Pandemic (H1N1) 2009 influenza A virus infection associated with respiratory signs in sloth bears (Melursus ursinus)

Summary

In 2009, a pandemic influenza A virus (pH1N1) spread globally in humans and infected a broad range of captive animals with close human contact. In February 2014, a pH1N1 virus was isolated from a sloth bear with respiratory signs at a US zoo, demonstrating that recurring epidemics present an ongoing threat to animals, including threatened species. This is the first report of pH1N1 infection in sloth bears. To understand the sloth bear virus within the global context of pH1N1, phylogenetic trees were inferred including full-length sequences from available non-human, non-swine hosts, representing four families in the order Carnivora and one order of birds. A combination of phylogenetic and epidemiological evidence strongly suggests the sloth bear was infected with a human-origin pH1N1 virus, supporting the implementation of biosecurity measures to protect human and animal health.

1 ∣. INTRODUCTION

The capacity of influenza A viruses (IAVs) to periodically spill over into non-host species presents an ongoing pandemic threat. Minimizing opportunities for the adaptation of IAV in a new reservoir species is important to help prevent the emergence of novel strains. During 2009–2010, the newly emerged pH1N1 virus infected a wide diversity of non-human hosts, including both mammals and birds, both domestic (cat, dog, ferret) and wild species, and species with no prior record of influenza infection, including black-footed ferret, skunk, cheetah, badger, binturong, and giant panda (Britton, Sojonky, Scouras, & Bidulka, 2010; Crossley et al., 2012; Li et al., 2014; Schrenzel et al., 2011). In many of these exotic species, infection was suspected to have been transmitted from human caretakers, a process termed “reverse zoonosis.” The risks presented by reverse zoonosis have been most apparent in swine, where the large-scale reverse zoonotic transmission of pH1N1 from humans greatly accelerated the evolution of IAV diversity in pigs (Nelson & Vincent, 2015), generating novel reassortant viruses with the capacity to transmit back to humans (Jhung et al., 2013). Here we report the first evidence of pH1N1 infection in sloth bears (Melursus ursinus), a Convention on International Trade in Endangered Species (CITES) Appendix I species listed by the International Union for Conservation of Nature (IUCN) as vulnerable and decreasing primarily due to habitat loss and poaching (CITES, 2017; Dharaiya, Bargali, & Sharp, 2016).

2 ∣. CASE REPORT

On 24 February 2014, an 8-week-old hand-reared sloth bear at Smithsonian’s National Zoo in Washington DC presented with respiratory signs including mucoid nasal discharge, coughing, and sneezing, as well as lethargy, mild pyrexia, and decreased appetite and weight gain. Nasal swabs were collected 25 February 2014 and tested for IAV by real-time reverse transcriptase polymerase chain reaction (rRT-PCR) and virus isolation. IAV viral RNA was detected, and H1N1 was isolated. The isolate (A/sloth bear/District of Columbia/6365/2014(H1N1)) was sequenced at the National Veterinary Services Laboratories of USDA-APHIS and found to be closely related to pH1N1 viruses circulating in humans.

Human-to-bear transmission of the pH1N1 virus, concurrently circulating in the human population, was strongly suspected. Several of the sloth bear cub’s caretakers had developed influenza-like illness within 1 month prior to the onset of respiratory signs in the cub. The four other juvenile and adult sloth bears housed separately in the same building also presented with respiratory illness during this time. The cub, hand-reared in isolation from the other bears, was the last to develop clinical signs. Bear-to-bear or bear-to-human transmission could not be confirmed. The cub was treated with supportive care, and clinical signs resolved within 12 days. Nasal swabs collected from the cub on 3/12/14 and 3/27/14 after clinical signs had resolved were negative on rRT-PCR. Serum collected from the cub on 2/28/14, 3/12/14, and 3/27/14 was tested by hemagglutination inhibition and showed a rising titre to pH1N1 (1:20, 1:80, 1:320, respectively).

