Skip to main content
NCBI home page
The Journal of Infectious Diseases logoLink to The Journal of Infectious Diseases
. 2015 Nov 17;213(5):703–711. doi: 10.1093/infdis/jiv538

Humanized Mouse Model of Ebola Virus Disease Mimics the Immune Responses in Human Disease

Brian H Bird 1, Jessica R Spengler 1, Ayan K Chakrabarti 1, Marina L Khristova 2, Tara K Sealy 1, JoAnn D Coleman-McCray 1, Brock E Martin 1, Kimberly A Dodd 1, Cynthia S Goldsmith 3, Jeanine Sanders 3, Sherif R Zaki 3, Stuart T Nichol 1, Christina F Spiropoulou 1
PMCID: PMC4747627  PMID: 26582961

Abstract

Animal models recapitulating human Ebola virus disease (EVD) are critical for insights into virus pathogenesis. Ebola virus (EBOV) isolates derived directly from human specimens do not, without adaptation, cause disease in immunocompetent adult rodents. Here, we describe EVD in mice engrafted with human immune cells (hu-BLT). hu-BLT mice developed EVD following wild-type EBOV infection. Infection with high-dose EBOV resulted in rapid, lethal EVD with high viral loads, alterations in key human antiviral immune cytokines and chemokines, and severe histopathologic findings similar to those shown in the limited human postmortem data available. A dose- and donor-dependent clinical course was observed in hu-BLT mice infected with lower doses of either Mayinga (1976) or Makona (2014) isolates derived from human EBOV cases. Engraftment of the human cellular immune system appeared to be essential for the observed virulence, as nonengrafted mice did not support productive EBOV replication or develop lethal disease. hu-BLT mice offer a unique model for investigating the human immune response in EVD and an alternative animal model for EVD pathogenesis studies and therapeutic screening.

Keywords: Ebola virus disease, animal model, viral hemorrhagic fever, humanized mice, cytokine profile, virus pathogenesis

(See the editorial commentary by Prescott and Feldmann on pages 691–3.)

Ebola virus disease (EVD) is a viral hemorrhagic fever characterized by uncontrolled virus replication, prominent organ system dysfunction, coagulopathy, and high case-fatality ratios [1, 2]. While some viral hemorrhagic fevers are readily modeled in immunocompetent rodents, Ebola virus (EBOV) derived from human patients does not, without extensive serial passaging and adaptation, cause disease in rodents [3–5]. Adapted EBOVs recapitulate various facets of human EVD and have been used to screen antiviral compounds, test vaccines, and investigate adapted-EBOV pathogenesis in the context of a rodent immune system [3, 6–9].

Severe nonadapted EBOV disease can also be modeled in various nonhuman primate species [10], providing valuable insights into EVD pathogenesis and allowing preclinical testing of vaccines and therapeutics. However, nonhuman primates can be difficult to obtain and present practical challenges in high-containment biosafety level 4 (BSL-4) facilities. Additionally, so-called animal-rule guidelines from the Food and Drug Administration specify that multiple distinct animal model systems are preferred, when available, to more robustly recapitulate the pathogenic and immunomodulatory features of the targeted disease to meet regulatory requirements for countermeasure development and approval [11, 12].

A significant advance in improving rodent models was the development of highly immunodeficient NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) mice. These mice lack functional murine macrophages, dendritic cells, T cells, B cells, and natural killer cells, owing to mutations in multiple genes, including those encoding protein kinase and DNA-activated catalytic polypeptide (Prkdc; which results in severe combined immune deficiency), as well as knockout of the gene encoding interleukin 2 receptor γ chain [13, 14]. The mice, when humanized, typically have high levels of engraftment of functional human macrophages, dendritic cells, T cells, B cells, and natural killer cells [13, 15]. hu-BLT mice have allowed the first successful experimental pathogenesis and transmission investigations of various pathogens, including human immunodeficiency virus type 1, Mycobacterium tuberculosis, hepatitis C virus, and others that, like EBOV, specifically target human immune cells for replication and induce aberrant immune states ranging from profound immunosuppression to overly active, dysregulated proinflammatory states capable of host tissue injury [2, 16–21]. This model may allow for the detailed investigation of the specific contributions of the human cellular immune response to pathogenesis in vivo in an ever more sophisticated and refined manner.

In this study, we investigated EVD in hu-BLT mice infected with wild-type EBOV derived from human patients. hu-BLT mice were infected with the lowest passage stocks available of either the historical prototype EBOV (Mayinga 1976) or an isolate from the ongoing outbreak in West Africa (Makona 2014). High-dose challenge with Mayinga EBOV caused uniformly lethal EVD with clinical, immunological, histopathological, and immunohistochemical findings that correspond closely to those in the limited reports of patients with severe EVD. Subsequent studies with lower-dose challenges of either Mayinga or Makona revealed dose- and donor-dependent severity of EVD. Engrafting human immune cell components appeared to be essential for EBOV-mediated virulence, as no significant morbidity or mortality was observed in EBOV-infected nonengrafted controls. Our work is one of the first steps toward assessing EBOV pathogenesis in an in vivo model of the human immune system and may lead to further insights into the immune mechanisms underlying human EVD.

