Download West Nile Virus Infection in Birds and Mammals

West Nile Virus Infection in Birds
and Mammals
LAURA D. KRAMER AND KRISTEN A. BERNARD
Arbovirus Laboratories, Wadsworth Center,
New York State Department of Health, Slingerlands, New York 12159, USA
ABSTRACT: West Nile virus (WNV) was found throughout New York State in
year 2000. The epicenter was located in New York City with a high level of activity in the immediately surrounding counties, including Rockland, Westchester, Nassau, and Suffolk. During 2000, WNV testing was performed by the
Wadsworth Center on 3,687 dead birds, representing 153 species, 46 families,
and 18 orders. There were 1,203 WNV-positive birds, representing 63 species,
30 families and 14 orders. The percentage of WNV-positive birds was 33% for
all birds tested throughout the state, with no significant difference in infection
rates in migratory versus resident birds, although significantly more resident
birds were submitted for testing. The highest apparent mortality for the entire
season was observed in American crows in Staten Island, a location that also
showed the highest minimal infection rate in Culex pipiens complex mosquitoes.
Studies examining tissue tropism of WNV in corvids and noncorvids from the
epicenter and from remote locations indicated that the kidney was the most
consistently infected tissue in birds, regardless of level of infection. The brain
was the next most consistently positive tissue. The differences in infection
among the tissues were most apparent when low levels of virus were present.
Experimental mouse inoculation demonstrated a classical flavivirus infection
pattern.
KEYWORDS: West Nile virus; tissue tropism; birds; mammals
INTRODUCTION
West Nile virus (WNV) is a newly emerging mosquito-borne virus now found on
four continents, Africa, Asia, Europe, and North America. The infection in humans
has been characterized by a broad range of clinical symptoms, from asymptomatic
infection to mild nonspecific symptoms, including fever and headache, to more serious cases with myocarditis and encephalitis.1 In 1999, 62 cases of WNV were confirmed in New York City and in 2000, 14 cases. Serosurveys conducted in 1999
demonstrated that 2.6% of the population in an area of high viral activity had antibody to WNV,2 and in 2000, approximately 1% of the population demonstrated
seroconversion.2
The introduction of WNV to the northeastern United States in 1999 was accompanied by unprecedented mortality in indigenous and captive birds. This was unexAddress for correspondence: Laura D. Kramer, Ph.D., Director, Arbovirus Laboratories, Wadsworth Center, New York State Department of Health, 5668 State Farm Road, Slingerlands, NY
12159. Voice: 518-869-4524; fax: 518-869-4530.
[email protected]
84
KRAMER & BERNARD: WEST NILE VIRUS IN BIRDS AND MAMMALS
85
pected since, historically, in countries where WNV is enzootic, the virus has led to
minimal disease in avian species. WNV has been demonstrated to infect a wide variety of vertebrate species besides man and birds, as indicated by evidence of antibody to the virus and virus isolation.3 Studies during the 1950s in Egypt indicated
that the hooded crow and the house sparrow had antibody rates of 65 and 42%, respectively, and virus was isolated from the blood of an adult crow.4
The work presented here is aimed at elucidating WNV infection in vertebrates
other than man. Data from 2000 in New York State were analyzed to determine
which bird species were found infected most frequently with WNV in 2000, and
which avian tissues were most often infected. Five out of 149 wild mammals were
found to be infected with WNV in year 2000 in New York State. These included one
little brown bat (Myotis lucifugus), one big brown bat (Eptesicus fuscus), one squirrel (Sciurus carolinensis), one domestic rabbit (Oryctolagus cuniculus), and one
eastern chipmunk (Tamias striatus). A preliminary study delineating infection in
white mice as a model for infection in mammals was undertaken to begin to define
the pathogenesis of WNV.
