Infection with MERS-CoV causes lethal pneumonia in the common marmoset

Abstract

The availability of a robust disease model is essential for the development of countermeasures for Middle East respiratory syndrome coronavirus (MERS-CoV). While a rhesus macaque model of MERS-CoV has been established, the lack of uniform, severe disease in this model complicates the analysis of countermeasure studies. Modeling of the interaction between the MERS-CoV spike glycoprotein and its receptor dipeptidyl peptidase 4 predicted comparable interaction energies in common marmosets and humans. The suitability of the marmoset as a MERS-CoV model was tested by inoculation via combined intratracheal, intranasal, oral and ocular routes. Most of the marmosets developed a progressive severe pneumonia leading to euthanasia of some animals. Extensive lesions were evident in the lungs of all animals necropsied at different time points post inoculation. Some animals were also viremic; high viral loads were detected in the lungs of all infected animals, and total RNAseq demonstrated the induction of immune and inflammatory pathways. This is the first description of a severe, partially lethal, disease model of MERS-CoV, and as such will have a major impact on the ability to assess the efficacy of vaccines and treatment strategies as well as allowing more detailed pathogenesis studies.

Figure 1. Interaction between the MERS-CoV spike glycoprotein (S1) and its receptor dipeptidyl peptidase 4 (DPP4).

(A) Alignment of the amino acid residues from human, common marmoset and ferret DPP4 that have been identified to interact with the receptor binding domain of the MERS-CoV spike glycoprotein. (B) Interaction model (front, back and side view) of the MERS-CoV S1 and its cognate receptor human DPP4, with amino acid differences in common marmoset DPP4 are highlighted in red.

Figure 2. Experimental schedule and clinical parameters.

(A) Scheduled necropsies are indicated above and actual necropsies below the timeline. CM1–CM9 indicate the study animals. (B) Animals were scored daily and their mean clinical score ±SD calculated. The number of animals remaining in the experiment is indicated above the bar on each day.

Figure 3. Radiographic alterations and lung pathology.

Dorsal-ventral and lateral thoracic x-rays from of common marmosets (CM) imaged prior to MERS-CoV inoculation (day 0) and on days 3 and 4 post-inoculation. Areas of interstitial infiltration, indicative of pneumonia, are highlighted (circle). Positional indicators are included (R – right, L – left). Gross pathology of the lungs from CM5 and CM9 necropsied on 4 dpi are shown below indicating extensive gross lesions. dpi are shown below indicating extensive gross lesions.

Figure 4. Histopathological changes in lungs of common marmosets inoculated with MERS-CoV.

Common marmosets were euthanized on day 3, 4 or 6 post inoculation and lung tissue was collected and stained with hematoxylin and eosin (H&E; panels A, C, E, G, I) or immunohistochemistry using a polyclonal α-MERS-CoV antibody (IHC; panels B, D, F, H, J). (A) Acute bronchointerstitial pneumonia centered on terminal bronchioles, with influx of inflammatory cells and thickening of alveolar septa in lung tissue collected on 3 dpi. Asterisk indicates essentially normal tissue. (B) IHC staining of sequential section of panel A reveals abundance of MERS-CoV antigen in affected areas. (C) Coalescing bronchointerstitial pneumonia inducing a diffuse lesion on 3 dpi. (D) IHC staining of sequential section of panel C reveals abundance of MERS-CoV antigen in affected areas. (E) Edema, hemorrhage and fibrin (asterisks) fill the alveolar spaces in lung tissue collected on 3 dpi. Arrowhead indicates syncytium. Inset highlights thickened alveolar interstitium with fibrin, edema and inflammatory cells. (F) IHC staining of sequential section of panel E. (G) Type II pneumocyte hyperplasia is visible on 6 dpi, as highlighted further in inset. (H) IHC staining of sequential section of panel G indicates viral antigen has been mostly cleared from remodeling tissue. (I) On 6 dpi, fibrin is consolidating into hyaline membranes (arrows). (J) IHC staining of sequential section of panel I. Magnification: A, B, C and D 4×; E, F, G, H, I and J 40×; inset in panel E and G 100×.

Figure 5. The cellular tropism of MERS-CoV.

(A) DPP4 is evident on type I pneumocytes (arrow), bronchiolar epithelial cells and smooth muscle cells. (B) In situ hybridization demonstrating the presence of viral RNA in type I pneumocytes (black arrow) and alveolar macrophages (blue arrow). (C) Type I pneumocytes (green) are lost following infection with MERS-CoV (lower right). Viral antigen (red) can be observed in type I pneumocytes (arrow). Magnification: A 500×; B 200×; C 500×.

Figure 6. Viral load from marmosets inoculated with MERS-CoV in nasal and throat swabs (A), blood (B) and tissues (C,D).

Nasal and throat swabs were collected at exams while blood samples were only collected at necropsy. RNA was extracted and viral load was determined as TCID₅₀ equivalents by qRT-PCR. The number of animals included in the analysis at each time point in (A) is indicated above the graph. Respiratory (C) and other tissues (D) were collected at necropsy on the indicated days post-inoculation. RNA was extracted and viral load determined as TCID₅₀ equivalents per gram of tissue by qRT-PCR.

Figure 7. Transcriptional signatures of MERS-CoV infection in marmoset lungs.

(A) 2-dimensional hierarchal clustering of DE transcripts from infected lungs. The heatmap shows log2 expression of transcripts with greater than 2-fold change relative to uninfected marmoset lung among all the MERS-CoV-infected samples. The row dendrogram shows clusters of gene expression and the column dendrogram shows clusters of samples with similar transcriptional profiles. (B) Functional network of genes demonstrating the functional relationship between inflammatory mediators and connective tissue adhesions. (C) Molecule activity prediction (MAP) network of cytokines, chemokines, growth factors, and receptors at 4 dpi. Red molecules are those upregulated in the data set, teal molecules are those downregulated in the dataset, orange molecules are those predicted to be activated, and dark blue molecules are those predicted to be inhibited. Orange lines indicate a predicted activation effect, blue lines indicate a predicted inhibitory effect, and gray lines indicate a relationship without a known or predicted activity or inhibition. dpi. Red molecules are those upregulated in the data set, teal molecules are those downregulated in the dataset, orange molecules are those predicted to be activated, and dark blue molecules are those predicted to be inhibited. Orange lines indicate a predicted activation effect, blue lines indicate a predicted inhibitory effect, and gray lines indicate a relationship without a known or predicted activity or inhibition.