MATERIALS AND METHODS

Biosafety

All work with infectious virus or infected animals was conducted in a BSL-4 laboratory. Strict adherence to intralaboratory infection control practices was maintained to prevent cross-contamination between individual animals and virus groups. All materials and specimens were decontaminated following internally approved, validated, and extensively tested safety protocols prior to removal from the BSL-4 laboratory.

Ethics Statement

All procedures and experiments described herein were approved by the Centers for Disease Control and Prevention (CDC) Institutional Animal Care and Use Committee and were conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals [22]. The CDC is a research facility fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International.

Mice

A total of 26 humanized bone marrow/liver/thymus (hu-BLT) mice from 4 different donors and 3 nonengrafted NSG mice (stock no. 005557) were obtained from Jackson Laboratories (Bar Harbor, Maine). All animals were housed in a climate-controlled laboratory with cycles of 12-hour days and 12-hour nights. All animals were housed in an isolator-caging system (Thoren Caging, Hazleton, Pennsylvania) with a high-efficiency particulate arrestance–filtered (HEPA) inlet and exhaust air supply appropriate for immunodeficient mice. Mice were group housed with food and water ad libitum. After infection, each animal was observed at least once per day by experienced CDC veterinarians or animal health technicians. Animals were humanely euthanized if clinical illness scores (details of scoring methods are available upon request) indicated that the animal was in distress or in the terminal stages of disease.

Viruses and Infection

The virus stocks used in these experiments were the lowest-passage wild-type EBOV isolates available. EBOV Zaire Mayinga (Ebola virus/H.sapiens-tc/COD/1976/Yambuku-Mayinga) and EBOV Zaire Makona KP178538 (Ebola virus/H.sapiens-tc/LBR/2014/Makona) were passaged twice in Vero-E6 cells; all inoculated virus stocks contained the 7-uracil RNA residues at the glycoprotein-editing site [23, 24]. For model characterization, 3 BLT and 3 NSG mice were anesthetized with isoflurane vapors and intraperitoneally inoculated with a target dose of 1 × 105 TCID50 of EBOV diluted in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Grand Island, New York). Negative controls (3 BLT and 3 NSG mice) were inoculated with sterile DMEM. For electron microscopy and variant comparison studies, animals were inoculated intraperitoneally with a target dose of 1 × 103 TCID50 of EBOV Zaire Mayinga diluted in DMEM or 1 × 102 TCID50 of either the Mayinga or Makona EBOV variant diluted in DMEM.

Total RNA Extraction

Whole-blood specimens were collected on days 0, 2, or 7 after infection, from moribund animals at experimental end-points, or at the completion of the experiment. Liver, spleen, lung and kidney samples were collected and approximately 100 mg was added to 1.0 mL of MagMax lysis buffer (Life Technologies, Grand Island, New York) and homogenized. RNA was extracted on a 96-well ABI MagMax extraction platform (details available upon request).

EBOV Quantitative Reverse-Transcription Polymerase Chain Reaction (qRT-PCR) Analysis

EBOV RNA was detected using a qRT-PCR assay (One-Step Superscript III, HiFi Taq, Invitrogen) with primers and probes specific for the nucleoprotein gene of Zaire ebolavirus (available at: http://www.fda.gov/EmergencyPreparedness/Counterterrorism/MedicalCountermeasures/MCMLegalRegulatoryandPolicyFramework/ucm182568.htm#ebola). Viral RNA genome equivalents in blood, fluid, or tissue specimens were quantitated using a standard curve generated by serially diluting a known-titer stock virus. Results were normalized to 18S ribosomal RNA (Life Technologies) following manufacturer's protocols. EBOV RNA abundance in the Makona versus Mayinga study was calculated by comparing the cycle threshold values to an in vitro–transcribed S-segment RNA standard of known copy number.

Tissue-Specific Gene Induction

Liver and spleen tissues were collected at experimental end points of terminal disease or at the end of the experiment. Host gene induction was quantified using multiplex qRT-PCR arrays targeting the human antiviral response (PAHS-122Z, Qiagen, Carlsbad, California), following the manufacturer's protocols.

Cytokine Quantitation

Fresh plasma was collected and stored at −80°C until use in a 41-plex bead-based detection assay (HCYTMAG-60K-PX41, Millipore, Billerica, Massachusetts; Bio-Rad Magpix Hercules, California), following the manufacturers' protocols. The assay was determined by the manufacturer (Millipore) to be species specific, using normal mouse plasma. We further investigated this potential issue by titrating manufacturer-supplied mouse standards, controls, and noninfected mouse plasma, using human-specific multiplex beads. Most analytes had minimal detectable cross-reactivity, but tumor necrosis factor α (TNF-α), interferon γ (IFN-γ), interleukin 6 (IL-6), and macrophage-derived chemokine were found to have variable levels of nonspecific cross-reactivity with reconstituted mouse standards but not normal control mouse plasma.

Histologic Analysis, Immunohistochemical Analysis, and Electron Microscopy

Tissue specimens were fixed in 10% neutral buffered formalin and subjected to gamma irradiation (2 × 106 rad), sectioned into 4-µm-thick slices, and stained with hematoxylin and eosin following routine protocols. Immunohistochemical analysis with viral antigen-specific mouse or rabbit polyclonal antibodies was performed essentially as previously described but with use of a colorimetric polymer-based indirect immunoalkaline phosphatase system [25]. For electron microscopy, liver tissue was minced, fixed in buffered 2.5% glutaraldehyde, subjected to gamma irradiation (as described above), and embedded in a mixture of Epon substitute and Araldite as previously described [26].

