Special Focus: Modeling Lethal and Emergent Viral Diseases in Laboratory Animals–Original Article

Veterinary 2015, Vol. 52(1) 21-25 ª The Author(s) 2014 Experimental Aerosolized Guinea Pig– Reprints and permission: sagepub.com/journalsPermissions.nav Adapted Zaire (Variant: Mayinga) DOI: 10.1177/0300985814535612 Causes Lethal Pneumonia in Guinea Pigs vet.sagepub.com

N. A. Twenhafel1, C. I. Shaia1, T. E. Bunton1, J. D. Shamblin1, S. E. Wollen1, L. M. Pitt1, D. R. Sizemore1,M.M.Ogg1, and S. C. Johnston1

Abstract Eight guinea pigs were aerosolized with guinea pig–adapted Zaire ebolavirus (variant: Mayinga) and developed lethal interstitial pneumonia that was distinct from lesions described in guinea pigs challenged subcutaneously, nonhuman primates challenged by the aerosol route, and natural infection in . Guinea pigs succumbed with significant pathologic changes primarily restricted to the lungs. Intracytoplasmic inclusion bodies were observed in many alveolar macrophages. Perivasculitis was noted within the lungs. These changes are unlike those of documented subcutaneously challenged guinea pigs and aerosolized filoviral infections in nonhuman primates and cases. Similar to findings in subcutaneously challenged guinea pigs, there were only mild lesions in the liver and spleen. To our knowledge, this is the first report of aerosol challenge of guinea pigs with guinea pig–adapted Zaire ebolavirus (variant: Mayinga). Before choosing this model for use in aerosolized ebolavirus studies, scientists and pathologists should be aware that aerosolized guinea pig–adapted Zaire ebolavirus (variant: Mayinga) causes lethal pneumonia in guinea pigs.

Keywords aerosol, alveolar macrophages, animal model, ebolavirus, filovirus, guinea pig, lungs, Mayinga

Zaire ebolavirus (EBOV) is 1 of 5 single-stranded RNA EBOV 100% mortality when used to experimentally infect GPs by the belonging to the and the Ebo- subcutaneous route.4,6 One of these strains was selected for the lavirus. EBOV in humans causes . It is creation of a seed stock for use in animal model and counter- categorized by the National Institute of and Infectious measure development studies at the US Army Medical Diseases as a category A priority pathogen, and it is listed by Research Institute of Infectious Diseases (USAMRIID). Health and Human Services as a tier 1 ; conse- Detailed information on the pathology of this stock in GPs quently, all work must be conducted in high-containment challenged by the subcutaneous route was previously pub- laboratories ( 4). There has been an intense effort lished.4 This current study characterizes GPs challenged with in recent years to develop medical countermeasures that are the same strain of GP-EBOV by the aerosol, not subcutaneous, effective against natural and potentially weaponized (aeroso- route. lized) filoviral infections, including EBOV. This of research requires the use of animal models that closely reflect Materials and Methods the disease in humans, which is characterized by fever, bleed- ing, immune suppression, lymphocyte apoptosis, neutrophilia, Four male and 4 female 10- to 14-week-old Hartley GPs pur- increased liver enzymes, azotemia, and skin rash.12 Nonhuman chased from Charles River (Raleigh, North Carolina) weighing primates (NHPs), such as the rhesus macaque (Macaca mulatta), cynomolgus macaque (Macaca fascicularis), and 1US Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD African green monkey (Chlorocebus aetheops), are quickly becoming the gold standard for use in these studies; however, Supplementary material for this article is available on the Veterinary Pathology work in mice and guinea pigs (GPs; Cavia porcellus) is desir- website at http://vet.sagepub.com/supplemental. able in certain studies because of space, time, expense, and per- sonnel constraints. Wild-type EBOV does not cause morbidity Corresponding Author: Nancy A. Twenhafel, DVM, Diplomate ACVP, Pathology Division, US Army or mortality in mice and GPs; therefore, adapted strains have 2–4,9,11 Medical Research Institute of Infectious Diseases, 1425 Porter Street, Fort been developed. Sequential passage of EBOV (variant: Detrick, MD 21702. Mayinga) in GPs (GP-EBOV) yielded novel strains that caused Email: [email protected] 22 Veterinary Pathology 52(1)

