|
Journal of General Virology |
| SUMMARY | INTRO | METHODS | RESULTS | DISCUSSION | FOOTNOTES | REFS |
| First posted online 19 May 2000 | FULL-LENGTH ARTICLE |
| Rec 10 March 2000; Acc 20 April 2000 | DOI: 10.1099/vir.0.17031-0 |
K. Halpin,1,2 P. L. Young,3 H. E. Field2 and J. S. Mackenzie1
1 Department of Microbiology and
Parasitology, The University of Queensland, St Lucia, Queensland 4072,
Australia
2 Department of Primary Industries, Animal
Research Institute, 665 Fairfield Road, Yeerongpilly, Queensland 4105,
Australia
3 Queensland Agricultural Biotechnology Centre,
Gehrmann Laboratories, St Lucia, Queensland 4072, Australia
Since it was first described in Australia in 1994, Hendra virus (HeV) has caused two outbreaks of fatal disease in horses and humans, and an isolated fatal horse case. Our preliminary studies revealed a high prevalence of neutralizing antibodies to HeV in bats of the genus Pteropus, but it was unclear whether this was due to infection with HeV or a related virus. We developed the hypothesis that HeV excretion from bats might be related to the birthing process and we targeted the reproductive tract for virus isolation. Three virus isolates were obtained from the uterine fluid and a pool of foetal lung and liver from one grey-headed flying-fox (Pteropus poliocephalus), and from the foetal lung of one black flying-fox (P. alecto). Antigenically, these isolates appeared to be closely related to HeV, returning positive results on immunofluorescent antibody staining and constant-serum varying-virus neutralization tests. Using an HeV-specific oligonucleotide primer pair, genomic sequences of the isolates were amplified. Sequencing of 200 nucleotides in the matrix gene identified that these three isolates were identical to HeV. Isolations were confirmed after RNA extracted from original material was positive for HeV RNA when screened on an HeV Taqman assay. The isolation of HeV from pteropid bats corroborates our earlier serological and epidemiological evidence that they are a natural reservoir host of the virus.
Introduction |
A new zoonotic disease affecting horses and humans
in Australia was
reported by Murray et al. (1995
). Two outbreaks of this virus zoonosis occurred within 1
month of each other, at locations 800 km apart (Brisbane, in south east
Queensland, and Mackay, in central Queensland) in 1994. A third event
involving a single fatal equine case occurred near Cairns in North
Queensland in January 1999. To date, two humans and 16 horses have died
from this disease (Rogers et al., 1996; Murray et al., 1995; Field et al., 2000). Typically, infected horses had clinical signs of an
acute respiratory disease. The first fatal human case also died of an
acute respiratory illness (Selvey et al., 1995). The second fatal human case suffered from relapsing
encephalitis (O'Sullivan et al., 1997). The causal agent is Hendra virus (HeV), formerly known
as equine morbillivirus, a previously undescribed virus belonging to the
family Paramyxoviridae (Wang et al., 1998).
Another member of the family Paramyxoviridae
emerged in late
1998. Nipah virus was responsible for an outbreak of severe encephalitis
in humans in Malaysia and Singapore and is attributed with the deaths of
more than 100 people (Lye et al., 1999). Measures to control the concurrent outbreak of
respiratory disease in pigs resulted in the culling of over one million
pigs (almost half the national pig herd) and had major domestic and
international trade repercussions (Aziz et al., 1999). Nipah virus, serological evidence of which
was also found in pteropid bats (Field et al., 1999), has been shown to have strong molecular and
sequence similarities with HeV, and the two viruses cross-neutralize (Rota
et al., 1999). Neither has yet been
assigned to a genus; more likely HeV and Nipah virus will form a new genus
within the family Paramyxoviridae.