MATERIAL AND METHODS
Tissue Submission and Processing
Dead birds and mammals were submitted for necropsy to the Department of Environmental Conservation, Wildlife Pathology Unit, New York State. Multiple tissues
from suspect animals were forwarded to the Arbovirus Laboratories of the Wadsworth
Center, New York State Department of Health, for viral assay. Protocols for handling
tissues and for viral assays have been described previously.5 Briefly, 50 mm3 sections
of tissue were excised and triturated directly in lysis buffer containing guanidium
isothiocyanate (Qiagen, CA). RNA was extracted as per the manufacturer’s instructions, and reverse transcription–polymerase chain reaction (RT-PCR) was conducted
using an ABI Prism 7700 Sequence Detector using TaqMan One-Step RT-PCR master
mix (Applied Biosystems, Foster City, CA). Two sets of primers/probes (1160; 3111)
were used as described.5,6 Ct values—the PCR cycle at which an increase in the fluorescence rises above the threshold value determined by the standard curve—were recorded; from this the TaqMan score was calculated, that is, + to ++++.
Determination of Minimum Infection Rate (MIR) in Mosquitoes
MIRs were calculated assuming one infected mosquito per pool per 1000 mosquitoes tested. Mosquito pools were processed as described.5 Briefly, pools were
triturated using SpexCertiPrep 8000-D mixer mill (Metuchen, NJ). A 350 µL aliquot
was removed immediately for RNA extraction, which was conducted as described
above for vertebrate tissue, as was RT-PCR. The remainder of the triturated pool was
held for inoculation of cell culture to detect live infectious virus.
Cell Culture Assay
African green monkey kidney (Vero) cell cultures in 25-mm flasks were inoculated with 0.1 mL homogenized and clarified mosquito suspension. Monolayers were
86
ANNALS NEW YORK ACADEMY OF SCIENCES
checked daily for seven days for evidence of cytopathology (cpe). After seven days
or when cpe was observed, cells were spotted on slides, fixed and stained with primary antibody, H5.46 WNV monoclonal antibody, and secondarily with goat antimouse IgG fluorescein-conjugated antibody (Kirkegaard and Perry Laboratories,
Gaithersburg, MD). Fluorescence was evaluated using an Olympus BH-2 microscope equipped with FITC filter set.
Plaque Assay
Cell culture plaque assays to quantitate infectious virus were conducted as previously described.7 Briefly, six-well Costar plates with Vero cell monolayers were inoculated with homogenized and clarified tissue suspensions, 0.1 mL per well. A
single overlay of 2.0% oxoid agar with 0.8 mL of 1.0% neutral red was added;
plaques were read from day 3 to day 7.
Pathogenesis of West Nile Virus in Adult Mice
Female Balb/C mice, five weeks old, were inoculated intraperitoneally with
103 PFU WNV isolated from a pool of 50 Cx. pipiens mosquitoes from New York in
2000. Virus was diluted in low endotoxin phosphate-buffered saline containing 1%
fetal bovine serum (diluent). Three control mice were inoculated with diluent alone,
and three additional mice were not inoculated, but were housed with inoculated mice
as contact controls. The mice were weighed and observed for clinical signs daily.
Three mice each were sacrificed at 19 and 28 hours, days 2, 3, 4, 5, 8, and 9. Control
mice were held for 24 days. Heart blood was collected, and the brain, liver, spleen,
kidney, and heart were harvested for assay of infectious virus. All tissues were
weighed, and diluent was added to make a 20% homogenate for brain or 10% homogenate for all other organs. The limit of detection was 250 PFU/g for brain and 500
PFU/g or mL for other organs and serum.
RESULTS
Avian Infection
A comparison of infection rates in year-round residents, migrants, mixed populations (in which only a portion of the population migrates), and captive birds indicated that the predominant number of birds that died and were tested for WNV were
permanent residents (TABLE 1). However, there was no difference in the proportion
of birds from all categories that were infected, 30 to 43%.
