Pathology of Lagovirus europaeus GI.2/RHDV2/b (Rabbit Hemorrhagic Disease Virus 2) in Native North American Lagomorphs

Rabbit hemorrhagic disease virus (Lagovirus europaeus GI.1/RHDVa) is a Lagovirus in the family Caliciviridae, with a host range primarily limited to the European rabbit (Oryctolagus cuniculus; Abrantes et al. 2012). Following detection in China in 1984, the virus quickly spread through Asia, Europe, and Australia, causing heavy losses of commercial and wild O. cuniculus (Abrantes et al. 2012). A second pathogenic Lagovirus, rabbit hemorrhagic disease virus 2 (Lagovirus europaeus GI.2/RHDV2/b), emerged in France in 2010 (Le Gall-Reculé et al. 2013). This strain also exhibits high virulence in O. cuniculus and has spread rapidly throughout the world, leading to additional economic and ecologic disruptions due to its effect on domestic and wild rabbit populations (Rouco et al. 2019). Compared to GI.1/RHDVa, GI.2/RHDV2/b shows: a broader host range, including multiple hare species (e.g., brown hare [Lepus europaeus; Velarde et al. 2017] and mountain hare [Lepus timidus; Neimanis et al. 2018a]); a lower but highly variable mortality rate; the potential for longer disease duration; and the ability to cause lethal infection in young animals as well as adults (Velarde et al. 2017).

The first detection of GI.2/RHDV2/b in North America occurred in 2016 in domestic O. cuniculus, with limited, sporadic detections in subsequent years (Rouco et al. 2019). In April 2020, GI.2/RHDV2/b was reported in native North American lagomorphs associated with mortality events involving wild desert cottontail (Sylvilagus audubonii) and black-tailed jackrabbit (Lepus californicus) in New Mexico, US (World Organization for Animal Health 2020). In the following months, mortalities of multiple wild rabbit and hare species, including desert cottontail, mountain cottontail (Sylvilagus nuttallii), black-tailed jackrabbit, and antelope jackrabbit (Lepus alleni), were reported in Arizona, California, Colorado, Nevada, Texas, and Utah, in association with GI.2/RHDV2/b detection (US Geological Survey National Wildlife Health Center 2020). As rabbit hemorrhagic disease was unprecedented in native North American wildlife, little was known regarding either the pathology or virulence of GI.2/RHDV2/b in these species. Here, we report postmortem findings in desert cottontails and black-tailed jackrabbits naturally infected with GI.2/RHDV2/b and compare them to findings in Old World lagomorph species.

Seven desert cottontails and seven black-tailed jackrabbits from Texas (n=9), New Mexico (n=4), and Arizona (n=1), collected in April and May 2020, were received for postmortem examination at the US Geological Survey National Wildlife Health Center. Animals were necropsied, and liver, lung, kidney, spleen, brain, trachea, heart, eye, esophagus, stomach, duodenum, jejunum, cecum, and colon were collected in 10% neutral buffered formalin, processed routinely, sectioned at approximately 5 µm, and stained with H&E for light microscopic assessment (not all tissues were examined from all animals, depending on the degree of postmortem artifact). A subset of liver, kidney, and lung sections was stained with phosphotungstic acid–hematoxylin to highlight fibrin, while another subset of liver sections was stained with von Kossa to highlight calcification. Liver was tested for GI.1/RHDVa and GI.2/RHDV2/b by real-time-PCR at the US Department of Agriculture's Foreign Animal Disease Diagnostic Laboratory, Plum Island, New York (FADDL) (adapted from Duarte et al. 2015; US Department of Agriculture 2018). Age assessment was based on size, reproductive tract development, and presence of thymus. Tissues were assessed for the presence or absence of gross and microscopic lesions commonly described with GI.2/RHDV2/b infection, and any additional abnormal findings were noted. Due to postmortem or euthanasia artifacts, we were not able to assess for all microscopic lesions previously reported with GI.2/RHDV2/b infection in all of our cases. In cases where artifact interfered with reliable assessment for a particular lesion, the animal was excluded from the total number assessed for that lesion.