Only one of the other bears at the Zoo was tested subsequent to this 2014 respiratory disease episode and that adult male had seroconverted to pH1N1 with a positive titre (1:320) detected from serum collected in April 2014. Banked sera from five Zoo sloth bears dating back as far as 2002 were also tested, and none demonstrated seroconversion to pH1N1. Serologic evidence revealed three of these bears (all still in the facility through 2014) with titres to H3N2 (also a predominant subtype in humans), indicating an introduction of H3N2 IAV into the sloth bear collection prior to this 2014 introduction of pH1N1. In 2011, a respiratory outbreak in area keepers and bears was documented, and all sloth bears present at the zoo during that time had respiratory signs and subsequently seroconverted to H3N2. The two bears without serologic evidence of IAV exposure had been removed from the Zoo collection prior to the 2011 respiratory outbreak.

3 ∣. MATERIALS AND METHODS

Hemagglutination inhibition tests were performed on sera using antigens and antisera generated against human seasonal strains A/Brisbane/59/07 (H1N1), A/Michigan/2/2003 (H1N2), A/New York/18/2009 (H1N1), A/Wuhan/359/1995 (H3N2), A/Moscow/10/1999 (H3N2), and A/Brisbane/10/2007 (H3N2) using methods previously described (OIE, 2012).

Swabs were screened for the presence of IAV by rRT-PCR targeting the matrix and nucleoprotein genes (VetMAX^™-Gold SIV Detection Kit; Life Technologies, Carlsbad, CA, USA). Hemagglutinin (HA) and neuraminidase (NA) subtype determination was performed with rRT-PCR using a commercially available swine influenza viral subtyping kit (VetMAX^™-Gold SIV Subtyping Kit; Life Technologies).

For virus isolation, swabs were diluted 1:2 in cell culture medium and inoculated onto confluent Madin–Darby canine kidney (MDCK) cells treated with TPCK trypsin and supplemented with antibiotics. Flasks were incubated at 37°C and observed daily for cytopathic effect (CPE).

Whole-genome sequences were generated for virus isolates on the Ion Torrent Personal Genome Machine, as previously described. Briefly, sequencing was performed using the Ion PGM 200 v2 sequencing kit on an Ion 314 chip (Bowman et al., 2012). To understand the virus infecting the sloth bear within the global context of pH1N1, we inferred phylogenetic trees including full-length sequences from available non-human hosts, representing four families in the Carnivora order (Ursidae, Phocidae, Felidae, and Mustelidae) and one order of birds (Galliformes) (n = 23 for the HA segment; n = 27 for the NA segment, Table 1). Swine were excluded from the analysis due to the high frequency of reverse zoonosis of pH1N1 to swine since 2009 (>800 available HA sequences). As background, a representative data set of human pH1N1 viruses circulating during 2009–2014 was included (Table 1). The phylogeny was inferred using the time-scaled Bayesian approach using Markov chain Monte Carlo (MCMC) available via the BEAST v1.8.2 package (Drummond, Suchard, Xie, & Rambaut, 2012) and the high-performance computational capabilities of the NIH Biowulf Linux cluster (http://hpc.nih.gov). A strict molecular clock was used, with a constant population size and a general-time reversible (GTR) model of nucleotide substitution with gamma-distributed rate variation among sites. The MCMC chain was run separately four times for at least 350 million iterations with subsampling every 35,000 iterations, using the BEAGLE library (Suchard & Rambaut, 2009) to improve computational performance. All parameters reached convergence, as assessed visually using tracer v.1.6 (http://beast.bio.ed.ac.uk/tracer).

TABLE 1.

Human pH1N1 virus isolates identified in non-human hosts during 2009–2014. The isolate from the sloth bear cub in this report is shown in italics