Extensive serological sampling throughout eastern
Queensland revealed
that of over 5000 sera collected from 46 species, including 34 species of
wildlife, antibodies capable of neutralizing HeV have only been detected
in pteropid bats (Young et al., 1996). These fruit and nectar feeding bats, commonly known as
flying foxes, belong to the Order Chiroptera, Suborder
Megachiroptera, Family Pteropididae, genus Pteropus
(Hall, 1986). Over 25 % of sampled pteropid
bats, including representatives of all four main species in eastern
Australia (the grey-headed flying-fox, Pteropus poliocephalus; the
black flying-fox, Pteropus alecto; the little red flying-fox,
Pteropus scapulatus; and the spectacled flying-fox, Pteropus
conspicillatus), were identified as being seropositive. A serological
survey of 128 people with prolonged and close contact with pteropid bats
failed to detect any evidence of infection with HeV (Selvey et al.,
1996).
In this paper we describe the isolation of HeV from pteropid bats, corroborating our serological and epidemiological evidence that these animals are a natural reservoir host of this virus.
Methods |
Samples. Concurrent with serological
sampling of pteropid bat populations (Young et al., 1996) was the opportunistic sampling of recently
captured sick or injured wild pteropid bats in temporary captivity with
wildlife rescue groups. Samples of liver, kidney, lung, spleen and
reproductive organs were removed aseptically at necropsy of pteropid bats
which had recently died or been euthanized. Additionally, where possible,
immediately after euthanasia a blood sample was collected by cardiac
puncture for serology.
Virus isolation. Virus isolations were attempted in RK13 cells. This was found to be the most sensitive cell line to HeV in cell line sensitivity experiments conducted at the Queensland State Health Department (L. Hiley, unpublished results). Samples of liver, kidney, lung, spleen and reproductive organs were ground in a sterile medium consisting of PBS with 160 U/ml benzylpenicillin sodium, 1000 µ g/ml streptomycin and 4 µg/ml fungizone. The resulting homogenates were centrifuged at 3000 g for 10 min and the supernatant fluids were retained. Monolayers of RK13 cells grown in 24-well cell culture plates with 10 % Eagle’s minimum essential medium with Earle’s salts (10 % FCS, 2 µg/ml fungizone and 100 U/ml benzylpenicillin sodium) were inoculated with 0.2 ml of the supernatants of pteropid bat organ homogenates, absorbed for 60 min at 37 °C, washed with sterile medium and then maintained with 10 % Eagle’s minimum essential medium with Earle’s salts (2.5 % FCS, 2 µg/ml fungizone and 100 U/ml benzylpenicillin sodium). The 24-well plates were incubated at 37 °C with 5 % CO2. Cultures were observed for any signs of cytopathic effect (CPE). If no CPE was detected, inoculated cultures were harvested by a freeze–thaw at intervals of 7 days and passaged with attempts terminating after five serial passages.
To confirm virus isolation, a range of diagnostic tests were undertaken including the detection of HeV antigen by RT–PCR, immunofluorescent staining of infected cell sheets and a Taqman assay. Direct detection of antigen in tissues by immunostaining was not attempted. Contamination was avoided by observing accepted laboratory protocols and by returning to original tissue for confirmation of the presence of virus antigen.
RNA extraction. Upon detection of
CPE, and removal of some cell culture supernatant for further work
(described below), infected cells were lysed by freezing and thawing. RNA
was isolated by incubating 0.3 ml of this lysate with extraction buffer (4
M guanidine thiocyanate, 25 mM sodium citrate, 0.5 % sarcosyl, 0.1 %
2-mercaptoethanol, 2 M sodium acetate). Viral RNA was purified by
phenol–chloroform extraction and ethanol precipitation, and the RNA
was suspended in 18 µl of diethyl pyrocarbonate-treated water (Schorr
et al., 1994). For the purpose of the Taqman assay, RNA was
also extracted directly from tissue homogenate supernatant, as described
above.