The proportion of American crows found infected with WNV was compared with
the number of human8 and equine9 cases in the five boroughs of NYC, Long Island
(Nassau and Suffolk counties), and two other locations within the epicenter,
Westchester and Rockland counties (FIG . 1). Ten of 14 cases in humans in New York
State were detected on Staten Island (SI), where more than 90% of the crows tested
were infected, and the minimal infection rate in Cx. pipiens complex mosquitoes
(Cx. pipiens and Cx. pipiens-restuans combined) was 10.9. Three other boroughs in
NYC—Brooklyn, Manhattan, and Queens—had one to two human cases, and all
other locations had none. The greatest numbers of dead crows were submitted by
KRAMER & BERNARD: WEST NILE VIRUS IN BIRDS AND MAMMALS
87
FIGURE 1. Relative infection rates of crows, humans, and equines by location in epicenter of WNV outbreak, 2000. MIR, minimal infection rate in Culex pipiens and Culex pipiens-restuans complex mosquitoes combined. Number of human and equine cases and
percentage positive crows are reported. Numbers above bars represent number of crows tested.
TABLE 1. Infection in birds with different migration patterns
Behavior pattern
Number
tested
Species
Positive
Individuals
Percent
Year-round residents
2924
28
1116
38
True migrants
51
14
22
43
Mixed population
175
15
53
30
Captive birds
25
6
10
40
Rockland county, of which 75% were infected with WNV; the MIR in Cx. pipiens
complex mosquitoes in Rockland was 2.0, but no human or horse cases were observed. Suffolk had the second highest number of dead crows submitted for testing
and had the greatest number of equine cases in NYS (8), but infected horses also
were detected in Staten Island, the Bronx, and Nassau. More detailed information on
WNV infection in birds and mosquitoes can be found in Bernard et al.10
The MIRs of Cx. pipiens and Cx. pipiens-restuans complex mosquitoes were
combined. Other Culex species were not included in this summary because of problems encountered with identification of other species of Culex mosquitoes (unpublished data, Kramer). Vero cell cultures inoculated with positive mosquito pools of
all species indicated that infectious virus could be detected in 111/266 (41.7%)
88
ANNALS NEW YORK ACADEMY OF SCIENCES
TABLE 2. WNV cell culture results for mosquitoes collected in New York State
during 2000
Mosquito species
Ochlerotatus cantator
Oc. japonicus
Oc. triseriatus
Ae. vexans
Ae. species
An. punctipennis
Culex pipiens
Cx. pipiens-restuans
Cx. salinarius
Cx. species
Psorophora ferox
No. positive pools by No. positive pools by cell culture
RT-PCR
(n tested)
1
5
3
10
1
1
79
212
31
19
1
0 (1)
0 (5)
1 (3)
1 (10)
0 (0)
0 (1)
25 (60)
61 (146)
13 (26)
9 (13)
1 (1)
FIGURE 2. Relative TaqMan score of brain, heart, and liver compared to kidney.
The numbers above the bars indicate the number of tissues compared in parallel with the
kidney. 1+ represents Ct value 32–36; 2+ to 3+, 24–32; 4+, 14–25.
(TABLE 2). Passage through cell culture did not allow for amplification of initially
undetectable levels of infectious virus.
Tissue Tropism in Birds
TaqMan RT-PCR results from kidney, heart and brain tissues from 49 crows were
compared in order to determine tissue tropisms of WNV (FIG . 2). Birds found infected with low, medium, and high concentrations of viral WNV RNA during 2000 were
selected. Multiple tissues from the crows were tested in parallel, but not all tissues
were available from all birds. Each tissue was scored as its TaqMan score relative to
the TaqMan score of the kidney of the same bird, and the mean was taken. Kidney
was the most consistently positive tissue, regardless of the level of infection. The
heart, brain and liver of birds with TaqMan scores of 3–4+ in the kidney had levels
KRAMER & BERNARD: WEST NILE VIRUS IN BIRDS AND MAMMALS
89
of viral RNA equal to that of the kidney, and these tissues therefore were equally
good indicators of infection. Similarly, heart and brain of birds with TaqMan scores
of 2+ in the kidneys were infected to an equal extent; liver, however, had 0.7 (70%)
of the level of viral RNA. The heart, brain, and liver of birds with kidneys that were
low positives (1+) had approximately 0.2 (20%) of the kidney’s level of viral RNA.