Virus name	Region	Host status	Date	Accession	Gene	Reference
A/canine/Beijing/cau2/2009/H1N1	Asia	Captivity	11/7/09	AEM89464	NA	Lin et al. (2012)
A/canine/Beijing/cau9/2009/H1N1	Asia	Captivity	11/20/09	AEM89474	NA	Lin et al. (2012)
A/canine/Korea/1/2010(H3N1)	Asia	Captivity	2/5/10	AED99983	NA	Song et al. (2012)
A/canine/Korea/VC125678/2012(H3N1)	Asia	Captivity	3/13/10	AIT38386	NA	Song et al. (2012)
A/cat/France/0514/2009/H1N1	Europe	Unknown	12/2009	CCB84488 CCB84487	HA NA
A/cat/IA/26991/2009/H1N1	North America	Captivity	2009	ADA70107 ADA70110	HA NA	Sponseller et al. (2010)
A/cat/Italy/304678-1/2009/H1N1	Europe	Semi-captivity	12/17/09	ADF58335 ADF58337	HA NA	Fiorentini et al. (2011)
A/cat/Italy/304678-2/2009/H1N1	Europe	Semi-captivity	12/17/09	ADF58336 ADF58338	HA NA	Fiorentini et al. (2011)
A/cat/OR/29573/2009/H1N1	North America	Captivity	11/9/09	ADB66689 ADB66692	HA NA
A/cat/PA/30187/2009/H1N1	North America	Unknown	11/7/09	ADD64912 ADD64913	HA NA
A/cheetah/CA/30954/2009/H1N1	North America	Captivity	11/17/09	ADD91632 ADD91633	HA NA	Crossley et al. (2012)
A/cheetah/California/D0912239/2009/H1N1	North America	Captivity	11/2009	AEI84811 AEI84813	HA NA	Crossley et al. (2012)
A/elephant seal/California/1/2010/H1N1	North America	Wild	05/2010	AFV31454 AFV31456	HA NA	Goldstein et al. (2013)
A/elephant seal/California/2/2010/H1N1	North America	Wild	04/2010	AGA19350 AGA19352	HA NA	Goldstein et al. (2013)
A/ferret/OR/23775/2009/H1N1	North America	Captivity	10/5/09	ADD37824 ADD37827	HA NA
A/ferret/OR/27004-3/2009/H1N1	North America	Captivity	10/23/09	ADD60249 ADD60252	HA NA
A/ferret/OR/27004/2009/H1N1	North America	Captivity	2009	ADA70111 ADA70114	HA NA
A/ferret/Taiwan/E01/2013/H1N1	Asia	Captivity	07/2013	AHY95082 AHY95085	HA NA	Lin et al. (2014)
A/ferret/Washington/WADDL13230/2009/H1N1	North America	Captivity	11/26/09	ADP37367 ADP37368	HA NA
A/giant panda/Ya an/01/2009/H1N1	Asia	Captivity	11/2009	AGR33808 AGR33811	HA NA	Li et al. (2014)
A/sloth bear/District of Columbia/6365/2014/H1N1	North America	Captivity	2/25/14	AIK26333 AIK26335	HA NA
A/turkey/Chile/28317-6504-3/2009/H1N1	South America	Captivity	8/17/09	ACV52963 ACV52964	HA NA	Mathieu et al. (2010)
A/turkey/Ontario/FAV110-4/2009/H1N1	North America	Captivity	11/3/09	ADI52836 ADI52832	HA NA	Berhane et al. (2010)
A/turkey/Ontario/FAV110/2009/H1N1	North America	Captivity	10/23/09	ADI52837 ADI52833	HA NA	Berhane et al. (2010)
A/turkey/Ontario/FAV114-17/2009/H1N1	North America	Captivity	11/3/09	ADI52835 ADI52831	HA NA	Berhane et al. (2010)
A/turkey/Ontario/FAV117-1C/2009/H1N1	North America	Captivity	12/7/09	ADI52834 ADI52830	HA NA	Berhane et al. (2010)
A/turkey/Virginia/4135/2014/H1N1	North America	Captivity	2/3/14	AIK26323 AIK26325	HA NA

4 ∣. RESULTS

The time-scaled maximum clade credibility (MCC) tree (Figure 1) indicates that the majority of pH1N1 infections detected in non-human/non-swine hosts occurred during the initial year of the H1N1 pandemic (2009–2010), when pH1N1 activity was highest in humans (Figure 2). During this time, reverse zoonosis of pH1N1 occurred in North America, South America, Asia, and Europe (Table 1). Cases were reported primarily among captive animals with high contact rates with humans, including pets (e.g., cats, dogs, ferrets; Lin et al., 2012; Lin, Wang, Wu, Chi, & Wang, 2014; Song et al., 2012; Sponseller et al., 2010), poultry (e.g., turkeys; Berhane et al., 2010; Mathieu et al., 2010), and wild animals maintained in captivity (e.g., cheetahs, giant pandas; Figure 1, Table 1). The pH1N1 virus also was isolated from wild elephant seals in California (Goldstein et al., 2013) and from cats maintained in a colony in Italy with little human contact (Fiorentini et al., 2011).