Full-Length Genome Sequencing of Stock EBOV and Liver Specimens

Complete genome consensus sequences were generated for the inoculated stock virus and from 3 representative EBOV-infected mouse liver specimens as described by Towner et al [27].

Statistical Analyses

All analyses were completed using PRISM v5.0 (GraphPad, La Jolla, California) or, for gene induction arrays, using the manufacturer's online calculator (available at: http://www.sabiosciences.com/pcrarraydataanalysis.php; accessed 28 September 2015). For the remaining analyses, significant differences were determined using a t test with Holm–Sidak corrections for multiple comparisons.

RESULTS

Wild-Type EBOV Replication and EVD in hu-BLT Mice

EVD development in immunocompetent mice requires virus adaptation [3]. To investigate whether engrafting human hematopoietic stem cells is sufficient for EVD development, 3 hu-BLT mice from a single donor were infected with 1 × 105 TCID50 of EBOV Mayinga and monitored for clinical signs of disease. Engraftment levels ranged 26.5%–54.9% total human CD45+ cells in peripheral blood (Supplementary Figure 1). In the pilot experiment, by day 3 after infection, all BLT mice developed moderate-to-severe disease, ruffled fur, weight loss (−7% to −15% from starting weight), elevated cutaneous temperature, and hunched posture. By day 6 after infection, EBOV-infected hu-BLT mice were euthanized owing to severe clinical illness, weight loss (approximately 30% from starting weight), hypothermia, and moribundity (Figure 1). Rapid virus dissemination and replication were observed in EBOV-infected clinically ill mice, with approximately 2 × 104 TCID50/mL equivalents of EBOV in whole blood 2 days after infection (Figure 1C). At the time of euthanasia (day 6 after infection), viral genome titers in blood and tissue samples often exceeded 1 × 107 TCID50 equivalents. In contrast, no EBOV RNA was detected in liver specimens from EBOV-infected NSG mice at the time of euthanasia (day 16 after infection). To assess the inoculation dose effect on lethality and donor variability effects, subsequent studies reduced the EBOV Mayinga challenge dose to 1 × 103 (n = 6) or 1 × 102 (n = 4) TCID50 and included hu-BLT derived from 3 additional donors. Median survival times for BLT animals challenged with 1 × 103 or 1 × 102 TCID50 were 7.5 and 9 days after infection, respectively (Figure 1), with corresponding delays in weight loss and clinical disease progression.

Figure 1.

Figure 1.

Open in a new tab

Clinical course and Ebola virus (EBOV) RNA levels in hu-BLT mice. A, Survival curve of mock-infected hu-BLT mice, EBOV-infected hu-BLT mice (1 × 102 [n = 4], 1 × 103 [n = 6], or 1 × 105 [n = 3] 50% tissue culture infective doses [TCID50]), and EBOV-infected NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG) control mice (105 TCID50). B, Daily body weight changes in infected animals, by group, as a proportion of the starting weight on the day of infection (day 0). Weights were obtained up to 8 days post-infection for EBOV 102 and 103 groups. Error bars indicate ± 1 standard error of the mean. C, EBOV RNA in serially collected whole-blood specimens from infected hu-BLT mice (red) and mock-infected hu-BLT mice (black); D, EBOV RNA in tissues from mock-inoculated hu-BLT mice (black), EBOV-infected (1 × 105 TCID50) hu-BLT mice (red), and EBOV-infected (1 × 105 TCID50) NSG mice (green; liver only). Results for each animal with sufficient specimens for testing are depicted. All measurements are of EBOV genome RNA in whole blood or tissue homogenates, as determined by quantitative reverse-transcription polymerase chain reaction analysis normalized to 18S ribosomal RNA levels and quantitated on standard curves of stock virus serially diluted 10-fold.

To confirm that disease manifestations were not a result of viral adaptation, viral RNA extracted from hu-BLT mouse tissues was sequenced. Complete 18.9-kb genome consensus sequences were identical between stock virus RNA and viral RNA extracted from liver specimens from hu-BLT mice. None of the mutations previously identified as essential for the enhanced virulence of the mouse-adapted EBOV (MA-EBOV) [28], including key mutations in the nucleoprotein and VP24 genes, were detected (data not shown).

