Recombinant vesicular stomatitis virus vaccine vectors expressing filovirus glycoproteins lack neurovirulence in nonhuman primates

Abstract

The filoviruses, Marburg virus and Ebola virus, cause severe hemorrhagic fever with high mortality in humans and nonhuman primates. Among the most promising filovirus vaccines under development is a system based on recombinant vesicular stomatitis virus (rVSV) that expresses an individual filovirus glycoprotein (GP) in place of the VSV glycoprotein (G). The main concern with all replication-competent vaccines, including the rVSV filovirus GP vectors, is their safety. To address this concern, we performed a neurovirulence study using 21 cynomolgus macaques where the vaccines were administered intrathalamically. Seven animals received a rVSV vector expressing the Zaire ebolavirus (ZEBOV) GP; seven animals received a rVSV vector expressing the Lake Victoria marburgvirus (MARV) GP; three animals received rVSV-wild type (wt) vector, and four animals received vehicle control. Two of three animals given rVSV-wt showed severe neurological symptoms whereas animals receiving vehicle control, rVSV-ZEBOV-GP, or rVSV-MARV-GP did not develop these symptoms. Histological analysis revealed major lesions in neural tissues of all three rVSV-wt animals; however, no significant lesions were observed in any animals from the filovirus vaccine or vehicle control groups. These data strongly suggest that rVSV filovirus GP vaccine vectors lack the neurovirulence properties associated with the rVSV-wt parent vector and support their further development as a vaccine platform for human use.

Figure 1. Neurovirulence assay design.

(A) Illustration of the rVSV genomes used in the IT inoculation procedures. Note that the only difference between the rVSV vectors used in this study were the glycoproteins. (B) Schematic of the sampling days during the 21 day study. S = swab, B = blood, N = necropsy, * day 5 necropsy of 67-01, ** day 6 necropsy of 56-09.

Figure 2. Representative vehicle control histology with no lesions in the neural tissue.

(A) Frontal cortex (10×). (B) Basal ganglia (10×). (C) Cerebellum (10×). (D) Spinal Cord (Asterisk denotes central canal) (10×).

Figure 3. Representative rVSV-wt histology showing lesions in all neural tissue examined.

(A) Frontal cortex (10×) section with severe encephalitic changes including perivascular lymphohistocytic cuffs (arrows) and aggregates of lymphocytes in the neuroparenchyma (*). (B) Frontal cortex (10×) section with perivascular cuff of lymphocytes and histocytes (arrow). (C) Cerebellum (10×) section with aggregates of lymphocytes in the parenchyma (arrows) admixed with increased numbers of reactive glial cells. (D) Spinal cord (10×) section with gliosis admixed with regions of perivascular inflammation (arrows). (E) Frontal cortex (40×) section depicting large numbers of perivascular lymphocytes and histocytes infiltrating into the adjacent gray matter. (F) Basal ganglia (40×) section depicting large numbers of lymphocytes and histocytes both around a meningeal vessel and invading into the adjacent tissue.

Figure 4. Representative rVSV-ZEBOV-GP histology.

(A and B) Frontal cortex (10×) sections with no lesions. (C) Cerebellum (10×) section with no lesions. (D) Spinal cord (10×) section with no lesions. (E) Frontal cortex (40×) section with a mild perivascular cuff of lymphocytes. (F) Occipital cortex (40×) section with a mild perivascular cuff of lymphocytes.

Figure 5. Combined histological scores of neural tissue.

Graph displaying the mean histological values of the neural tissue from left and right hemisphere for the frontal cortex (FC), basal ganglia (BG), thalamus (TH), occipital cortex (OC), and cerebellum/brainstem (CB/BS), plus scores from the spinal cord (SC). Error bars, standard deviation. VC = vehicle control.

Figure 6. Representative rVSV-MARV-GP histology.

(A) Frontal cortex (10×) section from 59-09 that had a small perivascular cuff of lymphocytes. (B) Frontal cortex (10×) section with no lesions. (C) Cerebellum (10×) section with no lesions. (D) Spinal cord (10×) section with no lesions. (E) Frontal cortex (40×) section with a mild perivascular cuff of lymphocytes. (F) Frontal cortex (40×) section with a scant perivascular cuff of lymphocytes.