FIGURE 1.

MCC tree of pH1N1 in humans and non-human hosts. Time-scaled Bayesian MCC tree inferred for H1 sequences from 325 pH1N1 viruses collected during 2009–2014 from humans (n = 300; 50 randomly sampled per year) and 23 non-human hosts, including A/sloth bear/District of Columbia/6365/2014(H1N1). Viruses from non-human hosts are indicated by colour, the shade of which corresponds to the family: green = Felidae, light blue = Phocidae, dark blue = Mustelidae, red = Ursidae, and brown = all avian species. Posterior probabilities are provided for key nodes. Closely related viruses from the same host species are collapsed as triangles

FIGURE 2.

pH1N1 activity in the USA, 2009–2014. Proportion of total influenza virus specimens that were positive for pH1N1 was calculated for each week from the beginning of the pandemic in the USA (week 35 of 2009), as reported by the Centers for Disease Control and Prevention (CDC, 2014b). The non-human hosts infected with pH1N1 viruses in the USA and Canada during each annual epidemic are listed as follows

Cases of pH1N1 declined in humans after the first year of the pandemic, when the H3N2 subtype became predominant again, and no infections of pH1N1 in non-human/non-swine hosts were observed in the USA or Canada from mid-2010 to the end of 2013 (Figure 2, Table 1). However, pH1N1 re-emerged as the dominant subtype during the influenza epidemic of 2013–2014 in the USA, when pH1N1 represented ~95% of all subtyped IAVs in humans during the winter peak. During the epidemic’s highest level of influenza activity (December 2013–February 2014), two new cases of reverse zoonosis of pH1N1 were identified, both in the Mid-Atlantic region: A/turkey/Virginia/4135/2014(H1N1) on 3 February 2014, and the isolate reported here, A/sloth bear/District of Columbia/6365/2014(H1N1) on 25 February 2014 (Figure 2, Table 1). The District of Columbia was one of only four US states still reporting high ILI activity during the eighth week of 2014 (CDC, 2014a), consistent with a human source for the bear’s infection.

5 ∣. DISCUSSION

A combination of phylogenetic and epidemiological evidence strongly suggests that a sloth bear cub at the Smithsonian National Zoo was infected with a human-origin pH1N1 virus in February 2014. This finding confirms that the risk of reverse zoonosis for a broad diversity of species has persisted into the post-pandemic period, consistent with what has been described in US swine (Nelson, Stratton, Killian, Janas-Martindale, & Vincent, 2015). Although the cub in this report recovered with supportive care, pH1N1 infection has been associated with severe morbidity and mortality in diverse species including the domestic cat, striped skunk, Bornean binturong, and American badger (Britton et al., 2010; Campognolo et al., 2011; Fiorentini et al., 2011; Schrenzel et al., 2011). Biosecurity measures should be considered to protect human and animal health, including recommendations that animal caregivers in zoos and others with close animal contact receive annual influenza vaccinations. These findings also highlight a central outstanding question regarding the extent of unreported reverse zoonosis of pH1N1 to different species, as detection generally is limited to closely monitored animals with relatively severe symptoms. Another outstanding question is whether (and why) pH1N1 has a greater capacity for reverse zoonosis than the H3N2 subtype that has circulated in humans since 1968. Evidence of past H3N2 virus infection in the sloth bear’s banked sera suggests that reverse zoonosis transmissions involving H3N2 have gone unreported. Further surveillance and testing are greatly needed to understand the scope and risk of reverse zoonosis of IAVs in different species.

Impacts.

• An infection of a sloth bear with the pH1N1 (2009) influenza A virus is confirmed for the first time. Human origin of infection is strongly suspected.

• Recurring pH1N1 epidemics in humans present an ongoing threat to a wide variety of animals, including threatened species.

• Increased awareness of the risk of reverse zoonosis in zoological settings supports enhanced biosecurity measures, including annual influenza vaccination for staff with animal contact.