Focal Infection of Tissues
A study was conducted to determine the uniformity of viral infection in various
organs of crows. The level of viral RNA was determined in 8 to 15 separate sections
excised from each of 12 individual paired crow brains and kidneys (TABLE 3). There
was insufficient tissue available to excise 15 sections in every case. All but four positive tissues demonstrated variable levels of viral RNA (i.e., from Ct value 36 to 14,
TaqMan score 1+ to 4+) in excised tissue sections. Four tissues demonstrated Ct values < 25 (i.e., 4+ levels of viral RNA in all sections: 3 brains, crows 10, 11, 12 and
one kidney, crow 12). In two of the 12 crows (crows 1 and 3), brains were negative
when the corresponding kidneys had several sections that were low positives (Ct 36
to 32); in one additional crow (crow 2), the brain was negative (− or ±) when the kidney was highly positive (Ct 29 to14). In one crow (crow 4), 13 of 15 brain sections
were highly positive with Ct values < 29, but only one of 11 kidney sections was a
low positive with a Ct value of 34. The variability observed among sections of both
brain and kidney was more than 10-fold in some instances.
Tissue Tropism in Mammals
The WNV-inoculated mice appeared healthy until day 6 when mild clinical signs
were evident in one mouse. Neurologic signs, ranging from ataxia and weakness to
bilateral hindleg paralysis, were observed on days 7 through 9 in two of the six remaining mice. The maximum weight loss was 9%. No clinical signs were observed
in the mock-inoculated or contact control mice. The following gross pathologic
changes were observed in some, but not all mice: myocarditis (days 4, 5, 8, and 9);
splenomegaly and lymphadenopathy (day 5); hemorrhagic meninges, hepatic lipidosis, lymphonecrosis of spleen and lymph nodes, reduced thymic size, and intestinal
distention (days 8 and 9). Virus in the serum reached a peak of 105.4 PFU/mL after
three days and was cleared rapidly with only one mouse having a detectable viremia
day 4 (FIG . 3). Infectious virus was also present early in the spleen of two mice at 19
h; one spleen contained 103 PFU/mL in the absence of detectable viremia. The titer
in the spleen then increased to a peak at day 3 to 4. The heart was first positive at 28
h, and then increased in titer until day 4. On each of days 5, 8, 9, one heart was positive with low levels of virus, 103.2–103.5 PFU/g. Virus was first detected in the kidney after 2 days. The level of infectious virus in this tissue increased slowly, reaching
a peak on day 5, with 105.2 PFU/g virus. The brain of one mouse was infected at 28
hours. The concentration of virus in the brain remained low until day 5, when one
mouse had a significantly increased titer, 104.8 PFU/g. By day 8, one mouse was very
sick, with 106.6 PFU/g virus in its brain, and 103.2 PFU/g virus in its heart. Another
mouse, which was mildly ill on day 8 with ruffled fur but no neurologic signs, was
found dead on day 9. The kidney and brain of this mouse had detectable virus, but
with titers below the peak, 102.7 and 104.9 PFU/g, respectively.
TABLE 3. Focus of West Nile virus in the brain and kidney of infected American crows
90
ANNALS NEW YORK ACADEMY OF SCIENCES
KRAMER & BERNARD: WEST NILE VIRUS IN BIRDS AND MAMMALS
91
FIGURE 3. Experimental infection of Balb/C mice with WNV. Mice were inoculated
intraperitoneally with 103 PFU WNV isolated from a pool of Culex pipiens in NY, 2000. Each
bar represents results for a tissue from an individual mouse. White bar = serum; diagonally
hatched ba r= spleen; black ba r = brain; stippled ba r= heart; horizontally hatched ba r= kidney.
No bar indicates that the sample was below the limit of detection (250 to 500 PFU/g or mL).