RT–PCR. RT–PCR was
performed with the Expand RT preamplification system (Boehringer Mannheim)
according to the manufacturer’s recommendations. The oligonucleotide
primer pair were designed from the sequence published by Murray et
al. (1995) and amplified a 200 bp
region in the matrix (M) protein (forward primer 5´ GGC TAC AAC GAG
AAA TTT GTG 3´; reverse primer 5´ TTC TAG CAT TGT CCT TGG GAT
3´). For RT, 2.5 µl of template RNA was annealed with 10 pmol of
reverse-sense primer in a
total volume of 5 µl at 65 °C for 10 min. The mixture was then
chilled on ice. RT
reagents were added to give a total volume of 20.5 µl containing
synthesis buffer (250 mM Tris–HCl, 200 mM KCl, 2.5 mM
MgCl2, 2.5 % Tween 20, v/v, pH 8.3), 100 mM dithiothreitol, 1
mM (each) deoxynucleoside triphosphates (dATP, dCTP, dGTP, dTTP), 20 units
of RNase inhibitor and 50 units of Expand reverse transcriptase. This
mixture was incubated at 42 °C for 60 min and then chilled on ice.
The resulting cDNA was used as a template to amplify a 200 bp region of
the M gene of HeV. For PCR, a total volume of 50 µl was prepared,
containing synthesis buffer, 2.5 mM MgCl2, 10
pmol each primer, 2 µl of the reverse-transcribed product and 1.5
units Tth DNA
polymerase (Biotech International). Each sample was overlaid with mineral
oil. Amplification was performed in a DNA thermocycler (DNA Thermal Cycler
480, Perkin Elmer) as follows. Samples were initially heated at 94 °C
for 2 min; this was followed by 30 cycles consisting of denaturation (94
°C, 1 min), primer annealing (45 °C, 1 min) and extension (72
°C, 1 min). Samples were finally held at 15 °C. Size
fractionation of PCR product was performed by electrophoresis in 1.2 %
agarose gel prepared in 0.5x TBE buffer containing 0.5 µg/ml of
ethidium bromide at 80 V for 60 min, and was visualized on a UV
transilluminator.
Nucleotide sequencing of the target genomic region. RT–PCR product (40 µl) was separated from the primers and dNTPs by filtration through a QIAquick PCR purification column (Qiagen). All DNA sequencing was performed using the recovered DNA, the oligonucleotide primer pair used in the RT–PCR and dideoxy sequencing chemistry utilizing the ABI PRISM Dye Terminator Sequencing Ready Reaction kit, with AmpliTaq DNA polymerase FS according to the manufacturer’s instructions (PE Applied Biosystems). After recovery, sequencing products were resolved on an ABI automated A373 sequencer (PE Applied Biosystems) according to the manufacturer’s instructions.
Immunofluorescent antibody (IFA)
staining. When CPE was detected, cell culture supernatant
was used to inoculate 16-well glass chamber slides (Nunc) containing RK13
cells (2x104 cells per well). The chamber slides were incubated
at 37
°C in a humid atmosphere containing 5 % CO2. When CPE was
detected in these wells, supernatant was aspirated and the cell sheet was
fixed in cold 90 % acetone for 15 min, and then stained by an indirect IFA
technique. Antiserum to HeV collected from clinical horse and human cases
was used as the primary antibody. A fluorescein isothiocyanate-labelled
protein G conjugate (Zymed) was used to stain the preparations. Uninfected
cells, overlaid with horse serum positive for HeV antibodies and stained
with the conjugate as described above, were used as indicators of
specificity. Immunofluorescence was examined with a fluorescence
microscope (Olympus).
