Pathogenic Differences between Nipah Virus Bangladesh and Malaysia Strains in Primates: Implications for Antibody Therapy

Abstract

Nipah virus (NiV) is a paramyxovirus that causes severe disease in humans and animals. There are two distinct strains of NiV, Malaysia (NiVM) and Bangladesh (NiVB). Differences in transmission patterns and mortality rates suggest that NiVB may be more pathogenic than NiVM. To investigate pathogenic differences between strains, 4 African green monkeys (AGM) were exposed to NiVM and 4 AGMs were exposed to NiVB. While NiVB was uniformly lethal, only 50% of NiVM-infected animals succumbed to infection. Histopathology of lungs and spleens from NiVB-infected AGMs was significantly more severe than NiVM-infected animals. Importantly, a second study utilizing 11 AGMs showed that the therapeutic window for human monoclonal antibody m102.4, previously shown to rescue AGMs from NiVM infection, was much shorter in NiVB-infected AGMs. Together, these data show that NiVB is more pathogenic in AGMs under identical experimental conditions and suggests that postexposure treatments may need to be NiV strain specific for optimal efficacy.

Figure 1. NiV_M and NiV_B comparison in human endothelial cells and AGMs.

Growth kinetics in endothelial cell lines at MOI 5 (a) and primary endothelial cell cultures at MOI 1 (b). Lighter colors represent brain endothelial cell experiments and darker colors represent lung endothelial cell experiments; Red hues: NiV_M; Blue hues: NiV_B. Error bars represent standard deviation of the mean. ANOVA with Dunnett’s Multiple Comparison Test; n = 4. **p-value < 0.01; ***p-value < 0.001. Limit of detection 25 PFU/ml. (c) Kaplan-Meier survival curves for AGMs infected with NiV_M (red) and NiV_B (blue). Log-rank (Mantel-Cox) Test; n = 4 for both cohorts. *p-value < 0.05.

Figure 2. Viral load in swabs and blood.

NiV_M (red hues) and NiV_B (blue hues) viral RNA genomic equivalents detected by qRT-PCR from sampling at days 0, 1, 3, 5, 7, 10, and 15 in nasal swabs (a), oral swabs (b), rectal swabs (c), or circulating in blood (d). Plus sign denotes infectious virus isolation from sample day. Error bars represent standard deviation. ANOVA with Dunnett’s Multiple Comparison Test was performed on the mean values of all animals from each group; n = 4. n.s., not significant; **p-value < 0.01; ***p-value < 0.001.

Figure 3. Viral load in tissues.

NiV_M (red hues) and NiV_B (blue hues) viral RNA genomic equivalents detected by qRT-PCR from tissues at study endpoint for each AGM in pulmonary associated tissues (a), neural tissues (b), lymphoid tissues (c), or other typical NiV-infected tissues (d). Plus sign denotes infectious virus isolation from sample; underlined plus sign denotes virus isolated from every AGM in group for that particular sample. R (right), L (left), RU (right upper), RM (right middle), RL (right lower), LU (left upper), LM (left middle), LL (left lower), LN (lymph node). Error bars represent standard deviation.

Figure 4. H&E and immunohistochemistry of AGM lung, spleen, and kidney.

Representative H&E of lung (b,d), spleen (f,h), and kidney (j,l) and representative immunohistochemistry for NiV antigen of lung (a,c), spleen (e,g), and kidney (i,k). Cells immunopositive for NiV antigen appear brown. NiV_M in left panels and NiV_B in right panels. (a) inset a magnification of endothelial cells immunopositive for NiV antigen in lung. (d) inset a magnification of area marked by white asterisk showing an endothelial cell syncytium in lung. (h) inset a magnification of area marked by white asterisk showing large syncytium in the spleen; black asterisks mark other areas of syncytia in the spleen. Representative normal splenic germinal center architecture in (f) whereas the white pulp is disrupted in (h). (l) inset a magnification of area marked by white asterisk showing a syncytium of endothelial cells in a glomerular tuft. Lung and spleen tissue section images taken at 20X magnification and kidney tissue section images taken at 40X magnification.

Figure 5. Histopathology scores for lung and spleen.

(a) Lung scores based on two histopathology categories for all lung lobes and all indicated animals 1) perivascular edema and vasculitis and 2) pneumonia-interstitial, subacute with fibrin and edema. (b) Spleen scores based on histopathology associated with lymphoid necrosis, lymphoid necrosis, hemorrhage, and fibrin. Scoring index based on percent affected tissue: 0- no lesion (none); 1- minimal change (10% and less); 2- mild change (11–25%); 3- moderate change (26–50%); 4- marked change (51–75%); 5- severe change (76–100%). Malaysia-Full represents n = 4 NiV_M cohort. Malaysia-Succumb represents n = 2 of NiV_M cohort that succumbed to NiV_M. Bars show distribution of scoring and median scores. A Mann-Whitney test comparing the sum rank of each group was used: (a) *p-value = 0.0111, ***p-value < 0.0001; (b) n.s. not significant, **p-value = 0.0041, ***p-value = 0.0004.

Figure 6. Human mAb m102.4 activity against NiV_B in vitro and in vivo.

(a) Percent neutralizing antibody dose response of m102.4 against NiV_M (red) and NiV_B (blue) infection in Vero cells. Error bars show standard deviation. (b) Diagram of the m102.4 treatment regimen and sampling days post NiV_B challenge in AGMs. Black arrows; D1/D3 treatment. Green arrows; D3/D5 treatment. Blue arrows; D5/D7 treatment. * depicts the day of NiV_B challenge. (c) Kaplan-Meier survival curves for the D1/D3 group (gray, n = 3), D3/D5 group (green, n = 3), the D5/D7 group (blue, n = 3), and the control group (red, n = 2). Log-rank (Mantel-Cox) Test; *p-value < 0.05 comparing the NiV_B D5/D7 cohort with the previously reported data from the NiV_M D5/D7 cohort.

Figure 7. Immunohistochemistry of AGM tissue after NiV_B challenge and m102.4 treatment.

Lack of NiV antigen in representative m102.4 D1/D3 and D3/D5 treated tissues and localization of NiV antigen in representative control and D5/D7 treated tissues by immunohistochemical staining. Lung and spleen were labeled with an N protein-specific polyclonal rabbit antibody and images taken at 20X magnification.

Figure 8. Seroconversion of AGMs post-challenge and m102.4 treatment.

Detection of NiV F specific antibodies from m102.4 treated and non-treated AGMs. D1/D3 group (gray, n = 3), D3/D5 group (green, n = 3), the D5/D7 group (blue, n = 3), and the control group (red, n = 2). Mean fluorescence intensities (MFI) are shown on the y-axis and represent binding of specific Ig (IgG, and IgM) to NiV F. Error bars represent the standard deviation of fluorescence intensity across 100 beads for each sample.