Elephant endotheliotropic herpesviruses (EEHVs) can cause fatal hemorrhagic disease in Asian (Elephas maximus) and African (Loxodonta africana) elephants. Of the seven known EEHV species, EEHV1 is recognized as the most common cause of hemorrhagic disease among Asian elephants in human care worldwide. Recent data collected from ex situ Asian elephants located in multiple North American and European institutions suggest that subclinical EEHV1 infection is common in this population of elephants. Although fatal EEHV1-associated hemorrhagic disease has been reported in range countries, data are lacking regarding the prevalence of subclinical EEHV infections among in situ Asian elephants. We used previously validated EEHV-specific quantitative real-time PCR assays to detect subclinical EEHV infection in three regionally distinct Asian elephant cohorts, totaling 46 in situ elephants in South India, during October and November 2011. Using DNA prepared from trunk washes, we detected EEHV1, EEHV3/4, and EEHV5 at frequencies of 7, 9, and 20% respectively. None of the trunk washes was positive for EEHV2 or 6. At least one EEHV species was detectable in 35% (16/46) of the samples that were screened. These data suggest that subclinical EEHV infection among in situ Asian elephants occurs and that Asian elephants may be natural hosts for EEHV1, EEHV3 or 4, and EEHV5, but not EEHV2 and EEHV6. The methodology described in this study provides a foundation for further studies to determine prevalences of EEHV infection in Asian elephants throughout the world.
Elephant endotheliotropic herpesviruses (EEHVs) are members of the Betaherpesvirinae and have been assigned to the genus Proboscivirus, which contains seven known members or species: EEHV1A, EEHV1B, EEHV2, EEHV3, EEHV4, EEHV5, and EEHV6 (Richman et al. 1999; Ehlers et al. 2001; Fickel et al. 2001; Garner et al. 2009; Latimer et al. 2011). The probosciviruses most commonly associated with morbidity and mortality in captive Asian (Elephas maximus) elephants are EEHV1A and EEHV1B (Richman et al. 1999; Fickel et al. 2001; Stanton et al. 2013), which account for the majority of fatal cases of herpesvirus-associated hemorrhagic disease examined in detail (Richman and Hayward 2011). Death due to EEHV1 infection is associated with widespread capillary endothelial-cell necrosis resulting in diffuse hemorrhagic disease, subcutaneous edema of the head and proboscis, lameness, and ultimately myocardial failure (Richman et al. 1999, 2000a, b). Death attributable to EEHV1 infection most commonly occurs in prereproductive, subadult Asian elephants (Richman and Hayward 2011).
Although EEHV1 is the proboscivirus most commonly associated with pathology, other probosciviruses produce morbidity and mortality in elephants. At least two African elephant calf deaths have been associated with EEHV2 infection (Richman et al. 1999) and EEHV3 and EEHV4 have each been associated with the death of an Asian elephant calf (Latimer et al. 2011). Blood samples collected from an apparently healthy adult Asian elephant gave rise to the discovery of EEHV5 (Latimer et al. 2011); however, since that time, an EEHV5 outbreak was detected in a herd of Asian elephants at a zoo in North America (Atkins et al. 2013). The outbreak resulted in the illness of one adult Asian elephant cow and subclinical infection of the remaining six elephants in the herd (Atkins et al. 2013). A transient detection of EEHV6 was found in the blood of a 1-yr-old African elephant calf with mild symptoms (Latimer et al. 2011).
Although the vast majority of reports describing EEHV1-associated deaths have originated from institutions holding captive Asian elephants in North America and Europe, EEHV1 is not exclusive to captive elephant populations. Case reports have emerged from range countries, including nine cases of EEHV1-associated deaths involving camp and free-ranging elephants in South India, as well as the death of a wild-born orphan elephant calf in Cambodia (Reid et al. 2006; Zachariah et al. 2013). These reports described Asian elephants with typical pathologic lesions associated with acute EEHV1 infection and were confirmed by PCR assays specific for EEHV1 DNA. Evidence of EEHV1 infection in South India is particularly concerning because the region holds the largest population of Asian elephants. Although EEHV1 mortalities have been reported in range countries, the prevalence and impact of proboscivirus infection among camp or free-ranging populations is unknown.
