressed in infected versus uninfected lung samples—ranked by FDR step-up value—and hierarchical clustering was performed to obtain a heatmap. B, Enrichment analysis with these 25 genes was performed via ShinyGo 0.76 [34] to identify the top 20 most enriched gene sets. C, Principal component analysis (PCA) was performed on all brainstem samples. The top 2 principal components and their relative contributions are shown. Tissues from infected animals (circles) and tissues from the uninfected animal (squares) are further broken down to highlight the brainstem tissue from the only marmoset showing neurological signs—marmoset No. 340 is an outlier by PCA. D, OPALIN, a marker of oligodendrocytes in gene expression profiling [35], and its 15 nearest neighbors based on single cell type RNA expression data [36, 37] were selected and hierarchical clustering was performed on all brainstem samples. The only marmoset with neurological signs (No. 340, magenta) is a clear outlier and the only marmoset showing high levels of expression of oligodendrocyte-related genes. *An orthologue or “like” protein via the National Center for Biotechnology Information (NCBI).