DISCUSSION
WNV has a very broad host range. In 2000 in the northeastern US, more than 60
species of birds became infected and died from this virus. This is in sharp contrast
to the biology of WNV in Europe, Asia and Africa, where there are few reports of
bird mortality. Laboratory experiments with Hooded Crows demonstrated a high
mortality in young birds and development of immunity in adult birds.11 Fourteen orders of birds were found infected with WNV in New York State in 2000.10 The predominant number of dead birds tested were members of the Order Passeriformes,
followed by Galliformes, Columbiformes, and Charadriiformes. When species that
had a minimum of 10 individuals tested were compared, the percent of positive birds
ranged from 15% (Ciconiformes) to 44% (Psittaciformes).10
In New York State, within the epicenter of viral activity—the five boroughs of
NYC, Long Island (Nassau and Suffolk counties), Rockland and Westchester counties—American crows demonstrated significantly greater mortality from WNV than
did other species of birds.10 In other locations in the state, no difference in infectivity
was observed among bird species found dead. A sharp increase in sightings of dead
crows preceded the first human case detected in NYC,12 possibly providing an early
92
ANNALS NEW YORK ACADEMY OF SCIENCES
warning of increased risk to man. The highest MIR in mosquitoes on Staten Island
correlated with the greatest proportion of infected crows and the greatest number of
human cases. Other locations had lower MIRs, but five of eight had MIRs greater
than 1. An MIR of 1 with St. Louis encephalitis virus, a related flavivirus, in Florida
has been understood to signify increased risk to humans.13 It is important to recognize, however, that the WNV MIRs reported here are a measure of the number of
mosquitoes with detectable viral RNA and not in many cases, infectious virus. This
is not surprising since the limit of detection of the TaqMan RT-PCR assay was determined to be 0.08 PFU or 10−1.1 log10 PFU.5 Thus the significance of the MIRs
must be interpreted with caution. Mosquitoes that have evidence of viral RNA only
and no infectious virus represent incompetent vectors, but theoretically they might
have become competent with an increased extrinsic incubation period had they lived
out their lifespan. Calculation of MIRs for each species also is dependent on correct
identification of mosquitoes, which in some cases was problematic in 2000 (unpublished data, Kramer).
Although the greatest number of birds submitted for testing were year-round residents of New York State, there was no difference in infection rates among birds with
different breeding behavior. Viremic migratory birds are suspected to contribute to
the movement of WNV in southern Europe.14 Similarly, migrating species on the
East Coast of the US may provide a mechanism of viral transport over distance and
thereby a means of viral spread to new areas throughout both temperate and tropical
regions of the Western Hemisphere. However, the true likelihood that this would
happen is unclear because the length of time that birds have viremic titers sufficient
to infect vector mosquitoes may be short in duration. Birds that will be most important to the transmission cycle probably gather in large groups in locations with high
numbers of ornithophilic mosquitoes.
The analysis of RNA levels in brain, heart, kidney and liver indicated that the kidney, and secondarily the brain, was the avian tissue most consistently infected with
WNV. These results confirmed those of Steele et al.15 They examined birds infected
with WNV in 1999, following introduction of the virus into New York, and found
that the kidney was the tissue most frequently infected, but that the brains of most
birds were also infected with WNV. Furthermore, the examination of the focal infection within the kidney and brain, as determined by TaqMan assay of multiple sections, suggests that the kidney displays consistently higher levels of viral antigen
than does the brain. This observation is further supported by immunofluorescent
staining of frozen tissue sections of the brain and kidney (unpublished data). WNV
antigen appears much more focal in the brain than in the kidney, which showed evidence of staining throughout.
The experimental infection in mice demonstrated a classical flavivirus pattern of
infection. After peripheral inoculation, virus was found in lymphoid tissues, followed by the heart and kidney. Peak viremia occurred on day 2 to 3. Virus was found
in the brain as early as 28 h and consistently on days 2 and 3. It is important to note
that the viremia was high for all three of these time points; therefore, the virus measured in the brain may represent virus in the blood. By day 4, virus unambiguously
had invaded the brain. Thus, the NY strain of WNV was neuroinvasive in adult mice.