Histopathologic and Clinical Chemistry Findings in hu-BLT Mice

Histopathologic assessment was performed on BLT mice challenged with 1 × 105 TCID50. Results were consistent among all animals. The most severe pathologic finding was observed in the liver, adrenal gland, and perirenal/periadrenal soft tissues, although all tissues examined displayed individual cell necrosis or apoptosis (Figure 2A). The liver showed randomly distributed single-cell necrosis or small aggregates of necrotic hepatocytes. Inflammation associated with necrotic foci was mixed, with neutrophils and macrophages predominating. Intracytoplasmic eosinophilic globular to fibrillar inclusion bodies, typical of human EBOV infections [29], were seen in large numbers of hepatocytes; frequently, every hepatocyte was affected in a high-magnification field. Portal inflammation was prominent, mixed, and often associated with cellular and nuclear debris. Foci within the adrenal cortex showed cells with swollen, vacuolated cytoplasms and overtly necrotic cells. Necrotic foci in the adrenal gland were associated with minimal inflammation; typical inclusion bodies were observed in regions of degeneration and necrosis. Perirenal and periadrenal fibroadipose soft tissues showed necrotic cells, cellular and nuclear debris, and frequent fibrin thrombi. Endothelial cells in these regions were plump and reactive, often protruding conspicuously into the vessel lumen. Electron microscopy of liver tissue confirmed viral replication, as evidenced by production of viral nucleocapsids (Figure 2B). Gross findings of necropsy included pale, friable, enlarged livers in lethally infected animals (Figure 2C). Regardless of infection status small spleens were observed in all hu-BLT mice. No signs of frank hemorrhage were observed. In marked contrast, nonengrafted NSG mice that survived EBOV infection were grossly normal upon necropsy.

Figure 2.

Figure 2.

Open in a new tab

Histopathologic, immunohistochemistry, electron microscopy, and clinical chemistry findings in Ebola virus (EBOV)–infected hu-BLT mice. A, EBOV-infected hu-BLT mouse tissues (400× original magnification), detected by hematoxylin-eosin (H-E) and immunostaining of EBOV antigen. Upper left panel, Mouse liver H-E staining showing characteristic eosinophilic filamentous inclusion bodies (arrows). Lower left panel, Mouse liver EBOV immunostaining of inclusion bodies (arrows) and sinusoidal lining cells (asterisk). Upper middle panel, Adrenal gland H-E staining showing sparse necrosis (arrow) and vacuolar degeneration (asterisk). Lower middle panel, Adrenal gland immunostaining for EBOV antigens in regions associated with necrotic and degenerating adrenal cortical cells (arrow). Upper right panel, Immunostaining of endothelial cells in a peri-renal vessel (arrows). Lower right panel, Mouse liver vessel immunostaining showing prominent EBOV antigen in circulating mononuclear cells (arrow). B, Electron microscopy of mouse liver, showing filamentous viral nucleocapsids cut longitudinally (arrow) and in cross-section (arrowhead). Bar represents 500 nm. L, lipid. C, Gross presentation of hu-BLT mouse liver depicted in panel B. Left panel, Diaphragmatic surface. Right panel, Visceral surface. Liver was collected 8 days after infection with 1 × 103 50% tissue culture infective doses (TCID50) of EBOV. Black bar, 1 cm. D, Whole-blood chemistry findings for EBOV (1 × 105 TCID50)–infected or mock-infected hu-BLT mice. Abbreviations: ALP, alkaline phosphatase; ALT, alanine transferase; AST, aspartate transferase.

Significant increases in alkaline phosphatase, alanine aminotransferase, and aspartate aminotransferase levels were detected in each EBOV-infected animal (Figure 2D). Complete blood counts from BLT mice at the time of euthanasia revealed moderate but nonsignificant leukopenia and thrombocytopenia (data not shown).

Elevated Human Proinflammatory Cytokines-Chemokines and Antiviral Genes in Infected hu-BLT Mice

In patients with EVD, rapid viral replication induces numerous proinflammatory cytokines, including TNF-α, monocyte chemoattractant protein 1 (MCP-1), macrophage inflammatory protein 1β (MIP1-β), IFN-γ–inducible protein 10 (IP-10), interleukin 1 (IL-1), IL-6, interleukin 8 (IL-8), interleukin 10 (IL-10), and IL-1 receptor A (IL-1RA), leading to the so-called cytokine storm [30–32]. Plasma from Mayinga variant–infected and mock-inoculated hu-BLT mice was collected at the time of euthanasia, and levels of 41 human-specific cytokines were analyzed using multiplex assays. Consistent with human disease, significant increases over mock-infected control levels were observed in several proinflammatory mediators, including IFN-α2, TNF-α, IL-1α, IL-15, IL-1RA, MCP-1, IP-10, and granulocyte colony-stimulating factor (Figure 3A). Induction of other clinically significant analytes, including IL-6, IL-8, and IL-10, was also elevated but not significantly so.

Figure 3.

Figure 3.

Open in a new tab

Human cytokine modulation and antiviral gene expression in Ebola virus (EBOV)–infected hu-BLT mice. A, Human cytokines found to be significantly modulated after EBOV infection as determined by multiplex bead-based assays. Shaded bars, EBOV-infected hu-BLT mice; white bars, mock-infected hu-BLT mice. Error bars indicate ± 1 standard error of the mean. *** P < .001, **P < .01, and *P < .05. B, Human antiviral gene arrays on RNA extracted from hu-BLT mouse liver and spleen samples collected 6 days after infection with 1 × 105 50% tissue culture infective doses of EBOV (n = 3). Values denote fold changes in EBOV-infected mice, compared with mock-infected hu-BLT mice. Expression of all genes presented is significant (P < .05), compared with expression for mock-infected animals. Error bars indicate 95% confidence intervals. Abbreviations: G-CSF, granulocyte colony-stimulating factor; GM-CSF, granulocyte macrophage colony-stimulating factor; IFN-α2, interferon α2; IFN-γ, interferon γ; IL-1α, interleukin 1α; IL-1RA, interleukin 1RA; IL-6, interleukin 6; IL-8, interleukin 8; IL-10, interleukin 10; IL-15, interleukin 15; IP-10, interferon γ–inducible protein 10; MCP-1, monocyte chemoattractant protein 1; MCP-3, monocyte chemoattractant protein 3; MDC, macrophage-derived chemokine; MIP-1β, macrophage inflammatory protein 1β; sCD40L, soluble CD40L; VEGF, vascular endothelial growth factor.