Figure 1. Lung; Hartley guinea pig (GP), case No. 8. Gross ventral view with attached heart and noncollapsing red lung lobes that have rounded edges, appear consolidated and firm, and contain many variably sized lighter areas. Inset: Gross dorsal view; lung lobes are sharply mottled red and noncollapsing. Figure 2. Lung; Hartley GP, case No. 4. Interstitial pneumonia with congestion. Inset: Normal lung; Hartley GP, historical control. Hematoxylin and eosin (HE). Figure 3. Lung; Hartley GP, case No. 4. Alveolar spaces are filled with cellular and karyorrhectic debris, inflammation, and edema. HE. Figure 4. Lung; Hartley GP, case No. 5. The alveolar space contains karyorrhectic debris, viable and degenerate Twenhafel et al 23

506.1 to 670.3 g were used in the virulence confirmation study. contained diffuse changes. Intracytoplasmic inclusion bodies Animals were exposed in groups of 4 in a whole-body auto- were readily identified in alveolar macrophages. Perivasculitis, mated bioaerosol exposure system.8 The aerosol challenge edema, fibrin, congestion, hemorrhage, mesothelial cell hyper- dose for each animal was calculated with Guyton’s formula, trophy, type II pneumocyte hyperplasia, and rare pleuritis were which estimates the minute volume based on mass to determine multifocally present in all animals. BALT was detectable in the inhalation dose. The aerosol challenge was generated with a some animals and usually depleted of lymphocytes. Collison nebulizer to produce a highly respirable aerosol (flow Mild changes were observed in the liver, including multifo- rate, 7.5 + 0.2 l/min). The system generated a target aerosol of cal vacuolar degeneration, few small foci of hepatocellular 1- to 3-mm mass median aerodynamic diameter, determined by necrosis, and, in one animal, rare intracytoplasmic inclusion TSI Aerodynamic Particle Sizer. Samples of the prespray sus- bodies (Fig. 7). Within the splenic white pulp, there was mild pension and aerosol, collected from the exposure chamber with to moderate lymphocytolysis (Fig. 8), tingible body macro- an all-glass impinger during each challenge, was agarose pla- phages, and mild lymphoid depletion. que titrated to determine the inhaled plaque-forming units Positive EBOV IHC (Supplemental Table 2, Figs. 9–12) (pfu). Additionally, 100 ml of the prespray suspension and dilu- was demonstrated in the lung, alveolar macrophages, intravas- ent was plated onto blood agar to rule out potential contami- cular monocytes, perivascular spindle cells, pneumocytes, nants. Actual aerosol doses ranged from 1901 to 1992 pfu bronchiolar epithelial cells; cells of spleen (likely fibroblastic (mean, 1953 + 31.4 pfu). reticular cells, dendritic cells, and macrophages); cells of liver The GP-EBOV stock used in this study was created from (likely stellate and Kupffer cells); and, rarely, hepatocytes. ‘‘parent’’ stock No. 1394 by passage on VeroE6 cells. The iden- The current work was compared to a serial sampling study tity of this new stock was confirmed by agent-specific reverse performed at USAMRIID in 2010 that challenged by aerosol transcription real-time polymerase chain reaction assays, whole and used a created from the same parent stock. Hematox- sequencing, and electron microscopy (data not shown). ylin and eosin and IHC slides were examined from the USAM- Necropsies of each animal were performed, and lung, liver, RIID pathology archives, and aerosolized GPs of the archived and spleen were collected for histopathology and immunohisto- study developed marked to severe interstitial pneumonia that chemistry (IHC). Tissues were routinely processed and stained was similar, if not identical, to that noted in our study begin- with hematoxylin and eosin. Indirect IHC was performed on ning day 5 postexposure (PE; data not shown). the lung, liver, and spleen with an Envision-PO kit. A mouse monoclonal anti-EBOV antibody was used at a dilution of 1:8000. After deparaffinization and peroxidase blocking, sec- Discussion tions were covered with primary antibody and incubated. Sec- Necrotizing interstitial pneumonia was the primary lesion iden- ondary antibody was applied, rinsed with substrate-chromogen tified in GPs following aerosol challenge with GP-EBOV (days solution, and then stained with hematoxylin. 6–7 PE). This is distinct from pulmonary changes documented in a serial sampling study of subcutaneous-challenged GPs with GP-EBOV. In that report, diffuse thickening of alveolar Results interstitium by lymphocytes and macrophages occurred late At necropsy, every animal had mottled dark red lungs that were in the course of disease (days 7–9 PE).4 Additional dissimilar firm and noncollapsing (Fig. 1). Of special note, none of the findings included systemic vasculitis, endothelial cell infec- GPs exhibited a cutaneous macular rash, which is characteris- tion, and many intracytoplasmic inclusion bodies noted within tically present in human cases and aerosol-challenged splenic macrophages and hepatocytes.4 Similarities of subcuta- NHPs.5,10,12 Histologically, marked to severe interstitial pneu- neous- and aerosol-challenged GPs include lack of macular monia was present in 100% of the GPs (Supplemental Table 1, rash, moderate splenic lymphocytolysis with only mild loss Figs. 2–6). Pneumonia was characterized by thickened alveolar of splenic white pulp, and moderate disseminated EBOV septa that contained fibrin, cellular and karyorrhectic debris, immunopositivity; small foci of necrosis with mild EBOV heterophils, and macrophages that extended into alveolar immunopositivity in the liver; and infection of the mononuclear spaces. Multifocally, there was necrosis and disruption of the phagocytic system and interstitial cells.4 septa. There was infrequent necrosis of bronchiolar epithelial Results of this study contrast findings in documented cases cells, and inflammation and necrosis extended into the adjacent of NHPs challenged with aerosolized filoviruses1,5,10 where the interstitium, resulting in a vague bronchointerstitial pattern. pulmonary changes include infection and destruction of The pattern was not apparent in severely affected sections that respiratory-associated lymphoid tissues, increased alveolar