To conduct studies regarding the epidemiology of EEHV infection we developed and validated a sensitive quantitative real-time PCR (qPCR) assay that simultaneously detects both EEHV1A and EEHV1B, and demonstrated that EEHV1 was detectable in trunk secretions from a group of five healthy adult Asian elephants, thereby suggesting that Asian elephants can be persistently infected with EEHV1 (Stanton et al. 2010). Other researchers have since detected EEHV1 in conjunctival swabs, vaginal mucosal swabs, and trunk secretions from captive Asian elephants at multiple institutions in Europe, suggesting that subclinical EEHV1 infection may be common in captive populations (Schaftenaar et al. 2010; Hardman et al. 2012). To expand our ability to use qPCR to detect EEHV infection in clinically ill and healthy elephants, we developed and validated four additional qPCR assays that specifically detect EEHV2, EEHV3 and 4 with a single assay (EEHV3/4), EEHV5, and EEHV6 (Stanton et al. 2012). Although the ability to detect subclinical EEHV infection has improved greatly, the prevalence of subclinical proboscivirus infection among in situ or ex situ elephants has not been fully determined.
On the basis of recent reports of subclinical EEHV1 infection of captive Asian elephants in North America and Europe, as well as previous documentation of EEHV1-associated deaths in South India, we hypothesized that EEHV1, and possibly other EEHV species, would be detectable in trunk secretions or conjunctival swabs of Asian elephants in South India. To test this hypothesis, we traveled to the Wildlife Disease Research Laboratory in Kerala, India and used previously published methodology and qPCR assays to perform a cross-sectional study in which DNA prepared from trunk washes and conjunctival swabs, collected from three geographically distinct cohorts of Asian elephants in South India, were screened for the presence or absence of EEHV1, EEHV2, EEHV3/4, EEHV5, and EEHV6 DNA.
MATERIALS AND METHODS
These studies were conducted during October and November 2011 with the approval and participation of the Department of Forest and Wildlife in Kerala and Tamil Nadu, India. We collected samples from three Asian elephant cohorts in two states in South India: Mudumalai National Park, Tamil Nadu, India (11°35′N, 76°33′E); Kodanad Elephant Orphanage, Kodanad, Kerala, India (0°10′0″N, 76°31′0″E); and Guruvayur Sri Krishna Temple Elephants, Guruvayur, Kerala, India (10°35′40.2″N, 76°2′20.58″E; Fig. 1).
Animals and samples
Forty-six elephants were included in the study. The age range of the group was 3 mo to 72 yr and the median age was 35 yr. The cohort includes elephants that lived primarily in captive situations, such as the Guruvayur Temple elephants and juvenile orphan calves in Kodanad, to elephants that interacted extensively with free-ranging elephants such as the Kumki elephants at the elephant camp at Mudumalai National Park. All the elephants included in this study were routinely under human care at some point.
Trunk washes were collected and processed as described by Stanton et al. (2010). Trunk-wash samples were stored at 4 C until processed for DNA purification. Conjunctival swabs were collected using sterile cotton-tipped applicators placed in the conjunctival sac of one eye of an elephant, then stored in 1 mL of sterile saline solution at 4 C until processed for DNA using a commercially available DNA purification kit and the manufacturer's recommended protocol (DNeasy Blood and Tissue Kit, Qiagen Inc., Valencia, California, USA).
Quantitative real-time PCR
Following trunk-wash DNA preparation, 5 of 60 µL from each DNA preparation (n = 46) were screened using five independent qPCR assays: Asian elephant tumor necrosis factor-alpha (TNF); EEHV1; EEHV2; EEHV3/4; EEHV5; and EEHV6 (Stanton et al. 2012, 2013). The TNF assay detects Asian elephant genomic DNA and is included as an internal PCR amplification control to determine if samples contain amplifiable elephant genomic DNA (Stanton et al. 2013). As previously described, a single qPCR assay is used to detect a sequence of the EEHV3 and EEHV4 terminase gene that is 100% identical (Stanton et al. 2012). This assay will be referred to as the EEHV3/4 assay. All qPCR assays were performed using PCR primers, 5′-hydrolysis probes, and qPCR reaction reagents as previously described (Stanton et al. 2010, 2012). All qPCR assays were performed in duplex format.
The only modification to the previously established trunk-wash screening protocol was the use of a lightweight, rugged, easily portable, real-time qPCR platform (MiniOpticon System, Bio-Rad, Hercules, California, USA). Sample analysis was performed using software provided with the qPCR platform (CFX Manager, Bio-Rad). Assays performed on this instrument were indistinguishable from those conducted earlier (Stanton et al. 2012, 2013; data not shown). Test samples were considered positive if their relative fluorescence unit (RFU) value was 10% greater than the average RFU value for negative controls and the threshold cycle (Ct) value was <40, and if it could be reproduced in an independent assay. Negative control reactions were identical to test samples except that no template DNA was added to the reaction. Sample preparation controls, in which water was substituted for test sample (trunk wash or conjunctival swab), were also included in the screening program to control for potential contamination during DNA preparation.
