Pathogenic differences of cynomolgus macaques after Taï Forest virus infection depend on the viral stock propagation

Abstract

Taï Forest virus (TAFV) is a negative-sense RNA virus in the Filoviridae family. TAFV has caused only a single human infection, but several disease outbreaks in chimpanzees have been linked to this virus. Limited research has been done on this human-pathogenic virus. We sought to establish an animal model to assess TAFV disease progression and pathogenicity at our facility. We had access to two different viral stock preparations from different institutions, both originating from the single human case. Type I interferon receptor knockout mice were inoculated with TAFV stock 1 or stock 2 by the intraperitoneal route. Inoculation resulted in 100% survival with no disease regardless of viral stock preparation or infectious dose. Next, cynomolgus macaques were inoculated with TAFV stock 1 or stock 2. Inoculation with TAFV stock 1 resulted in 100% survival and robust TAFV glycoprotein-specific IgG responses including neutralizing antibodies. In contrast, macaques infected with TAFV stock 2 developed disease and were euthanized 8-11 days after infection exhibiting viremia, thrombocytopenia, and increased inflammatory mediators identified by transcriptional analysis. Histopathologic analysis of tissue samples collected at necropsy confirmed classic filovirus disease in numerous organs. Genomic differences in both stock preparations were mapped to several viral genes which may have contributed to disease severity. Taken together, we demonstrate that infection with the two TAFV stocks resulted in no disease in mice and opposing disease phenotypes in cynomolgus macaques, highlighting the impact of viral stock propagation on pathogenicity in animal models.

Fig 1. Intraperitoneal infection of TAFV in mice.

IFNAR^-/- mice were infected i.p. with a low-dose (100 TCID₅₀) of TAFV stock 1, 2, or EBOV or with a high-dose (100,000 TCID₅₀) of TAFV stock 1, 2 or RESTV. At 6 dpi, 4 mice in each group were euthanized for viral load analysis. (A) Mouse body weight changes and (B) survival over time. (C) Viral RNA in blood, liver, or spleen. (D) TAFV GP-specific IgG in serum at study end (28 dpi). Mean and standard error of the mean are displayed in (A). Geometric mean and geometric standard deviation are displayed in (C, D). Statistical significance calculated by Mantel-Cox test or 2-way ANOVA with Tukey’s multiple comparisons is indicated as ***p<0.001 and ****p<0.0001.

Fig 2. Clinical and serological findings in NHPs after TAFV infection.

Cynomolgus macaques (n = 8) were intramuscularly infected with 1x10³ PFU of either TAFV stock 1 or stock 2. (A) Survival curve, (B) clinical score, (C) TAFV RNA in the blood, (D) platelet counts, and (E-F) liver and (G-H) kidney enzyme levels. (I) TAFV GP-specific IgG levels in serum. (J) Serum neutralization presented as 50% reduction of GFP-positive cells (FRNT₅₀) at the time of infection (day 0) and euthanasia (day 28; study end). ALT, alanine aminotransferase; AST, aspartate aminotransferase; BUN, blood urea nitrogen; CREA, creatinine. Statistical significance calculated by Mantel-Cox test is indicated as **p<0.01.

Fig 3. Transcriptional response of NHPs to TAFV infection.

(A) Bar graph of the number of DEG identified at 3, 8, and 11 dpi relative to 0 dpi for TAFV stocks 1 and 2. (B) Venn diagram of DEG for 3 (top), 8 (middle), and 11 (bottom) dpi relative to 0 dpi. (C) Heatmap of average TPM values for stock 1 (top) and stock 2 (bottom) at 3 dpi. (D) Bubbleplot representing GO terms for 8 (left) and 11 (right) dpi relative to 0 dpi. Color indicated -log10(q) and size indicated the number of genes within the GO term. (E) Heatmap of average TPM values for 8 and 11 dpi (left) and unique to 11 dpi (right) from panel D. Color is based on scaled and centered TPM values in (C, E). #, unique to TAFV stock 2 at 8 dpi.

Fig 4. MaSigPro analysis of transcriptional changes between TAFV stocks.

(A) Average TPM of the gene expression clusters identified by MaSigPro using two-way forward regression analysis. (B) Bubbleplot representing GO terms for genes from Cluster 1 and 2 from panel A. Color indicated -log10(q) and size indicated the number of genes within the GO term. (C) Heatmap of average TPM values mapping to GO terms in Panel B for Cluster 1 (left) and Cluster 2 (right). Color is based on scaled and centered TPM values.

Fig 5. Histopathology and immunohistochemistry (IHC) of liver and spleen samples.

Tissue samples were collected at the time of euthanasia and stained with hematoxylin & eosin (H&E) or for TAFV antigen immunoreactivity. (A, B) TAFV stock 1 liver; (C,D) TAFV stock 1 spleen; (E,F) TAFV stock 2 liver; (G,H) TAFV stock 2 spleen. All images are 100x.

Fig 6. Histopathology of tissues from NHPs infected with TAFV stock 2.

Tissue samples were collected at the time of euthanasia (8–11 dpi) and stained with hematoxylin & eosin (H&E) or for TAFV antigen immunoreactivity. (A,D) inguinal lymph node; (B,E) gastro-duodenal junction; (C,F) lung; (G,J) testicle; (H,K) epididymis; (I,L) ovary. All images are 100x.