Figure 4. (Continued) heterophils, and alveolar macrophages. Inset: Degenerate alveolar macrophage with intracytoplasmic inclusion body. HE. Figure 5. Lung; Hartley GP, case No. 5. There is inflammation and karyorrhectic debris surrounding blood vessel (asterisk). Disrupted alveoli contain cellular debris, macrophages, and degenerate heterophils, and there is mesothelial cell hypertrophy (arrow). HE. Figure 6. Lung; Hartley GP, case No. 8. Histology of euthanized animal is similar to the animal found dead in Figure 5. Inflammation surrounds blood vessel (asterisk) and infiltrates the interstitium and alveoli; there is also karyorrhectic and cellular debris, mesothelial cell hypertrophy, and type II pneu- mocyte hypertrophy (arrow). Inset: Historical control, Hartley GP, normal lung. HE. 24 Veterinary Pathology 52(1)

Figure 7. Liver; Hartley GP, case No. 2. There is disorganization of hepatic cords, hepatocellular degeneration and necrosis, and few hetero- phils (arrow). Inset: Rare intracytoplasmic inclusion body. HE. Figure 8. Spleen; Hartley GP, case No. 6. Inset: Generally, the amount of white pulp appears similar to historical control (left). There is lymphocytolysis as well as moderate numbers of viable lymphocytes (retained white pulp; right). HE. Figure 9. Lung; Hartley GP, case No. 7. There is multifocal immunopositivity of the lung. Ebolavirus immunohistochemistry (EBOV IHC). Figure 10. Lung; Hartley GP, case No. 1. There is robust staining of alveolar macrophages. EBOV IHC. Figure 11. Spleen; Hartley GP, case No. 3. There is moderate disseminated immunopositivity of the red and white pulp. Inset: Robust staining of cells. EBOV IHC. Figure 12. Liver; Hartley GP, case No. 2. Multifocal immunopositivity in the nonparenchymal cells. Note the stained monocyte in portal vein (arrow). Inset: Rare hepatocytes are stained. EBOV IHC. Twenhafel et al 25 macrophages with occasional intracytoplasmic inclusion bod- Declaration of Conflicting Interests ies, mild inflammation, vasculitis, fibrin, and hemorrhage. In The author(s) declared no potential conflicts of interest with respect to human cases, there is pulmonary congestion and ‘‘absence of the research, authorship, and/or publication of this article. significant inflammatory cellular infiltration.’’12 Additional key features of aerosolized EBOV in NHP and human infection Funding include marked depletion of splenic white pulp, severe hepato- The author(s) received no financial support for the research, authorship, cellular degeneration and necrosis with many intracytoplasmic and/or publication of this article. inclusion bodies, systemic vasculitis with fibrin deposition in multiple organs and tissues, and a macular rash.5,7,12 References Briefly, the pathogenesis that we propose for aerosol GP- 1. Alves DA, Glynn AR, Steele KE, et al. Aerosol exposure to the EBOV based on this study is as follows: Inhaled virus infects Angola strain of virus causes lethal viral hemorrhagic alveolar macrophages, few bronchiolar epithelial cells, and fever in cynomolgus macaques. Vet Pathol. 