Using three-week-old outbred mice, Halevy et al. found similar results after peripheral inoculation using an Israeli strain of WNV.16
KRAMER & BERNARD: WEST NILE VIRUS IN BIRDS AND MAMMALS
93
ACKNOWLEDGMENTS
We would like to acknowledge the considerable contributions of the Arbovirus
Laboratories’ staff, including Elizabeth B. Kauffman, Susan A. Jones, Alan P. Dupuis II, Kiet A. Ngo, Greg D. Ebel, Donna M. Young, and David Nicholas. We thank
Ward Stone and the Wildlife Pathology Unit of the Department of Environmental
Conservation, the Division of Epidemiology of the New York State Department of
Health, the New York City Department of Health, and the Centers for Disease Control and Prevention.
REFERENCES
1. SAMPSON, B.A., C. AMBROSI, A. CHARLOT, et al. 2000. The pathology of human West
Nile Virus infection. Hum. Pathol. 31: 527–531.
2. CDC. 2001. Serosurveys for West Nile virus infection—New York and Connecticut
counties, 2000. Morb. Mortal. Wkly. Rep. 50: 37-39.
3. HAYES, C.G. 2000. West Nile Fever. In The Arboviruses: Epidemiology and Ecology,
Vol. V. T.P. Monath, Ed.: 59–88. CRC Press, Inc. Boca Raton, FL.
4. TAYLOR, R.M., T.H. WORK, H.S. HURLBUT & F. RIZK. 1956. A study of the ecology of
West Nile virus in Egypt. Am. J. Trop. Med. Hyg. 5: 579.
5. SHI, P-Y., E.B. KAUFFMAN, P. REN, et al. 2001. High throughput detection of West Nile
virus RNA. J. Clin. Microbiol. 39: 1264–1271.
6. LANCIOTTI, R.S., A.J. KERST, R.S. NASCI, et al. 2000. Rapid detection of west nile virus
from human clinical specimens, field-collected mosquitoes, and avian samples by a
TaqMan reverse transcriptase-PCR assay. J. Clin. Microbiol. 38: 4066–4071.
7. REISEN, W.K., R.P. MEYER, S.B. PRESSER & J.L. HARDY. 1993. Effect of temperature on
the transmission of western equine encephalomyelitis and St. Louis encephalitis
viruses by Culex tarsalis (Diptera:Culicadae). J. Med. Entomol. 30: 151–160.
8. CDC. 2000. Update: West Nile Virus activity—Eastern United States, 2000. Morb.
Mortal. Wkly. Rep. 49: 1044–1047.
9. TROCK, S.C., B. MEADE, A.L. GLASER, et al. 2001. West Nile outbreak among horses in
New York State, 1999 and 2000. Emerg. Infect. Dis. In press.
10. BERNARD, K.A., J.G. MAFFEI, S.A. JONES, et al. 2001. Comparison of West Nile virus infection in birds and mosquitoes in New York State in 2000. Emerg. Infect. Dis. 7: 679–685.
11. WORK, T.H., H.S. HURLBUT & R.M. TAYLOR. 1953. Isolation of West Nile virus from
hooded crow and rock pigeon in the Nile Delta. Proc. Soc. Exp. Biol. Med. 84: 719–722.
12. EIDSON, M., L.D. KRAMER, W. STONE & I. GOTHAM. 2001. Dead bird surveillance as an
early warning system for West Nile virus. Emerg. Infect. Dis. 7: 631–635.
13. DAY, J.F. & L.M. STARK. 1996. Transmission patterns of St. Louis encephalitis and
eastern equine encephalitis viruses in Florida: 1978–1993. J. Med. Entomol. 33:
132–139.
14. RAPPOLE, J.H., S.R. DERRICKSON & Z. HUBALEK. 2000. Migratory birds and spread of
West Nile virus in the Western Hemisphere. Emerg. Infect. Dis. 6: 319–328.
15. STEELE, K.E., M.J. LINN, R.J. SCHOEPP, et al. 2000. Pathology of fatal West Nile virus
infections in native and exotic birds during the 1999 outbreak in New York City, New
York. Vet. Pathol. 37: 208–224.
16. HALEVY, M., Y. AKOV, D. BEN-NATHAN, et al. 1994. Loss of active neuroinvasion of
West Nile virus: pathogenicity in immunocompetent and SCID mice. Arch. Virol.
137: 355–370.