RNA extracted from liver and spleen specimens, the main target organs of EBOV, was analyzed by targeted qRT-PCR analysis using a human antiviral gene array. The specificity of the array was confirmed by analysis of RNA from virus-stimulated C57Bl/6J mice; human array gene targets that cross-reacted with murine genes were excluded, except in 3 specific instances noted below. Significant human gene upregulation over levels in mock-inoculated hu-BLT mice was observed in both the liver and the spleen (Figure 3B). In the liver, the most robust upregulation was observed in ISG15 (36.4-fold), IL8 (18.0-fold), CXCL10/IP-10 (17.7-fold), CXCL11/IP-9 (18.8-fold), and CD80 (4.1-fold). ISG15, IP-9, and CD80 results were included here despite some detectable cross-reactivity with murine genes, as they were not found to be upregulated during separate mouse specific antiviral panels run in parallel, suggesting that, for the gene targets, the increases were due to human specific contributions to upregulation of these genes. In the spleen, most prominent upregulation was observed for Cat L/Cathepsin L (35.8-fold), MXA (16.2-fold), and OAS2 (14.5-fold).

Donor Variation in Severity of EVD Caused by EBOV Mayinga and EBOV Makona Variants

Susceptibility to wild-type EBOV infection suggests that hu-BLT mice could be a suitable model for investigating differential pathogenesis of the EBOV Mayinga and EBOV Makona variants. hu-BLT mice were mock inoculated (n = 2) or inoculated with 1 × 102 TCID50 of low-passage isolates of EBOV Mayinga (n = 4) or EBOV Makona (n = 4) variants and followed for 21 days after infection. hu-BLT mice from 2 donors (A and B) were equally distributed in experimental groups. In both of the variant challenge groups, variability between donors was observed. Generally, animals derived from donor A appeared to be more susceptible to severe disease. Among EBOV Mayinga–infected animals, both animals from donor A developed severe disease by days 8–9 after infection; donor B animals appeared either refractory to infection or supported low-level virus replication (Figure 4 and Table 1). Similarly, half of the animals (both from donor A) infected with EBOV Makona died from disease or were euthanized owing to severe illness on days 9 and 11 after infection, and all survivors were derived from donor B. Among those with severe disease, weight loss began around day 5 (Figure 4B). EBOV RNA loads among animals that died from disease were consistent between EBOV Mayinga and EBOV Makona, with highest levels in liver and spleen, as in initial model characterization findings. EBOV RNA was detected in all tissues examined, except for those from one surviving animal that was euthanized 21 days after infection (Table 1).

Figure 4.

Figure 4.

Open in a new tab

Ebola virus (EBOV) Mayinga and EBOV Makona Ebola virus disease in hu-BLT mice derived from 2 donors. A, Survival curve of donor A and donor B derived hu-BLT mice inoculated with EBOV Makona (red square, n = 2; green triangle, n = 2), EBOV Mayinga (yellow diamond, n = 2; blue circle, n = 2), or Dulbecco's modified Eagle's medium (purple triangle, n = 2). B, Associated percentage weight change from the day of inoculation in hu-BLT mice mock inoculated or challenged with EBOV Makona or EBOV Mayinga (1 × 102 50% tissue culture infective doses).

Table 1.

Ebola Virus RNA Copy Number Detected in Blood and Tissue Specimens Collected From Mice at the Time of Euthanasia

Mouse No. Virus/Donor Fatal Time of Death,a d Blood Liver Lung Spleen Kidney Brain Testes Urine
84 Makona/A Y 11 3.56E + 06 4.97E + 07 1.56E + 04 1.32E + 07 1.50E + 06 1.05E + 05 NS NS
86 Makona/A Y 9 3.35E + 06 1.09E + 07 8.86E + 05 1.05E + 07 2.14E + 05 1.47E + 04 NS NS
34 Makona/B N 21 3.11E + 02 1.45E + 05 5.63E + 03 2.38E + 03 2.87E + 04 6.50E + 03 4.08E + 05 NS
31 Makona/B N 21 1.24E + 04 3.60E + 05 6.01E + 04 7.66E + 03 1.05E + 05 2.15E + 04 1.91E + 05 NS
81 Mayinga/A Y 9 NS 3.60E + 08 7.75E + 06 2.17E + 07 4.04E + 06 3.39E + 05 NS NS
80 Mayinga/A Y 8 2.14E + 07 6.35E + 07 1.90E + 06 1.88E + 07 4.60E + 06 6.13E + 05 NS 2.80E + 02
39 Mayinga/B Y 8 ND 1.50E + 01 ND ND ND ND NS NS
38 Mayinga/B N 21 NS ND ND ND ND ND NS NS
Open in a new tab

Abbreviations: ND, not detected; NS, not sampled.

a Data denote time of death from disease or euthanasia after infection.