Twelve animals were found dead, and two were found lethargic and euthanized. Five animals were submitted chilled, and nine were frozen prior to necropsy. Animals were in good to fair postmortem condition based on gross assessment. Body weights of desert cottontails ranged from 73 to 1,230 g, and weights of black-tailed jackrabbits ranged from 440 to 3,260 g. Adults were more common than juveniles (9/14 adult; 64%), and females were overrepresented (10/14 female; 71%). Of the six adult females, four were pregnant, and one was lactating. Most animals (10/14; 71%) were in poor body condition with depleted fat stores. Consistent gross findings included epistaxis (12/13; 92%; Fig. 1A); a tan friable liver with an enhanced reticular pattern or pinpoint red capsular foci (13/14; 93%; Fig. 1B); and wet and heavy lungs that were mottled pink to red or red to dark red (14/14; 100%; Fig. 1C); lungs of most (13/14; 93%) animals floated in formalin. Other common gross findings included pale tan kidneys (6/14; 43%) and meningeal congestion (5/13; 38%). Less common findings included petechiae on cecocolic serosa, epididymis, or ovary (3/14; 21%; Fig. 1D); perianal or periurethral blood (2/14; 14%); splenomegaly (2/14; 14%); and dark red tracheal mucosa (1/14; 7%). In 4/13 (31%) animals where eyes could be assessed, buphthalmia, hyphema, or hemorrhagic ocular discharge was suspected, although due to autolysis or trimming artifacts, ocular abnormalities could not be confirmed histologically.

Microscopic findings (Table 1 and Fig. 2) included dissociation (14/14; 100%) and necrosis or apoptosis (11/11; 100%) of hepatocytes; in all livers suitable for assessment of necrosis, panlobular (massive) dissociation and necrosis or apoptosis were present in at least some areas (Fig. 2C), while 3/11 (27%) livers also exhibited periportal to midzonal dissociation and necrosis or apoptosis (Fig. 2A, B). Dissociation was characterized by individualization and rounding of hepatocytes with loss of chord architecture (Fig. 2C, D). Areas of necrosis or apoptosis lacked significant heterophilic inflammation, although heterophils were present in low numbers in some areas (Fig. 2D). Less frequent hepatic findings included hemorrhage (6/9; 67%; Fig. 2B), lipid-type hepatocellular vacuolation (7/12; 58%), and hepatocellular mineralization (4/13; 31%). Evaluation of the lung revealed congestion (12/12; 100%), edema with increased numbers of intra-alveolar macrophages (12/13; 92%), intra-alveolar hemorrhage (11/12; 92%; Fig. 2E), interstitial heterophils or mononuclear cells (2/9; 22%), and pyknotic interstitial debris (1/9; 11%). In the kidney, there was frequent accumulation of hemoglobin, red blood cells, or protein in renal tubules (9/14; 64%) and acute tubular injury (3/8; 38%; Fig. 2F). Splenic lesions were limited to congestion (2/5; 40%). Additional findings included submucosal tracheal congestion (10/13; 77%) and luminal hemorrhage (2/13; 15%); epicardial or myocardial hemorrhage (3/14; 21%); intra-alveolar fibrin accumulation (1/8; 13%); and cerebral and cerebellar hemorrhage (1/14; 7%) and meningeal congestion (1/14; 7%).

Concurrent findings not typically associated with GI.2/RHDV2/b infection and considered most probably incidental included intestinal luminal metazoan parasites or coccidia (7/14; 50%); mesenteric cysticerci (5/14; 36%); mild lymphoplasmacytic renal interstitial infiltrates (3/14; 21%); 1–2-mm-diameter white foci in the liver corresponding with granulomas seen histologically, probably secondary to Taenia sp. migration (2/13; 15%); subcutaneous edema (2/14; 14%); intercostal muscular hemorrhage (1/14; 7%); and eosinophilic intranuclear inclusions in the renal tubular epithelia resembling herpesviral inclusions (1/14; 7%), although a PCR for herpesvirus was negative (Ehlers et al. 1999).

All animals tested positive for GI.2/RHDV2/b and negative for GI.1/RHDVa by real-time PCR at FADDL, and death in all cases was attributed to GI.2/RHDV2/b. Subjectively, gross lesions were more severe in the jackrabbits, and several microscopic lesions were more common in jackrabbits than in cottontails (Table 1). These included hepatocellular mineralization (57% vs. 0%); acute tubular injury (67% vs. 20%); renal tubular hemorrhage, hemoglobin, or protein (86% vs. 43%); tracheal congestion (100% vs. 50%); and tracheal luminal hemorrhage (29% vs. 0%). Hepatocellular mineralization has been previously described associated with lagoviral infection in Lepus spp., but not in O. cuniculi (Neimanis et al. 2018a). Comparing microscopic lesion prevalence in adults vs. juveniles, renal lesions were more common in adults, including renal tubular hemorrhage, hemoglobin, or protein accumulation (8/9; 89% vs. 1/5; 20%) and acute tubular injury (3/4; 75% vs. 0/4; 0%). We could not definitively determine the age of our animals, but the inclusion of juvenile desert cottontails weighing as low as 73 g and black-tailed jackrabbits weighing as low as 440 g confirms that young animals of both species are susceptible to fatal infection, consistent with reports of fatal GI.2/RHDV2/b infection in O. cuniculus as young as 11 d old (Velarde et al. 2017).

