Dengue virus-induced hemorrhage in a nonhuman primate model

Abstract

Lack of a dengue hemorrhagic animal model recapitulating human dengue virus infection has been a significant impediment in advancing our understanding of the early events involved in the pathogenesis of dengue disease. In efforts to address this issue, a group of rhesus macaques were intravenously infected with dengue virus serotype 2 (strain 16 681) at 1 x 10(7) PFU/animal. A classic dengue hemorrhage developed 3 to 5 days after infection in 6 of 6 animals. Blood chemistry appeared to be normal with exception of creatine phosphokinase, which peaked at 7 days after infection. A modest thrombocytopenia and noticeable neutropenia concomitant with slight decrease of hemoglobin and hematocrit were registered. In addition, the concentration of D-dimer was elevated significantly. Viremia peaked at 3 to 5 days after infection followed by an inverse relationship between T and B lymphocytes and a bimodal pattern for platelet-monocytes and platelet-neutrophil aggregates. Dengue virus containing platelets engulfed by monocytes was noted at 8 or 9 days after infection. Thus, rhesus macaques inoculated intravenously with a high dose of dengue virus produced dengue hemorrhage, which may provide a unique platform to define the early events in dengue virus infection and help identify which blood components contribute to the pathogenesis of dengue disease.

Figure 1

Dengue hemorrhage in rhesus monkeys. Rhesus monkeys were intravenously infected with dengue virus as described in “Nonhuman primates, virus, and cells.” Hemorrhagic manifestations were captured with digital camera on days 3 (RM2), 4 (RM4), and 5 (RM3) after infection. Different severity of hemorrhage, ranging from petechiae to severe coagulopathy, was seen. A classic clinical hemorrhage was observed in infected animal RM4. The top and bottom panels indicate that the images of the skin hemorrhage were captured from different parts of the body within the same animal.

Figure 2

Viral load in plasma. Blood was drawn at the indicated days, RNA was isolated from plasma, and purified viral RNA was quantified by real-time RT-PCR in specimens collected at alternate day (A) and daily (B) as described in “Quantitation of viral load with real-time RT-PCR.” The peak of dengue viremia in infected RM was from 3 to 7 days after infection. P.I. indicates postinfection.

Figure 3

Typical primary IgM and IgG antibody responses. Presence of dengue specific antibodies in the sera was assayed by ELISA as described in “Methods.” Variations of IgM response in individual RM were observed. But in general, a typical quick and robust response of IgM antibody (A) and a delayed response of IgG antibody (B) were registered. P.I. indicates postinfection.

Figure 4

Profiling of leukocyte subpopulation. Cell surface markers conjugated with proper fluorochrome, which can differentiate the leukocyte subpopulation, were used to stain the fresh-drawn blood and subjected to FACS as described in “FACS analysis and immunohistochemistry.” (A) Absolute counts of each leukocyte subpopulation with SE bar from each animal were presented. A noticeable reduction of monocytes on day 1 after infection was observed, and thereafter a rebounded pattern to normal level was registered. A slight fluctuation with a trend of gradual increase in lymphocytes during acute infection was seen. A consistent and gradual reduction of neutrophils was documented during the acute period, which returned to uninfected level 14 days after infection. (B) Percentage of lymphocyte subpopulation with SE bar from each animal was presented. P.I. indicates postinfection.

Figure 5

Strategy to profile the aggregation of platelets with neutrophils or monocytes. Whole blood flow cytometry was performed after samples were stained with specific cell surface markers conjugated with proper fluorochrome as described in “FACS analysis and immunohistochemistry.” The strategy to gate the specific platelet-leukocyte aggregation was described. The first step is to differentiate neutrophils from monocytes with CD14 surface marker, which then further identified with makers for platelets, respectively.

Figure 6

Profiles of platelet-monocyte or neutrophil-leukocyte aggregation. Kinetics of platelet-leukocyte aggregation were presented as percentage of the gated event. A similar pattern of platelet aggregation with monocytes (A) or neutrophils (B) was observed. P.I. indicates postinfection. Error bars indicate 1 SD.

Figure 7

Engulfment of platelets by monocytes/macrophages. Blood smears were prepared from dengue virus–infected rhesus monkeys and Wright Giemsa and immunofluorescence stainings were performed. Images were acquired on a Zeiss AxioImager A1 epifluorescence microscope with an AxioCam MRC5 camera. Images were captured with a Zeiss 100×/1.3 Plan Neofluar oil objective lens and then processed with AxioVision Release 4.5 software. (A-B) Wright Giemsa staining revealed that tangled platelets were engulfed by monocytes. (C) Immunofluorescence staining revealed that some of these platelets were positive for dengue viral antigen (3H5, red). Nuclear was stained with DAPI (blue) and SYTOX Green.