DISCUSSION

Herein we report that wild-type EBOV infection in hu-BLT mice recapitulates many of the features of severe human EVD. hu-BLT mice rapidly developed high levels of EBOV replication (Figures 1 and 4), profound tissue pathology (Figure 2), and alterations in key human cytokine-chemokine profiles (Figure 3), which are associated with fatal outcomes in patients [30, 31]. Classic target tissues such as the liver, despite being of murine origin, were severely affected, with frequent viral inclusions in hepatocytes indicative of massive viral replication and pathology (Figure 2).

Engrafting human immune response components into hu-BLT mice resulted in enhanced EBOV replication and virulence. In a previous report of a humanized mouse model lacking engraftment of matched liver and thymic tissue, NSG-A2 mice engrafted with hematopoietic stem cells purified from HLA-A2–positive donors challenged with 1 × 103 focus-forming units developed EVD with disease severity associated with engraftment levels; low engraftment was associated with an EVD-related lethality rate of 75% (approximately 12 to >25 days after infection) [33]. In the present study, engraftment levels in hu-BLT mice did not affect clinical course similarly. Our work advances the understanding of immune contributors to disease susceptibility and suggests that providing a more diverse set of human progenitor cells and immune cell education in the context of matched hematopoietic tissue may contribute to disease progression and severity.

Severe cases of human EVD are marked by uncontrolled virus replication and high case-fatality ratios approaching 60%–90%. In patients with EVD, high-level virus replication induces multiple proinflammatory chemokines/cytokines, leading to profound immune activation and dysregulation [30, 31, 34]. Fatal EVD is associated with increased viral load, blood urea nitrogen level, creatinine level, and alanine aminotransferase level. Signs and symptoms observed more frequently in fatal EVD include weakness, dizziness, and diarrhea [35, 36]. Experience from the current outbreak suggests that, while coagulation abnormalities are observed, bleeding appears to be a rare sign [35] and that thrombocytopenia, although common, can be present without evidence of coagulopathy [37]. The present study did not investigate blood coagulation parameters (prothrombin time or activated partial thromboplastin time) in EBOV-infected hu-BLT mice. While no frank hemorrhage was observed in this model, no conclusions regarding the presence or absence of coagulopathies can made without further study. Based on parameters investigated in these studies, infected hu-BLT mice typically developed high levels of EBOV replication, profound tissue pathology, and alterations in key human inflammatory gene activation and elevated proinflammatory cytokine levels, similar to what is reported in severe or fatal human EVD cases. Most notably, upregulation of TNF-α, RANTES, IL-1a, IL-1RA, IL-6, IL-8, and IL-15, all of which are associated with fatal human outcomes, were detected in hu-BLT mice [30, 31, 34]. These findings suggest that future research exploring the immunopathogenesis of EVD may be the most fruitful area where this mouse model may contribute to our understanding of this significant public health threat.

Despite their promise, humanized mice are not without limitations and challenges, including limited availability, expense, complex and as yet poorly understood interspecies interactions between human and murine cells and tissues, and, importantly, risk of graft-versus-host disease (GVHD) [38, 39]. Currently, researchers must closely monitor engrafted BLT cohorts for GVHD, which manifests clinically with severe, sudden-onset hair loss and, ideally, must use these mice within approximately 20 weeks of engraftment. GVHD may be overcome eventually by replacing the murine immunodominant major histocompatibility complex (MHC) protein with the human leukocyte antigen equivalent (HLA) via transgenic techniques to generate HLA-positive, MHC-negative mice for later use as human tissue recipients. Likewise, efforts are underway to enhance B-cell maturation and immunoglobulin M to immunoglobulin G isotype switching efficiency, which will be key for further use in humanized mouse EBOV vaccination studies [22].

The unprecedented emergence of EBOV in a previously unaffected geographic area in West Africa, with >28 000 suspected, probable, and confirmed cases to date [40], emphasizes the continued significant public health threat of viral hemorrhagic fevers and stresses the need for further development of safe and effective vaccines and antiviral therapeutics [41]. hu-BLT mice may provide a useful tool for investigating EBOV pathogenesis, and infection of such mice closely recapitulates many key immunologic facets of human EVD. Human donor variability may be perceived as a drawback or a unique advantage to the hu-BLT EVD model. While donor variability introduces corresponding variability into experimental data (Figure 4 and Table 1), it also reflects variation present in the human population and, in certain applications, may be more appropriate than the 100% lethal nonhuman primate model of EVD. Human donor variation should be considered in future hu-BLT study design and should be capitalized on to model human immune characteristics associated with EVD susceptibility, especially within and across donor groups, to investigate the response of the engrafted immune cell system to that of mock-infected engrafted controls. While the hu-BLT EVD model likely has a variety of applications, as previously discussed, the most valuable use is as an advanced model of humanized immune responses during EVD.

The ability of BLT mice to develop lethal EVD without the need for virus adaptation demonstrates an intriguing contribution of the engrafted human cells toward virulence in an otherwise highly resistant host. Further detailed studies of the immunological mechanisms underlying this observation, especially at early postinfection time points, could reveal how the engrafted human cells circumvent an otherwise highly restrictive species barrier to disease. These future studies may help to elucidate the mechanism of EVD immunopathogenesis in humans and unravel a central mystery of why humans are highly susceptible to severe and often fatal disease due to EBOV and other viral hemorrhagic fevers.