Our pathologic findings associated with GI.2/RHDV2/b infection in cottontails and jackrabbits are consistent with those reported in European lagomorphs (Velarde et al. 2017; Neimanis et al. 2018a, 2018b; Harcourt-Brown et al. 2020), with few exceptions. Other lesions reported with GI.2/RHDV2/b infection in O. cuniculus that were not found in our cases include glomerular or pulmonary fibrin thrombi (Neimanis et al. 2018b; Harcourt-Brown et al. 2020); significant heterophilic inflammation in the liver associated with areas of necrosis (Neimanis et al. 2018b); and various splenic lesions including lymphoid depletion, lymphocytolysis, macrophage hyperplasia, red pulp necrosis, and red pulp fibrin accumulation (Neimanis et al. 2018b). It is unknown whether the absence of these findings in our animals is due to species differences, differences in the viral strain, or other factors. The number of diagnostically suitable spleens in our case set was limited due to postmortem autolysis, reducing the opportunity to detect splenic lesions.

While individualization of hepatocytes has been described previously (e.g., Neimanis et al. 2018b), the severity of hepatocellular dissociation in the livers we examined has not been emphasized in previous reports. Similarly, ocular abnormalities such as buphthalmos and blood within or around the eye have not been previously associated with GI.2/RHDV2/b infection in rabbits or hares. Although ocular abnormalities were suspected at necropsy in several animals, they could not be confirmed histologically, and their association with GI.2/RHDV2/b infection is unclear.

Based on the microscopic findings, death was probably acute to subacute; however, most rabbits and hares in this report were in poor body condition. It is unknown whether animals in poor body condition may have been more likely to become infected with GI.2/RHDV2/b or to have developed lethal disease, or whether poor body condition was an incidental finding. In a previous report, body condition of wild lagomorphs with GI.2/RHDV2/b infection varied from normal to emaciated (Neimanis et al. 2018a).

Compared to desert cottontails (n=8) and black-tailed jackrabbits (n=7) in the National Wildlife Health Center diagnostic database that were examined prior to 2020 or tested PCR-negative for GI.2/RHDV2/b at FADDL, the presence of massive hepatocellular dissociation with necrosis or apoptosis was the most diagnostically distinct feature of GI.2/RHDV2/b infection, as both epistaxis (11/15; 73%) and pulmonary edema or congestion (11/15; 73%) were common postmortem findings in GI.2/RHDV2/b–negative animals or those predating the introduction of GI.2/RHDV2/b into the North American wild lagomorph population, while hepatocellular dissociation was not noted in any case. The relative specificity of hepatocellular necrosis or apoptosis for GI.2/RHDV2/b infection was also posited by Harcourt-Brown et al. (2020) for pet O. cuniculus.

In Europe, wild rabbit population declines and associated ecologic changes, including declines in endangered predator populations, have been attributed to the introduction of GI.2/RHDV2/b (Monterroso et al. 2016). Given the susceptibility of both Sylvilagus spp. and Lepus spp., further efforts are needed to understand, control, and mitigate the impacts of GI.2/RHDV2/b in native North American lagomorphs.

The authors dedicate this work to the memory of Robert (Dr. Bob) Dittmar, 1955–2020. We thank US Department of Agriculture Foreign Animal Disease Diagnostic Laboratory's Diagnostic Services Section for GI.1/RHDVa and GI.2/RHDV2/b PCR testing, Anne Justice-Allen for guidance on the manuscript, and Barbara L. Bodenstein and other field epidemiologists at the US Geological Survey (USGS) National Wildlife Health Center. We also thank Bob Dittmar and others at Texas Parks and Wildlife Department, Kerry Mower and others at New Mexico Department of Game and Fish, and Michelle Verant, Chanteil Walter, Nancy Hendricks, Steve Richey, Pete Reser, and Kristen Philbrook from National Parks Service for submitting animals for necropsy. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the US Government. Data generated during this study are publicly available as a USGS data release: https://doi.org/10.5066/P9ZINO2P.