Constant-serum varying-virus neutralization
test. Supernatant collected from CPE-producing wells was
diluted from 10–1 to 10–6 in PBS. Horse
serum positive for HeV antibodies was held constant at 1/40. Horse serum
negative for HeV antibodies, also diluted at 1/40, was used as a negative
control. Serum and supernatant were incubated for 1 h at 37 °C and
subsequently inoculated onto RK13 cell monolayers in a 24-well plate (100
µl per well). Wells were monitored for signs of CPE. Taqman assay. This two-step assay was
developed at Queensland State Health Laboratories (I. Serafin, K. Halpin
& G. Smith, unpublished data). The first step involved conventional
cDNA production from RNA (extraction method described previously) using
random primers. The HeV Taqman primers and probes were designed using the
program Primer Express (PE Applied Biosystems) and reviewed based on the
design guidelines. The 69 bp target region was based on the sequence of
the HeV M gene. The PCR involved 50 cycles of denaturation at 95 °C
for 15 s followed by a combined annealing/extension step for 1 min at 60
°C (PE Applied Biosystems). Serology. Blood, collected by cardiac
puncture immediately after euthanasia, was allowed to clot at room
temperature and centrifuged at 3000 r.p.m. for 10 min. Serum was removed
from this preparation and stored at –20 °C until used. Serum was
initially screened by ELISA. All incubations took place at 37 °C. The
96-well microtitre plates (Nunc) were coated with inactivated HeV-infected
cell extract treated with 10 % SDS at a 1/400 dilution with PBS (50
µl per well) by incubation at 37 °C for 1 h. Following a brief
rinse
(3x200 µl) with PBS containing 0.1 % Tween 20 (PBST), the plates were
blocked for
1 h by the addition of 200 µl per well of 1 % skim milk powder (SMP)
in PBS. Plates were washed again,
as described above. Pteropid bat serum (50 µl), diluted 1/100 with
PBST containing 1 % SMP, was added and incubated
for 30 min, followed by another wash. The conjugate (ProteinG–HRP,
ICN Biomedicals) was diluted at 1/10000 and incubated for 30 min. After a
final washing of the plates, 50 µl of TMB peroxidase substrate
(K-blue, Neogen Corporation) was added to
each well and the plates were incubated for 10 min. At this point the
reaction was stopped by the addition of 50 µl of 1 M sulfuric acid
and the absorbance was read at 450 nm. Positive and
negative control horse sera were included on each plate. Additionally, aliquots of serum were sent to the
Australian Animal
Health Laboratories, Geelong, Victoria, Australia where serum
neutralization tests using HeV were carried out in a biosafety level 4
laboratory. Electron microscopy. Preparations of
supernatant collected from CPE-producing wells were mounted on grids
previously rendered hydrophilic by glow-discharge and were negatively
stained in 2 % phosphotungstic acid and viewed in a Philips CM 10
microscope operated at 60 kV, using a magnification of x25000 plus the x10
binocular. Between July and December 1996, 465 pteropid bats
were sampled,
providing 652 tissues for virus isolation. Most of these animals came from
south-east Queensland. Many pteropid bats were taken into captivity as a
result of physical injury, commonly electrocution on power transmission
cables, or entanglement in barbed-wire fences. No lesions, other than
those from trauma, were detected at necropsy of the pteropid
bats. Isolation of HeV Through the RK13 cell culture system, three virus
isolates were
obtained. The first two isolates were from a female grey-headed flying-fox
(Bat 1), which was euthanized due to extensive injuries. She had recently
aborted twin foetuses after becoming entangled on a wire fence, in
Brisbane. CPE was noted in the RK13 cells approximately 60 h after
inoculation with uterine fluid collected from this animal. Reisolation of
this first isolate was attempted and CPE was noted in the RK13 cells on
third passage. The second isolate was from a pooled sample of liver and
lung collected from the aborted foetuses of Bat 1. This material underwent
three serial passages before CPE was noted in the RK13 cells. The third
isolate was from the lung of a Pteropus alecto foetus, collected
from a female (Bat 2) which was euthanized following the diagnosis of
severe spinal injuries. Bat 2 was also found in Brisbane. This material
also underwent three serial passages before CPE was noted in the RK13
cells. All other tissues collected from these bats were negative for virus
isolation. Amplification of genomic region and
sequencing
The three isolates all produced a PCR product of the
expected size (200 bp), and when this region was sequenced, it was found
that the sequences were identical to that of HeV. IFA staining Infected cells showed very intense fluorescence
throughout the cell
sheet (Fig. 1). Fig. 1 also illustrates the typical
CPE
observed. The CPE was characterized by extensive fusion and formation of
syncytia, features which are commonly observed with other paramyxoviruses.
