First Report of a Novel Hepatozoon sp. in Giant Pandas (Ailuropoda melanoleuca)

The first report of giant pandas (Ailuropoda melanoleuca) infected with a novel Hepatozoon species is presented. An intraleukocytic parasite was detected via routine blood smear from a zoo-housed giant panda at the National Zoological Park. Ribosomal DNA sequences indicated a previously undescribed Hepatozoon species. Phylogenetic and distance analyses of the sequences placed it within its own branch, clustered with Old World species with carnivore (primarily ursid and mustelid) hosts. Retrospective and opportunistic testing of other individuals produced additional positive detections (17/23, 73.9%), demonstrating 100% prevalence (14/14) across five institutions. All animals were asymptomatic at time of sampling, and health implications for giant pandas remain unknown.


INTRODUCTION
Hepatozoon spp. are apicomplexan hemoparasites capable of infecting a wide range of vertebrate taxa globally, including canids, ursids, felids, and numerous others (André et al. 2010;East et al. 2008;Kubo et al. 2006Kubo et al. , 2008Kubo et al. , 2010Pawar et al. 2011Pawar et al. , 2012. In North America, Hepa-tozoon infections have been documented in domestic canids, wild coyotes (Canis latrans), and other small carnivores (Kocan et al. 2000;Mercer et al. 1988). Clinical disease is largely uncommon but can affect domestic dogs, causing musculoskeletal lesions and potentially death (Baneth et al. 2003; Barton et al. 1985). Hepatozoon spp. have heteroxenous life cycles, requiring definitive invertebrate hosts and intermediate vertebrate hosts (Smith 1996 transmitted via host ingestion of the arthropod during grooming or consumption of infected prey or prey with infected ticks (Allen et al. 2011;Baneth et al. 2003;Smith 1996).
This report describes a novel Hepatozoon parasite found in captive zoo-housed giant pandas (Ailuropoda melanoleuca). Giant pandas are ursids native to central China and an iconic ''flagship'' species for conservation. Currently categorized as vulnerable by the IUCN, the species faces continued threats from habitat loss and fragmentation, starvation, and infectious diseases (Feng et al. 2015;Swaisgood et al. 2016;Zhang et al. 2008).
In March 2005, an intraleukocytic hemoparasite was detected on a routine blood smear of a male giant panda housed at the Smithsonian National Zoological Park (NZP) in Washington, DC (listed as GP4 in Table 1). The blood smear demonstrated neutrophil-associated parasites with a morphology similar to that of known Hepatozoon species. The specimen could not be identified to species by light microscopy.
DNA was extracted from the sample using DNeasy kits (Qiagen) and amplified via polymerase chain reaction (PCR) using a combination of 18S ribosomal DNA primer sets (Hep18S2-H, Hep18S4-L, Hep18S4-H, BT1-L, BT1-H, BTH1-L, and BTH1-H) in a 25-lL PCR. Sequencing was conducted on an ABI 3730, and sequences were aligned and edited using Sequencher 4.1. A total length of 1092 bp of 18S sequence was obtained for GP4. His mate, GP5 (MeiXiang, 113607), was also positive, and we obtained 1111 bp of sequence for her which was identical to that of GP4. The latter sequence was utilized in subsequent analyses. Blasts to GenBank sequences from morphologically identified specimens and phylogenetic analyses confirmed identification of the parasite lineage to genus (Hepatozoon).
The program RAxML 8.2.11 (Stamatakis 2014) implemented in Geneious (Kearse et al. 2012) was used to estimate phylogenetic relationships within genera using a maximum likelihood (ML) criterion and a general timereversible (GTR+I+gamma) model of nucleotide substitution, with 1000 bootstrap replications. jModelTest (Darriba et al. 2012;Guindon and Gascuel 2003) was used to identify the best evolutionary model for the sequence data. Under BIC and DT (decision theory) criteria, the GTR+I+G model was preferred, while the AICc criterion favored the slightly different TPM1uf+I+G model. The estimated I and G values in all models were nearly identical (0.49 and 0.89, respectively). Therefore, we selected the GTR+I+G model from the BIC for the ML analysis.
In the resulting phylogeny, the giant panda-associated Hepatozoon lineage occupied its own distinct, long branch, nested in a clade with Old World species H. felis, H. ursi, and H. martis (Fig. 1). The consensus phylogeny suggested a genetically unique parasite, most closely related to a sister clade containing H. ursi, a parasite of Asiatic bears (Kubo et al. 2008;Pawar et al. 2011), and H. martis, a mustelid parasite (Hodžića et al. 2018), but bootstrap support of 52% was marginal for this node, and it essentially collapses to a trichotomy. The giant panda Hepatozoon 18S sequence was 3.2 to 3.6% divergent (uncorrected) from those of H. ursi and H. martis, but 4.5% divergent from all other carnivore Hepatozoon sequences. Thus, we found support for a single clade containing ursid (including panda) and mustelid Hepatozoon-derived sequences. We also found two paraphyletic H. felis clades, with one falling out as a poorly supported sister clade to H. americanum. However, analyses with additional H. felis sequences sometimes removed the paraphyly and the two H. felis clades became poorly supported sister clades. Regardless, the trees and data support that the giant panda Hepatozoon is a distinct lineage and species and that it is most closely related to Hepatozoons with ursid and mustelid hosts.