2010;47:831–851. pneumocytes causing interstitial pneumonia. Infection 2. Bray M, Davis K, Geisbert T, et al. A mouse model for evaluation spreads to blood monocytes, liver, and spleen. Pneumonia of prophylaxis and of hemorrhagic fever. J Infect progresses rapidly, and animals die due to complications of Dis. 1999;179:248–258. severe interstitial pneumonia with few changes in the liver 3. Bradfute SB, Warfield KL, Bray M. Mouse models for filovirus and spleen, without widespread lymphoid destruction, with- infections. . 2012;4:1477–1508. out systemic vasculitis, and without a macular rash. Varia- 4. Connolly BM, Steele KE, Davis KJ, et al. Pathogenesis of experi- tions between models should be carefully weighed when mental Ebola virus infection in guinea pigs. J Infect Dis. 1999; considering the use of the GPs for EBOV studies involving 179:203–217. aerosol exposure. 5. Johnson E, Jaax N, White J, et al. Lethal experimental infection of This study was approved by the Institutional Animal rhesus monkeys by aerosolized Ebola virus. Int J Exp Pathol. Care and Use Committee, and research was conducted in 1995;76:227–236. compliance with the Animal Welfare Act and other federal 6. Mateo M, Carbonnelle C, Reynard O, et al. VP24 is a molecular statutes and regulations relating to animals at the time that determinant of Ebola virus virulence in guinea pigs. J Infect Dis. these studies were conducted. Experiments involving ani- 2011;204(suppl 3):S1011–S1020. malsadheredtoprinciples stated in the Guide for the Care 7. Murphy FA. Pathology of Ebola virus infection. In: SR Pattyn, ed. and Use of Laboratory Animals of the National Research Ebola Virus Haemorrhagic Fever. New York, NY: Elsevier/ Council. The USAMRIID is fully accredited by the Associ- North-Holland Biomedical Press; 1978:43–60. ation for Assessment and Accreditation of Laboratory Animal 8. Roy CJ, Reed DS, Hutt AJ. Aerobiology and inhalation exposure Care International. to biological select agents. Vet Pathol. 2010;47:799–789. The views, opinions, and/or findings contained herein are 9. Subbotina E, Dadaeva A, Kachko A, et al. Genetic factors of those of the authors and should not be construed as an official Ebola virus virulence in guinea pigs. Virus Res. 2010;153: Department of Army position, policy, or decision unless so 121–133. designated by other documentation. 10. Twenhafel NA, Mattix ME, Johnson JC, et al. Pathology of experimental aerosol Zaire ebolavirus infection in rhesus maca- Acknowledgements ques. Vet Pathol. 2013;50:514–529. We wish to gratefully acknowledge the numerous contributions of 11. Zumbrun EE, Abdeltawab NF, Bloomfield HA, et al. Develop- Pathology Division laboratory personnel, including Neil Davis, Phil ment of a murine model for aerosolized ebolavirus infection using Fogle, Gale Krietz, and Christine Mech; virology personnel William a panel of recombinant inbred mice. Viruses. 2012;4:3468–3493. M. Aguilar; Center for Aerobiology personnel; and Bill Discher for 12. Zaki SR, Goldsmith CS. Pathologic features of filovirus infections photographic assistance. in humans. Curr Top Microbiol Immunol 1999;235:97–116.