The IFA specificity was confirmed by the absence of fluorescent staining
in uninfected cells. Constant-serum varying-virus neutralization
test No cytopathic changes were observed at any dilution
where
horse serum positive for HeV antibodies was incubated with the virus
isolates. In the control, where horse serum negative for HeV antibodies
was incubated with the virus isolates, CPE was observed up to the
10–4 dilution. Taqman assay These isolates were positive when tested with the
Taqman assay. RNA
extracted from the uterine fluid of Bat 1, and RNA extracted from the
tissue homogenates of both foetuses (foetal lung and liver pool from Bat
1, and foetal lung from Bat 2) as well as supernatant from different
passages all gave positive results, confirming the original
isolation. Serology Bat 1 tested positive by HeV ELISA and by the HeV
serum
neutralization test (titre 1: 40). Serum collected from pooled organs of
the foetuses of Bat 1 was also ELISA-positive, but toxic on serum
neutralization test. Blood was not collected from Bat 2, or from the
foetus of Bat 2. Electron microscopy Preparations of supernatant collected from
CPE-producing wells,
examined by negative contrast electron microscopy, revealed virus
particles with typical paramyxovirus morphology. Fig.
2 shows
an example of the nucleocapsid filaments which were observed in the
infected supernatant. The ‘herring-bone’ morphology of the
nucleocapsid filaments produced by the close association of the N protein
with the genomic RNA, which is characteristic of the
Paramyxoviridae family, is clearly visible. Surface projections
were also visible on the virions. Following the first reported outbreak of HeV
infection of horses
(Murray et al., 1995 As previously described, our targeted serological
surveillance of
pteropid bats revealed antibodies capable of neutralizing HeV in all four
mainland pteropid species (Young et al., 1996 In targeting tissues for virus isolation, we
developed the hypothesis
that virus excretion might be associated with the birthing process. In
forming this hypothesis, we firstly observed that a serological study of
people with close and prolonged contact with pteropid bats had revealed no
evidence of infection (Selvey et al., 1996 Historically, bats, megachiropteran and
microchiropteran, have been
associated with many viruses both in terms of evidence of natural
infection and as possible reservoir hosts (Sulkin & Allen, 1974 The discovery of three new zoonotic viruses from
pteropid bats in
Australia is of wider interest. The distribution of the family
Pteropodidae worldwide encompasses south-east Asia, the Pacific
islands, India, Madagascar and much of Africa, with representatives of the
genus Pteropus found over the entire range except Africa
(Mickleburgh et al., 1992 The question of HeV transmission from pteropid bat
to susceptible
naïve host remains unanswered. The high seroprevalence rate in
pteropid bats indicates that infection is transmitted efficiently.
Transmission studies have proven that pteropid bats seroconvert and
develop subclinical disease when inoculated with HeV (Williamson et
al., 1998 We acknowledge Howard Prior, Queensland DPI
Animal Research
Institute, for assistance with electron microscopy, Paul Selleck and
colleagues at the Australian Animal Health Laboratories for performing the
Hendra virus serum neutralization tests, and Ina Serafin, Queensland State
Health Department, for assistance with the Taqman assay. Lester Hiley,
also from the Queensland State Health Department is to be thanked for
providing the results of his cell line sensitivity tests. We also thank
wildlife carers and groups, particularly ONARR, for their cooperation and
assistance in sampling pteropid bats. Aziz, J., Olson, J., Lee, O. B.,
Daniels, P., Adzhar,
A. B., Bunning, M., Field, H., Johara, M. Y. & Hooper, P. (1999).
Nipah virus infection of animals in Malaysia. In Abstracts of the
XIth International Congress of Virology, Sydney, Australia
(9th–13th August, 1999). Abstract VW31B.02, p. 38. Flannery, T. F. (1995).
Mammals of the
South-West Pacific and Moluccan Islands. Sydney: Reed
Books. © 2000
SGM This article is now available in the August 2000
print issue of JGV (vol. 81, 1927–1932). The complete issue of the
journal may be seen in electronic form on JGV
Online.