Conventional PCR was then performed on 23 archived or opportunistically collected blood and tissue samples acquired from 14 giant pandas between 1982 and 2006. Primer sets BT1 (432 bp amplicon) and BTH1 (751 bp amplicon) provided the greatest amplification and sequencing consistency and were tested on all available samples. Additional sequencing was performed as described, and for most individuals, up to 1113 bp were obtained. Whole blood samples were evaluated where possible, but when unavailable, other tissue types, plasma, or stained blood smears were substituted. For six pandas, multiple samples were used. The samples had been collected from seven adult males, six adult females, and one male neonate from five institutions (three in the USA, one in the United Kingdom, and one in China). All individuals were born in China (either wild-caught or captive-bred) with the exception of the neonate, which was born at the NZP but did not survive beyond a few days. None of the animals were reported to have clinical signs of disease consistent with hepatozoonosis at the time of sampling.
The PCR results from testing 14 captive giant pandas are summarized in Table 1. All individuals (14/14, 100%) sampled demonstrated positive tests for Hepatozoon sp., identical in overlapping sequence to that of GP4 and GP5. Positive detections were made in 17 samples out of 23 tested (73.9%). Of the negative results, four resulted from stained blood smears and two from whole blood samples. These were presumed to be false negatives, attributed to poor quality of archival samples and varying extraction efficiency and PCR sensitivity due to DNA inhibitors (Scopel et al. 2004;Shavey and Morado 2012). The sequence for giant panda 113607 was deposited in GenBank (accession number pending MK645858). Although the level of parasitemia was not quantified, infrequent findings on routine microscopy and occasional negative results by PCR suggest that the parasite may be present at low levels or intermittently in circulation (Otranto et al. 2011;Scopel et al. 2004). Challenges in detecting the parasite without molecular methods may explain why the parasite was not previously found. It is possible that the parasite is abundant in other tissues, as these may be sites of merogony and cyst formation in mammalian Hepatozoon life cycles (Smith 1996), but tissues were not thoroughly investigated here; testing of other individuals and tissue types may produce additional positive detections.
To our knowledge, this is the first report of a Hepatozoon infection described in giant pandas. The pathogenicity of this novel species and the health and conservation implications for its vertebrate host are unknown. Like many wildlife species found with asymptomatic Hepatozoon infections (Clark et al. 1973;McCully et al. 1975;Pawar et al. 2012), the individuals tested here displayed no apparent clinical illness attributable to hepatozoonosis. The high prevalence and penetration into the population demonstrated here may indicate low pathogenicity (Best et al. 2014).
Given the giant panda's vulnerable conservation status, however, the presence of any identified pathogen warrants consideration of clinical and conservation management implications. Infectious disease outbreaks can have considerable and disproportionate effects on small populations (De Castro and Bolker 2005;Smith et al. 2006). While Hepatozoon infection may be an incidental finding in otherwise healthy captive giant pandas, it may have significant consequences for juvenile, geriatric, or otherwise immunocompromised individuals (Kocan et al. 2000). Concomitant disease or coinfections could precipitate increased parasite load and potentially heightened pathogenicity (McCully et al. 1975;Simposon et al. 2013). Giant pandas are prone to a suite of health issues, including gastrointestinal disorders, infectious diseases, infertility, and others which are incompletely understood (Feng et al. 2015;Qiu and Mainka 1993;Williams et al. 2016;Zhang et al. 2008). The extent to which a background Hepatozoon infection may contribute to illness in these cases is unknown.
Considering a common geographic origin for 13/14 individuals, it is likely that the Hepatozoon infections were not locally acquired at receiving institutions but rather were already present in the animals upon arrival. The high prevalence demonstrated here could indicate: an enzootic infection of giant pandas, where they may be the parasite's natural host; a high rate of exposure; or an elevated susceptibility to a novel infection due to either host or agent factors. The genus has a wide distribution globally, including species identified from China Xu et al. 2015). Phylogenetic relationships also support a plausible Asiatic origin, as the species clusters with Old World Hepatozoon species. Its proximity to H. ursi is expected from an evolutionary standpoint and consistent with an Asiatic origin. Hepatozoon ursi is a parasite of other Asiatic ursids like Japanese black bears (Ursus thibetanus japonicas) and Indian sloth bears (Melursus ursinus) (Kubo et al. 2008;Pawar et al. 2011), and to date, North American ursids have not been reported with Hepatozoon infections.
The route of transmission remains to be described. Cases of tick-infested free-ranging giant pandas have been reported (Qiu and Mainka 1993), although possible vector species were not sought or identified in these cases. The positive detection in a neonate suggests a potential vertical transmission route (Allen et al. 2011;Murata et al. 1993). Because host specificity of this novel Hepatozoon species is unknown, the impact of an introduced hemoparasite on local mammalian populations is unclear. The potential risk to both native wildlife and giant pandas may warrant stringent Hepatozoon surveillance in the captive population.
Further research is necessary to assess whether this Hepatozoon species constitutes a potential pathogen of giant pandas and to characterize the extent of threat to the species. Epidemiologic studies may help determine whether there is a correlation between the degree of parasitemia and any occurrence of clinical signs. Prospective studies could investigate prevalence in captive individuals held globally in other facilities. The parasite's infectivity, pathogenicity, range, prevalence, and vector species in free-ranging populations are unknown, but should remain priorities for future investigation. Caution may also be indicated in the international movement of captive individuals. In domestic canids, hepatozoonosis is typically managed using antiprotozoal and palliative treatments (Allen et al. 2011). Efforts could be directed at identifying safe and effective antiprotozoal therapies and parasite management for giant pandas for implementation during routine quarantine procedures prior to animal transport. Thus, additional studies can elucidate pathogen, host, and vector relationships and identify the need for targeted conservation management actions for this vulnerable species.