We detected at least one EEHV species in 29% (6/21) of the trunk-wash samples collected from the cohort of camp elephants at Mudumalai National Park on 4 November 2011 (Table 1). Among this cohort, EEHV1 (14%), EEHV3/4 (10%), and EEHV5 (5%) were detected at varying frequencies, whereas EEHV2 and EEHV6 were not detected (Table 1). A single elephant, designated M20, had detectable EEHV1 and EEHV5 in one trunk-wash sample. After obtaining positive results of EEHV infection among this cohort of elephants, we returned to Mudumalai National Park on 11 November to obtain a limited number of recheck trunk-wash samples and opportunistic conjunctival swabs. The recheck trunk-wash sample for M20 again was positive for EEHV1 and EEHV5, but no other EEHV DNA was detectable (data not shown). Trunk-wash samples collected from M13 and M16 on 11 November were negative for all EEHVs. Elephants M10 and M14 were unavailable for rechecking EEHV shedding status. Conjunctival swabs collected from M13 and M20 were negative for all EEHVs and positive for the internal amplification control (TNF; data not shown).
EEHV screening for the cohort of Guruvayur Temple elephants collected on 28 October 2011 detected EEHV3/4 (11%) and EEHV5 (37%), but not EEHV2 and EEHV6 (Table 2). Notably, EEHV5 was detected at a substantially higher frequency when compared with the other EEHVs within this cohort and when compared with the prevalence of EEHV5 within the other two cohorts. The single trunk wash collected from elephant G5 was positive for EEHV3/4 and EEHV5.
Screening results for the Kodanad orphan elephant camp collected on 9 November 2011 yielded one positive trunk-wash sample, which was positive for EEHV5 (Table 3). All six conjunctival swabs collected from this cohort were positive for the internal PCR amplification control, but negative for all EEHVs for which we tested.
In this cross-sectional study evaluating the presence of proboscivirus DNA in trunk washes from 46 in situ Asian elephants in South India, the overall prevalence of infection with at least one EEHV species was 35% (16/46). Each EEHV species had a different prevalence in the overall study: EEHV1, 7% (3/46); EEHV2, 0% (0/46); EEHV3/4 9% (4/46); EEHV5, 20% (9/46); and EEHV6, 0% (0/46; Table 4). No proboscivirus DNA was detectable in the eight conjunctival swabs that were evaluated in the study.
This is the first report describing subclinical proboscivirus infection of Asian elephants in a native range country. Previous reports have described subclinical EEHV1 infection of captive Asian elephant populations in North America and Europe (Schaftenaar et al. 2010; Stanton et al. 2010; Hardman et al. 2012) and we did detect EEHV1 infection in this cohort of elephants. In a previous study, we detected EEHV1 DNA in trunk washes collected from a herd of five captive Asian elephants at the Houston Zoo and determined an overall prevalence of 31% over 15 wk (Stanton et al. 2010). That seemingly high prevalence is not surprising given that the previous study determined the period prevalence over 15 wk. When examining the cross-sectional prevalence, the data are similar to the results of our weekly trunk-wash screening program conducted at a North American zoo, in which the prevalence of EEHV1-positive trunk washes is low on a week-by-week basis (Stanton et al. 2010). Nearly all Asian elephants we have screened one to two times weekly for greater than 4 wk have detectable EEHV1 in trunk washes at some point (Stanton et al. 2010).
Hardman et al. (2012) detected EEHV1 DNA in swabs taken from the conjunctiva, palate, and vulva, as well as trunk washes, and conjunctival swabs proved to have the highest rate of EEHV1 detection. In the present study, we screened eight conjunctival swabs and detected no proboscivirus DNA. We believe these negative results are accurate because all of the conjunctival swabs were positive for our internal amplification control (TNF) validating that elephant DNA was purified from the swab. The results may be different from that of the previous study because different DNA purification methods, qPCR assays, and qPCR reagents were used, which can lead to discordant results between studies. In addition, EEHV1 shedding patterns may differ between elephants living under different conditions or in different environments. We found that collection of conjunctival swabs was less difficult than collection of trunk washes, suggesting that optimization of EEHV screening using conjunctival or other mucosal swabs could be useful for screening for EEHV infection from elephants that are not trained for providing trunk washes.
In this study we detected at least two EEHV species in addition to EEHV1. Atkins et al. (2013) reported that EEHV5 produced both clinical illness and subclinical infections in a herd of Asian elephants at a zoo in North America; however, we know of no reports of subclinical EEHV5 infection in other herds. In this study, EEHV5 was the most prevalent proboscivirus detected in trunk washes. We know of no reports of fatal hemorrhagic disease associated with EEHV5 infection. Whether EEHV5 is less pathogenic than other EEHVs is unknown.