Supplementary Data

Supplementary materials are available at http://jid.oxfordjournals.org. Consisting of data provided by the author to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the author, so questions or comments should be addressed to the author.

Supplementary Data
supp_213_5_703__index.html (875B, html)

Notes

Acknowledgments. We thank Dr Clifton Drew, for outstanding veterinary pathology support and insights; Bobbie Erickson, Tatyana Klimova, Anita McElroy, and César Albariño, for thoughtful suggestions and critical comments during these studies; and Drs Lenny Shultz, Yan Yang, and Jim Keck (Jackson Laboratories) and Drs Michael Brehm and Dale Greiner (University of Massachusetts Medical School), for their helpful technical discussions.

Disclaimer. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention or the US government.

Financial support. This work was supported by the Centers for Disease Control and Prevention (CDC; emerging infectious disease research core funds); the CDC Research Participation Program (to J. R. S., administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the CDC); and the National Institutes of Health Loan Repayment Program (award to J. R. S.).

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

  • 1.Feldmann H, Geisbert TW. Ebola haemorrhagic fever. Lancet 2011; 377:849–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Paessler S, Walker DH. Pathogenesis of the viral hemorrhagic fevers. Annu Rev Pathol 2013; 8:411–40. [DOI] [PubMed] [Google Scholar]
  • 3.Bray M, Davis K, Geisbert T, Schmaljohn C, Huggins J. A mouse model for evaluation of prophylaxis and therapy of Ebola hemorrhagic fever. J Infect Dis 1999; 179(suppl):S248–58. [DOI] [PubMed] [Google Scholar]
  • 4.Warfield KL, Bradfute SB, Wells J et al. Development and characterization of a mouse model for Marburg hemorrhagic fever. J Virol 2009; 83:6404–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Bradfute SB, Warfield KL, Bray M. Mouse models for filovirus infections. Viruses 2012; 4:1477–508. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Connolly BM, Steele KE, Davis KJ et al. Pathogenesis of experimental Ebola virus infection in guinea pigs. J Infect Dis 1999; 179(suppl):S203–17. [DOI] [PubMed] [Google Scholar]
  • 7.Wahl-Jensen V, Bollinger L, Safronetz D, de Kok-Mercado F, Scott DP, Ebihara H. Use of the Syrian hamster as a new model of Ebola virus disease and other viral hemorrhagic fevers. Viruses 2012; 4:3754–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bowen ETW, Platt GS, Lloyd G, Raymond RT, Simpson DIH. A comparative study of strains of Ebola virus isolated from southern Sudan and northern Zaire in 1976. J Med Virol 1980; 6:129–38. [DOI] [PubMed] [Google Scholar]
  • 9.Volchkov VE, Chepurnov AA, Volchkova VA, Ternovoj VA, Klenk HD. Molecular characterization of guinea pig-adapted variants of Ebola virus. Virology 2000; 277:147–55. [DOI] [PubMed] [Google Scholar]
  • 10.Nakayama E, Saijo M. Animal models for Ebola and Marburg virus infections. Front Microbiol 2013; 4(September):267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Snoy PJ. Establishing efficacy of human products using animals: the US food and drug administration's “animal rule”. Vet Pathol 2010; 47:774–8. [DOI] [PubMed] [Google Scholar]
  • 12.Sullivan NJ, Martin JE, Graham BS, Nabel GJ. Correlates of protective immunity for Ebola vaccines: implications for regulatory approval by the animal rule. Nat Rev Microbiol 2009; 7:393–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ishikawa F, Yasukawa M, Lyons B et al. Development of functional human blood and immune systems in NOD/SCID/IL2 receptor γ chain null mice. System 2005; 106:1565–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Shultz LD, Lyons BL, Burzenski LM et al. Human lymphoid and myeloid cell development in NOD/LtSz-scid IL2R gamma null mice engrafted with mobilized human hemopoietic stem cells. J Immunol 2005; 174:6477–89. [DOI] [PubMed] [Google Scholar]
  • 15.Shultz LD, Brehm MA, Garcia-Martinez JV, Greiner DL. Humanized mice for immune system investigation: progress, promise and challenges. Nat Rev Immunol 2012; 12:786–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Peters CJ, Zaki SR. Overview of viral hemorrhagic fevers. In: Guerrant RL, Walker DH, Weller PF, eds. Tropical infectious diseases: principles, pathogens & practice. 3rd ed. Philadelphia: Saunders Elsevier, 2011:441–8. [Google Scholar]
  • 17.Calderon VE, Valbuena G, Goez Y et al. A humanized mouse model of tuberculosis. PLoS One 2013; 8:e63331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.McEnery R. Will humanized mice move us closer to an AIDS vaccine? IAVI Rep Newsl Int AIDS vaccine Res 2012; 16:15–6. [PubMed] [Google Scholar]