Results
Fig. 1. Positive
immunofluorescent staining of RK13 cells infected with HeV isolated from
the uterine fluid of Bat 1.
Fig. 2. Electron micrograph of virus isolated
from
the uterine fluid of Bat 1. Magnification x200000. (a)
Nucleocapsid; (b) surface
projections.
Discussion
), the Queensland Department of Primary industries (DPI)
Animal and Plant Health Service undertook a search for the natural
reservoir host of the new virus. A wide range of domesticated and native
animals were surveyed for serological evidence of infection, but none was
found. Subsequent to the second reported outbreak (Rogers et al.,
1996), a multidisciplinary group
including representatives from the Queensland DPI, the University of
Queensland, and the Australian Animal Quarantine Service reviewed the
available laboratory and epidemiological data. As a result we postulated
that pteropid bats were a plausible natural host as they best fulfilled
the proposed characteristics of the most likely reservoir host of HeV
(Young et al., 1996).). However, the possibility existed that these animals were
infected with a related paramyxovirus and it was therefore important to
isolate virus from pteropid bats and compare it to HeV.). We interpreted these findings as possibly indicating
that the route of excretion of the virus was novel. Secondly, the
outbreaks of HeV infection in horses occurred during the birthing season
of three of the four seropositive pteropid species. We therefore included
uterine fluids, foetal tissues and placenta in virus isolation
attempts.). However, there have only been two previous
reports detailing the isolation of paramyxoviruses from megachiropteran
bats. Pavri et al. (1971) reported the isolation of a virus from a Leschenaultās
rousette fruit bat (Rousettus leschenaulti) in India; a virus which
was later identified as a new animal subtype of parainfluenza virus type
2. Mapuera virus, belonging to the genus Rubulavirus, was isolated
from the salivary glands of a yellow-shouldered bat (Sturnira
lilium) captured in the tropical rainforest of Brazil in 1979
(Henderson et al., 1995). Neither of these paramyxoviruses has been associated
with disease and/or infection in man, making our finding the first
reported isolation of a zoonotic paramyxovirus from bats. Additionally,
our earlier serological evidence of neutralizing antibodies to HeV in
pteropid bats (Young et al., 1996) led to an increased disease surveillance interest in
bats. Findings of a rabies-like virus, Australian bat lyssavirus, in
Australian pteropid bats (Fraser et al., 1996) and of infection with another previously undescribed
paramyxovirus, Menangle virus (Philbey et al., 1998), followed. Most recently, surveillance of
wildlife for the source of the recently emerged and zoonotic Nipah virus
in Malaysia revealed serological evidence of infection in two species of
Malaysian pteropid bats (Field et al., 1999), further emphasizing the close association between these
two new members of the family Paramyxoviridae.). Notwithstanding the generally restricted range of
individual species, the overlapping ranges of neighbouring species (Corbet
& Hill, 1992; Flannery, 1995) make feasible the inter-species transmission
of infectious agents, or the possibility of related viruses in other
pteropid species across the entire range. The first of these may be the
recent discovery of Nipah virus in Malaysia.). The isolation of virus
from foetal tissues and fluids suggests one possible mechanism for
transmission of infection to other animals, by way of direct contact with
aborted foetuses and other products expelled during the birthing process.
That this mode of transmission would be purely a chance event supports its
consideration as a possible mechanism of transmission. If the route of
excretion made the virus more readily accessible (for example, in urine or
faeces) one would expect an increased chance of exposure by susceptible
hosts and hence an increase in the number of spillover infections and
cases. However, much remains to be learned about the nature of the
infection in pteropid bats, including the sites of virus replication,
persistence and the behaviour of the virus in other genera and
species.
References
Supplementary material
An adult
grey-headed flying fox (Pteropus poliocephalus). Brisbane,
Australia.
A colony of little red
flying foxes (Pteropus scapulatus) flying out of their camp at dusk.
Indooroopilly Island in Brisbane, Australia.
Grey-headed flying foxes
(Pteropus poliocephalus) in their camp. Brisbane, Australia.
JGV
Direct table of contents