We also obtained positive qPCR results using an assay that detects both EEHV3 and EEHV4 at an overall prevalence of 9%. A limitation of the assay is identifying the specific EEHV detected and attempts to obtain EEHV DNA sequence information from these samples failed. However, this is the first report of EEHV3 or EEHV4 detection in apparently healthy Asian elephants, suggesting that, similar to EEHV1 and EEHV5, EEHV3 or EEHV4 can produce subclinical infection in Asian elephants.
One limitation to our studies is the lack of sequence confirmation from positive trunk-wash samples. Generally, reasons for sequencing selected samples would be to determine whether the qPCR was detecting contaminant nucleic acids that were not EEHV or to rule out laboratory contamination. As we have shown previously (Stanton et al. 2010, 2012), all of the EEHV sequence-specific hydrolysis probes used in this study generate Cts only in the presence of the specific EEHV they were designed for, even when carried out in PCR amplifications over 40 cycles with other EEHV types. Furthermore, some of the EEHVs are only 10–20% diverged within genes detected by these assays (Richman and Hayward 2011), indicating that the level of specificity of these assays for detecting EEHV sequences is strong and the chance of amplifying non-EEHV nucleic acids is likely small. As discussed in “Materials and Methods,” we also applied two criteria to define a positive. Finally, all of our assays detect their respective EEHV targets down to 10 copies with 100% sensitivity and down to 1 copy with approximately 75% sensitivity (Stanton et al. 2010, 2012). When assaying samples with a single EEHV copy, we always obtained Ct values of <40 and negative controls or samples with significant levels of other EEHV types did not generate any Ct during 40-cycle PCR amplifications. Therefore, we feel that any Ct value obtained during a 40-cycle amplification is likely to represent detection of the specific EEHV targeted by the hydrolysis probe, provided that the nontemplate controls are negative. Logistic constraints prevented us from conducting standard curves for each experiment and, although it is difficult to compare our results with those we obtained in earlier studies, the Ct values obtained in this study likely indicate that several of our samples fall within the lower range of detection in our assays (i.e., 1–10 copies). To address concerns that the positives detected in this study were derived from laboratory contamination, we prepared several mock DNAs using water in place of trunk-wash material during the sample preparation phase. These mock samples were then subjected to qPCR analysis along with the test samples. In addition to normal nontemplate controls included in the qPCR reactions, these samples did not generate any Cts when the reactions were carried out to 40 cycles of amplification. Therefore, on the basis of extensive testing of these sequence-specific qPCR assays and rigorous inclusion of multiple controls, we believe that the positive samples obtained in this study are correlated with EEHV DNAs present within trunk-wash samples obtained from the indicated elephants.
More important, neither EEHV2 nor EEHV6 was detected in this cohort of elephants. This is not surprising given that neither of these viruses has been identified in samples from Asian elephants. These viruses have only been associated with infection of African elephants with EEHV2 (producing lethal hemorrhagic disease in at least two calves) and EEHV6 (detected in a mildly ill adult) (Richman et al. 1999; Latimer et al. 2011). The absence of detectable EEHV2 or EEHV6 DNA in trunk secretions in this cohort may suggest that these viruses do not produce subclinical infection of Asian elephants. It is possible that EEHV2 and EEHV6 are natural African elephant probosciviruses because they have been detected in benign lung nodules in adults and skin nodules of juveniles (Hayward 2012).
Our data strongly suggest that proboscivirus infection is not unique to captive elephants in North America and Europe. Consistent with our findings, EEHV-associated deaths have now been documented in the same geographic regions as the elephants in this study (Zachariah et al. 2013). Additional research should focus on identifying the genotypes of the EEHVs circulating among elephants in the Indian subcontinent and whether they are similar to the diverse types characterized in captive elephants living outside of range countries. Our data add to the growing body of work suggesting that EEHVs have coevolved with both African and Asian elephants and leaves open the possibility that some probosciviruses are native to African elephants and some are native to Asian elephants. This work also provides a foundation for future studies to determine the overall prevalence of proboscivirus infection in elephants throughout the world.
We thank Gary Hayward, Laura Richman, Simon Long, and Erin Latimer of Johns Hopkins University and the National Elephant Herpesvirus laboratory as well as the Kerala Forestry Department for use of their EEHV diagnostic laboratory facility in the Waynad Wildlife Sanctuary in Kerala. We thank Mark Lawson from Bio-Rad laboratories for assisting with the use of the Mini-Opticon and CFX manager software. Financial support for these studies was provided by the International Elephant Foundation and the Houston Zoo. S.A.N. was supported by National Institutes of Health training grant T32-AI-07471.