  • 19.Nischang M, Sutmuller R, Gers-Huber G et al. Humanized mice recapitulate key features of HIV-1 infection: A novel concept using long-acting anti-retroviral drugs for treating HIV-1. PLoS One 2012; 7:e38853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Dorner M, Ploss A. Deconstructing hepatitis C virus infection in humanized mice. Ann N Y Acad Sci 2011; 1245:59–62. [DOI] [PubMed] [Google Scholar]
  • 21.Mota J, Rico-Hesse R. Humanized mice show clinical signs of dengue fever according to infecting virus genotype. J Virol 2009; 83:8638–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.National Research Council of the National Academies. Guide for the care and use of laboratory animals. 8th ed Washington, DC: National Academy Press, 2011. [Google Scholar]
  • 23.Volchkov VE, Becker S, Volchkova VA et al. GP mRNA of Ebola virus is edited by the Ebola virus polymerase and by T7 and vaccinia virus polymerases. Virology 1995; 214:421–30. [DOI] [PubMed] [Google Scholar]
  • 24.Sanchez A, Yang ZY, Xu L, Nabel GJ, Crews T, Peters CJ. Biochemical analysis of the secreted and virion glycoproteins of Ebola virus. J Virol 1998; 72:6442–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Zaki SR, Shieh WJ, Greer PW et al. A novel immunohistochemical assay for the detection of Ebola virus in skin: implications for diagnosis, spread, and surveillance of Ebola hemorrhagic fever. Commission de Lutte contre les Epidémies à Kikwit. J Infect Dis 1999; 179(suppl):S36–47. [DOI] [PubMed] [Google Scholar]
  • 26.Goldsmith CS, Whistler T, Rollin PE et al. Elucidation of Nipah virus morphogenesis and replication using ultrastructural and molecular approaches. Virus Res 2003; 92:89–98. [DOI] [PubMed] [Google Scholar]
  • 27.Towner JS, Sealy TK, Ksiazek TG, Nichol ST. High-throughput molecular detection of hemorrhagic fever virus threats with applications for outbreak settings. J Infect Dis 2007; 196:S205–12. [DOI] [PubMed] [Google Scholar]
  • 28.Ebihara H, Takada A, Kobasa D et al. Molecular determinants of Ebola virus virulence in mice. PLoS Pathog 2006; 2:e73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Martines RB, Ng DL, Greer PW, Rollin PE, Zaki SR. Tissue and cellular tropism, pathology and pathogenesis of Ebola and Marburg viruses. J Pathol 2015; 253:153–74. [DOI] [PubMed] [Google Scholar]
  • 30.Wauquier N, Becquart P, Padilla C, Baize S, Leroy EM. Human fatal Zaire Ebola virus infection is associated with an aberrant innate immunity and with massive lymphocyte apoptosis. PLoS Negl Trop Dis 2010; 4:pii:e837. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Villinger F, Rollin PE, Brar SS et al. Markedly elevated levels of interferon (IFN)-gamma, IFN-alpha, interleukin (IL)-2, IL-10, and tumor necrosis factor-alpha associated with fatal Ebola virus infection. J Infect Dis 1999; 179(suppl Il):S188–91. [DOI] [PubMed] [Google Scholar]
  • 32.McElroy AK, Erickson BR, Flietstra TD et al. Ebola hemorrhagic Fever: novel biomarker correlates of clinical outcome. J Infect Dis 2014; 210:558–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Lüdtke A, Oestereich L, Ruibal P et al. Ebola virus disease in mice transplanted with human hematopoietic stem cells. J Virol 2015; 89:4700–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.McElroy AK, Akondy RS, Davis CW et al. Human Ebola virus infection results in substantial immune activation. Proc Natl Acad Sci U S A 2015; 112:4719–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Schieffelin JS, Shaffer JG, Goba A et al. Clinical illness and outcomes in patients with Ebola in Sierra Leone. N Engl J Med 2014; 371:2092–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Qin E, Bi J, Zhao M et al. Clinical features of patients with Ebola virus disease in Sierra Leone. Clin Infect Dis 2015; 61:491–5. [DOI] [PubMed] [Google Scholar]
  • 37.Lyon GM, Mehta AK, Varkey JB et al. Clinical care of two patients with Ebola virus disease in the United States. N Engl J Med 2014; 371:2402–9. [DOI] [PubMed] [Google Scholar]
  • 38.Covassin L, Jangalwe S, Jouvet N et al. Human immune system development and survival of non-obese diabetic (NOD)-scid IL2rγnull (NSG) mice engrafted with human thymus and autologous haematopoietic stem cells. Clin Exp Immunol 2013; 174:372–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Greenblatt MB, Vbranac V, Tivey T, Tsang K, Tager AM, Aliprantis AO. Graft versus host disease in the bone marrow, liver and thymus humanized mouse model. PLoS One 2012; 7:e44664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Centers for Disease Control and Prevention. 2014 Ebola outbreak in West Africa—case counts, 2015. http://www.cdc.gov/vhf/ebola/outbreaks/2014-west-africa/case-counts.html Accessed 6 April 2015.
  • 41.Baize S, Pannetier D, Oestereich L et al. Emergence of Zaire Ebola virus disease in Guinea—preliminary report. N Engl J Med 2014; 371:1418–25. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_213_5_703__index.html (875B, html)
supp_jiv538_jiv538supp_fig1.docx (186.4KB, docx)

Articles from The Journal of Infectious Diseases are provided here courtesy of Oxford University Press

RESOURCES